Perspectives on the Performance of the Continental Economies

Perspectives on the Performance of the Continental Economies

CESifo Seminar Series edited by Hans-Werner Sinn Michael M. Hutchison and Frank Westermann, editors, Japan’s Great Stagnation: Financial and Monetary Policy Lessons for Advanced Economies Jonas Agell and Peter Birch Sørensen, editors, Tax Policy and Labor Market Performance Marko Köthenbürger, Hans-Werner Sinn, and John Whalley, editors, Privatization Experiences in the European Union Jay Pil Choi, editor, Recent Developments in Antitrust: Theory and Evidence Ludger Woessmann and Paul E. Peterson, editors, Schools and the Equal Opportunity Problem Bruno S. Frey and Alois Stutzer, editors, Economics and Psychology: A Promising New Field Mark Gradstein and Kai A. Konrad, editors, Institutions and Norms in Economic Development Robert Fenge, Georges de Ménil, and Pierre Pestieau, editors, Pension Strategies in Europe and the United States Steven Brakman and Harry Garretsen, editors, Foreign Direct Investment and the Multinational Enterprise Reinhard Neck and Jan-Egbert Sturm, editors, Sustainability of Public Debt Roger Guesnerie and Henry Tulkens, editors, The Design of Climate Policy Stephan Klasen and Felicitas Nowak-Lehmann, editors, Poverty, Inequality, and Policy in Latin America Gregory D. Hess, editor, Guns and Butter: The Economic Laws and Consequences of Conflict Timothy Besley and Rajshri Jayaraman, editors, Institutional Microeconomics of Development Paul DeGrauwe, editor, Dimensions of Competitiveness Vivek Ghosal, editor, Reforming Rules and Regulations Edmund S. Phelps and Hans-Werner Sinn, editors, Perspectives on the Performance of the Continental Economies See http://mitpress.mit.edu for a complete list of titles in this series.

Perspectives on the Performance of the Continental Economies

edited by Edmund S. Phelps and Hans-Werner Sinn

The MIT Press Cambridge, Massachusetts London, England

© 2011 Massachusetts Institute of Technology All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher. For information about special quantity discounts, please email special_sales@mitpress. mit.edu This book was set in 10/13 pt Palatino by Toppan Best-set Premedia Limited. Printed and bound in the United States of America. Library of Congress Cataloging-in-Publication Data Perspectives on the performance of the continental economies / edited by Edmund Phelps and Hans-Werner Sinn. p. cm. – (CESifo seminar series) Includes bibliographical references and index. ISBN 978-0-262-01531-8 (hardcover : alk. paper) 1. Europe, Western–Economic conditions–1945- 2. Economic indicators–Europe, Western. I. Phelps, Edmund S. II. Sinn, Hans-Werner. HC240.P383 2011 330.94–dc22 2010035922 10

9

8

7

6

5

4

3

2

1

Contents

Contributors vii Series Foreword ix

1

Introduction: Gauging and Explaining Economic Performance in Continental Europe 1 Edmund S. Phelps and Hans-Werner Sinn

2

Entrepreneurship in Europe and the United States: Security, Finance, and Accountability 27 Roman Frydman, Omar Khan, and Andrzej Rapaczynski

3

Europe’s Venture Capital Institutions Are Good Enough Richard Robb

4

Promoting Entrepreneurship: What Are the Real Policy Challenges for the European Union? 91 Anders N. Hoffmann

5

Innovations to Foster Risk-Taking and Entrepreneurship Robert J. Shiller

135

6

Europe: Cultural Adjustment to a New Kind of Capitalism? Harold James

151

7

Venturesome Consumption, Innovation, and Globalization 169 Amar Bhidé

8

Cyclical Budgetary Policy and Economic Growth: What Do We Learn from OECD Panel Data? 223 Philippe Aghion and Ioana Marinescu

65

vi

Contents

9

Policies to Create and Destroy Human Capital in Europe James J. Heckman and Bas Jacobs

10

Market Forces and the Continent’s Growth Problem Gylfi Zoega

11

Controversies about Work, Leisure, and Welfare in Europe and the United States 343 Robert J. Gordon

12

Revisiting the Nordic Model: Evidence on Recent Macroeconomic Performance 387 Jeffrey D. Sachs

13

The Welfare State and the Forces of Globalization Hans-Werner Sinn

14

Payroll Taxes, Wealth, and Employment in Neoclassical Theory: Neutrality or Nonneutrality? 429 Hian Teck Hoon

15

Economic Culture and Economic Performance: What Light Is Shed on the Continent’s Problem? 447 Edmund S. Phelps Index

483

253

323

413

Contributors

Harvard University

Philippe Aghion

Harvard University

Amar Bhidé

New York University

Roman Frydman Robert Gordon

Northwestern University

James Heckman

University of Chicago

Anders Hoffmann

Singapore Management University

Hian Teck Hoon Bas Jacobs

FORA

Erasmus University Rotterdam

Harold James Omar Khan

Princeton University Columbia University University of Chicago

Ioana Marinescu Edmund Phelps

Columbia University

Andrzej Rapaczynski Richard Robb Jeffrey Sachs

Columbia University

Columbia University Columbia University

Robert Shiller Yale University Hans-Werner Sinn Gylfi Zoega

CESifo

University of London

Series Foreword

This book is part of the CESifo Seminar Series. The series aims to cover topical policy issues in economics from a largely European perspective. The books in this series are the products of the papers and intensive debates that took place during the seminars hosted by CESifo, an international research network of renowned economists organized jointly by the Center for Economic Studies at Ludwig-Maximilians-Universität, Munich, and the Ifo Institute for Economic Research. All publications in this series have been carefully selected and refereed by members of the CESifo research network.

1

Introduction: Gauging and Explaining Economic Performance in Continental Europe Edmund S. Phelps and Hans-Werner Sinn

Felix qui potuit rerum cognoscere causas. (Happy is he who can know the causes of things.) Virgil, Georgics1

At the Lisbon Summit of March 2000, the European heads of state proclaimed their goal of making Europe “the most competitive and dynamic knowledge-based economy in the world” by 2010. Now the decade has come to an end and the transformation of the European economy is not in sight. It is widely perceived by now that continental western Europe has gone from catching up to falling back, its torpor exceeded only by Japan. This was confirmed in the data of the past dozen years when, in continental western Europe, GDP growth was slower than that in any other region of the world, except for Japan (figure 1.1).2 The underperformance in regard to productivity and the performance in some other dimensions are the subject of this book. What ails the economies of continental western Europe? Is it mainly some deficiency in their economic system? Is it mainly some effect of their welfare systems of social insurance and assistance: either the taxes to pay for them or the entitlements themselves? Or are all these systems healthy, even by advanced standards, yet weighed down by adverse market prospects in the continental economies—notably, the Continent’s grim demographic future? The state of Europe’s economies raises issues on which economists have been divided for some time. Some adherents of neoclassical economic theory see no causal role at all for a country’s “economic system.” They recognize a need for such legal institutions as rights to hold wealth, from land to government bonds, and rights to contract one’s labor for goods. They also recognize the convenience of facilities that

2

Edmund S. Phelps and Hans-Werner Sinn

180 170

Real GDP Index, 1995=100

Developing Asia 136.1%

160

Middle East 74.1% Eastern Europe 66.9% World 57.7% Latin America 48.5%

150 140

Africa 78.3%

Hong Kong, Singapore, South Korea, Taiwan 72.5%

United States 44.3% EU27 35.4%

130 120 Japan 17.2%

110 100 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Figure 1.1 Economic growth in selected regions and countries

reduce the transaction costs of borrowing from or partnering with other investors. In their view, however, these and other arrangements spring reliably into place: a country is able to evolve the economic system that is best for it. The Panglossian view that market solutions are efficient because they result from a competitive trial and error process is extended by analogy to government institutions. With the economic system of the Continent having been optimized, so far as one can tell, no improvement can be expected from altering it. This may be named the Rational Institutions theory.3 The same might be thought about a country’s social system. The Continent could be supposed to have its reasons for its social insurance and social assistance entitlements. Indeed some economists of neoclassical persuasion, including some at this conference, hypothesize that the Continent has preferences for work and saving, and indeed economic institutions that are no different from those in a great many other economically advanced countries; the primary difference, they say, is that most people on the Continent have an outsized need for “security,” which their welfare state is well designed to provide. Thus everything in this world is for the best—or for Pareto efficiency, at any rate. There is no free lunch. If institutions are to be changed, it will simply be a matter of whose ox is gored.

Gauging & Explaining Economic Performance in Continental Europe

3

The methodology of this volume (and of the conference from which it originates) holds that in a real-life society the economic institutions are apt to be suboptimal. One of the reasons could simply be that institutional evolution is much too slow to adapt to changing challenges of nature, markets, and knowledge to be able to even approximate the optimum. Another is that the international competition for the best institutional designs is far from a price mechanism that would enable an Invisible Hand to find the optimal institutional solutions. In fact, as Hans-Werner Sinn’s Selection Principle suggests, there is every reason to expect market failure in systems competition. The Selection Principle says that governments correct market failure and carry out activities that cannot easily be handled by markets. Thus, if the market is reintroduced through the back door of systems competition, it is likely to suffer from the same kind of market failure that caused the government intervention in the first place.4 Yet another reason is that no real-life society can know all the illeffects of the economic institutions it has and all the good effects of the economic institutions it lacks, as Thrainn Eggertsson has argued.5 In any society there is likely to be uncertainty (in the sense of Knight) about the net benefit of some of its institutional innovations. There is likely too to be Knightian uncertainty about any further innovation in institutions. Thus a society may accept that there may well “exist” some advantageous redesigns of its social systems, but is at sea over which redesigns would deliver an overall improvement and which would not. Another possibility is that the inefficiency of some of society’s institutions is not in dispute yet it goes on “irrationally” operating with its existing institutions because it cannot agree or muster a majority on which of the less inefficient ones to select. So it is possible for political reasons as well as cognitive ones that the Continent’s systems are “inefficient.” Thus the Rational Institutions theory is not only lacking any foundation; it is highly objectionable as well. 1.1

Comparative Evidence on Economic Performance

Most seasoned and expert observers of the economies of the West— roughly those of the member countries of the OECD—have the impression that most, if not all, of the ones in continental western Europe have performance characteristics that are worse on the whole than do the economies in the rest of the West. It is worthwhile to check such perceptions against the evidence.

4

Edmund S. Phelps and Hans-Werner Sinn

Let us consider the unemployment rates in 1995 and 2007. The data suggest that unemployment rates tend to be higher on the western Continent than the usual comparator countries outside the Continent: Finland, Sweden, Canada, United Kingdom, and United States (see table 1.1). Data on labor force participation rates in the same two periods also suggest a tendency toward less participation on the Continent than in the comparator countries. The difference between the former group, comprising most of the western Continent, and the comparator group could be interpreted as reflecting differences in the mental stimulus, intellectual challenge, and other attractive features found in jobs in the one group and the other. That is, after all, one of the reasons why we are interested in unemployment and nonparticipation: the data tell us something about people’s identification with their work and with the market system as such.6 There is direct evidence, however, on how rewarding, relatively speaking, is the employee experience on the Continent. Personal accounts by employees and journalists describe a workplace that is quite stultifying.7 Table 1.1 Unemployment rate and participation rate of men and women in G10 countries excluding Japan

Country

Unemployment rate in 1995

Unemployment rate in 2007

Participation rate men 2006

Participation rate women 2006 63.9

France

9.9

8.0

74.2

Germany

7.1

6.4

81.4

68.5

11.3

5.9

74.6

50.8

Italy Netherlands

6.8

3.3

81.9

69.4

Spain

18.7

8.1

82.5

61.1

Eurozone without Sweden, Denmark, United Kingdom

10.0

6.8

79.3a

64.2a

Canada

9.5

6.0

82.2

73.5

Finland

16.7

6.6

76.2

73.2

Sweden

7.7

4.6

82.6

77.7

United Kingdom

8.6

5.5

83.2

70.3

United States

5.6

4.6

81.9

69.3

Source: OECD, Economic Outlook 1982, Employment Outlook 2007. a. EU 15.

Gauging & Explaining Economic Performance in Continental Europe

5

A few household surveys collect evaluations of employees in a large number of economies (table 1.2). A highly reputed survey supports the view that the continental workplace is relatively unrewarding. The data on reported “job satisfaction” show a clear and significant difference between the levels on the Continent—particularly France, Germany, Spain, Italy, and the Netherlands—and the levels in most of the comparator countries: Denmark, Sweden, Canada, Iceland, and the United States. A difficulty with the satisfaction data is that respondents who do not regard themselves as “very satisfied” may mean that they feel they are underrated or unjustly treated by their employer. The respondents may not have been thinking of fulfillment, self-actualization, or selfrealization. For this reason it is good to have reported “job involvement,” or employee engagement, which we may presume to be more about the job than the employee. The ranking of the countries in this respect suggests that France, Germany, Italy, and the Netherlands are low, relatively speaking, in job involvement—thus low in mental stimulation and intellectual challenge. France and Germany, which are so Table 1.2 Indicators of economic performance in the G10 from world values surveys: Job satisfaction and employee engagement Country

Job satisfaction

Job satisfaction index

Employee engagement

Employee engagement index

Canada

7.89

0.76

2.70

0.85

France

6.76

0.00

1.74

0.00

Germany

6.98

0.15

1.79

0.04

Italy

7.26

0.34

2.03

0.26

Japan

7.66

0.61

2.20

0.41

Netherlands

7.48

0.49

2.16

0.37

Spain

7.02

0.18

2.31

0.50

Sweden

7.93

0.79

2.63

0.79

United Kingdom

7.42

0.45

2.80

0.94

United States

7.84

0.73

2.87

1.00

Note: Job satisfaction is measured on a scale from 1 to 10, 10 being the highest, and employee engagement (degree of pride derived from the job) is on a scale from 1 to 3, 3 highest; both are averaged for 1990 to 1993 (Human Beliefs and Values, Inglehart et al., Ann Arbor: University of Michigan Press, 1997). The job satisfaction index and the employee engagement index scores are mapping the job satisfaction and employee engagement measures onto the interval [0, 1].

6

Edmund S. Phelps and Hans-Werner Sinn

sophisticated, are at the bottom. The French Revolution and German Socialism still cast a long shadow. A very familiar indicator of economic performance in a quite different dimension is hourly productivity. The more esteemed variant is total factor productivity. (A country might manage to have a higher hourly productivity than another but only by means of having saved much more or by somehow having borrowed overseas much more against future income—either one coming at the cost of past or future consumption, and those differences could mask the lesser efficiency or greater technological backwardness of the former economy.) Yet there are elements of arbitrariness in the adjustment made to go from labor productivity to total factor productivity.8 Furthermore official estimates of factor productivity levels, as distinct from growth rates, are unobtainable. An investigator is forced to “roll his own.” With regard to hourly productivity, it is widely acknowledged that by the middle of the 1990s several countries on the Continent had reached levels in the neighborhood of the highest levels found in the comparator countries. This is shown in table 1.3 But the usual perforTable 1.3 Productivity comparisons

Country

Hourly productivity level 1995 at current $ prices and OECD PPP

Hourly productivity level 2006 at current $ prices and OECD PPP

Hourly productivity index 2006 at constant prices (base 1995)

GDP per person in working age 2006 at current prices and OECD PPP (1,000$)

France

32.2 (41.7)

50.1 (57.2)

120.9

47.8 (54.6)

Germany

31.9 (41.9)

47.1 (52.0)

120.3

48.2 (53.2)

Italy

29.6 (30.5)

38.1 (41.6)

105.6

43.5 (47.5)

Netherlands

33.6 (40.2)

51.5 (57.5)

113.7

54.5 (60.8)

Spain

26.8 (24.9)

39.3 (37.5)

106.7

42.5 (40.6)

Eurozone without Sweden, Denmark, United Kingdom

29.4 (34.5)

43.4 (47.0)

116.2

46.4 (50.3)

Canada

27.5 (24.3)

41.4 (43.8)

119.0

53.4 (56.5)

Finland

26.5 (34.6)

41.1 (50.1)

130.0

49.1 (59.8)

Sweden

28.4 (37.3)

44.7 (55.5)

133.7

53.2 (66.1)

United Kingdom

25.3 (25.5)

41.4 (49.6)

126.0

49.8 (59.6)

United States

31.5

50.4

126.2

64.8

Note: Figures in brackets are at the current exchange rates.

Gauging & Explaining Economic Performance in Continental Europe

7

mance indicators in nearly all the continental economies are now poor compared with the other countries in the G7. The Big Three—Italy most markedly, Germany and, to a lesser extent, France—steadily lost ground to the United States owing to their slowdowns and the US speed-up early in the present decade. In Germany hourly productivity sank to 93.5 percent of the US level in 2006 from 101.2 percent in 1995. Of course, it would give a biased picture to compare with the US level only the levels on the Continent of the standout countries—France, Germany, and the Netherlands. Those countries’ levels might better be compared with the levels in California, New York, and Pennsylvania. In the same spirit one might compare with the average level in the United States the average level in the continental part of the eurozone. This ratio had sunk to 86.1 percent by 2006 from 93.3 percent in 1995 (see table 1.3). Some caution in the interpretation of the data is warranted, however, insofar as international differences of production patterns make unambiguous comparisons difficult.9 The purchasing power parities that are used to make national currency values comparable and to eliminate exchange rate volatility are very sensitive to the composition of the commodity baskets used for the calculations. As a rule, a country’s productivity appears higher in international comparisons if this country’s commodity basket is chosen to calculate the purchasing power parities. (In 2006, for example, the PPP based on the German basket was 1.34 dollars for the euro, while it was 1.15 with the OECD basket and 0.92 with the US basket.) Thus indexes of the growth rate of productivity may have greater reliability. However, as the third column of table 1.3 shows, since 1995 the hourly productivity numbers in the continental European countries all increased less than in the countries in the control group, with the slight exception of Canada, which is outperformed by France. Thus these indexes give the same impression gained by looking at the absolute values. A further comparison, shown in the last column of the table, refers to GDP per person of working age. This comparison confirms and strengthens the impression of European backwardness gained by looking at the productivity figures. France, Germany, and Italy are outperformed by every single one of the comparator countries by a wide margin. The huge gap in productivities could be explained by Europeans’ preference to work less and enjoy their leisure time, but it is more likely the result of inflexible wage structures that create more open and hidden unemployment in Europe.

8

Edmund S. Phelps and Hans-Werner Sinn

As table 1.1 shows, even the open unemployment rates in Europe’s Big Three are much higher than in the comparator group. A high rate of unemployment usually results from a high degree of wage rigidity. The rigidity can be due to strong union influence or due to the explicit or implicit minimum wages resulting from legal minimum wages or a welfare state that pays generous wage replacement incomes. The latter interpretation is supported by table 1.4, which shows that France and Germany suffer from record levels of unemployment among the unskilled. Both countries have strong welfare states that defend high minimum replacement incomes. To the extent that low employment is not a result of individual choice but of institutional rigidities, GDP per person in working age shows Europe’s deficiency even better than hourly productivities. While hourly productivities can be increased by simply increasing wages and wiping out the less productive parts of the labor force, GDP per person capture the deficiencies in the wage formation process. Only this measure avoids gauging a mere selection effect. The impact of wage rigidities on European performance has declined a bit in recent years. Table 1.5 shows that the wage gap between the Continent’s Big Three and the comparator countries declined from 1995 Table 1.4 Unemployment rates among the G10 excluding Japan in 2005

Country

Total (%)

Less than higher secondary educationa (%)

France

8.8

12.4

Germany

9.1

20.2

Italy

7.8

7.7

Netherlands

4.9

5.8

Spain

9.2

Eurozone without Sweden, Denmark, United Kingdom

8.4

Canada

6.8

9.8

Finland

8.4

10.7

Sweden

5.8

8.5

United Kingdom

4.8

6.6

United States

5.1

9.0

9.3 na

Source: OECD Economic Outlook Database, OECD Employment Outlook. a. Persons aged 25 to 64.

Gauging & Explaining Economic Performance in Continental Europe

9

Table 1.5 Wage costs per hour at OECD $PPP Country

Wage cost 1995

Wage cost 2006

France

20.8 (27.0)

25.6 (29.3)

Germany

20.3 (26.7)

27.8 (30.7)

Italy

20.7 (21.3)

24.8 (27.1)

Netherlands

20.8 (24.8)

30.2 (33.7)

Spain

16.0 (14.8)

20.3 (19.4)

Eurozone without Sweden, Denmark, United Kingdom

na

na

Canada

16.3 (14.5)

na

Finland

16.1 (21.0)

24.0 (29.2)

Sweden

na

na

United Kingdom

16.5 (16.6)

26.9 (32.2)

United States

19.3

29.7

Note: Numbers in brackets show wage costs at current exchange rates.

to 2005. Europe’s labor costs used to be much higher, but measured at PPPs or exchange rates alike, they are now roughly in line with those of the comparator countries. It might be asked why, especially in the twenty-first century, many economists in weighing the performance of an economy place huge emphasis on productivity as measured by the quantity of output delivered at the end of the sausage machine. Indeed a provisional finding of happiness research is that, beyond some point, a higher productivity level is not conducive to greater “life satisfaction.”10 One would suppose that the quality of the business experience—conceiving products to produce, the adventure of developing a product, the challenge of evaluating and using new products, and so forth—has become critical to the reward gained from participation in the economy. One answer, no doubt familiar, is that higher productivity makes it affordable for society to pursue new collective goals that would otherwise be too expensive, such as public programs aimed at adding to longevity and tax credits aimed at increased integration of the disadvantaged into the market economy. Another answer is that a higher level of productivity makes it possible for a larger number of people to be able to earn the wage they need to be able to afford to opt for jobs offering mental stimulation and challenge. For this reason it is important to add another indicator of

10

Edmund S. Phelps and Hans-Werner Sinn

performance: the compensation of labor, out of which comes the paycheck and the “social contributions” for the employee. The wage data here suggest that hourly total compensation on the Continent has sunk ever farther than productivity has in the past ten years. The gross hourly wage has sunk farther below the US level—more than has productivity (at constant prices). This strengthened the competitiveness of European workers but came with substantial work dissatisfaction (tables 1.3 and 1.5). What conclusion does this collection of evidence suggest?11 It appears reasonable to infer that the western Continent is not structured for the appreciably higher economic performance found outside the Continent and that there are one or more serious faults in the Continent’s economic system(s) or social system(s), or both, gauged against the best-performing (though undoubtedly imperfect) systems in use outside the Continent—a statement that is not about social manners, high culture, clothes sense, behavior of children, treatment of the elderly, and much else. This proposition, whether true or false, has nothing to do with the current growth rate of the economy (or the growth rate of productivity) in any period. It is a proposition about the potential of the economy as it is structured at a specified time to deliver, under given circumstances, low unemployment, to induce high participation, to offer high job satisfaction and job involvement, and, finally, high levels of productivity. Yet the first reaction to this proposition before taking adequate time to think about it is to say, “how implausible, since this is essentially the same economy that delivered spectacular prosperity, employee engagement and the rest during the ‘glorious years’ from 1955 to 1975.” The error in logic is to judge the economic performance of a country under spectacularly advantageous circumstances against the performance of a comparator country (or countries) under normal conditions. In those “glorious years,” the dust of World War II had just settled and Europeans were America’s low-wage competitors that succeeded in playing the card of absolute cost advantages and swimming in its slip stream. Also Europe experienced rapid population growth with hoards of young and energetic people entering the labor markets. Europe must also have gained by being able to imitate new technologies from those of the United States. While the United States had benefited largely from the influx and capture of talent and know-how before and after World War II, Europe, being deprived of its leading position during the war, was eventually able to recoup. The low-hang-

Gauging & Explaining Economic Performance in Continental Europe

11

ing fruit of unexploited technologies overseas was abundant outside the Continent. All these forces operated to cause rapid growth and high employment for about twenty years. These forces spent themselves, however. By the end of the 1970s wealth on the Continent was fully recovered, the low-hanging fruit was mostly gone, and absolute cost advantages had disappeared due to revaluations and extremely rapid wage increases. With normal circumstances prevailing from the 1980s onward, the Continent’s lackluster performance characteristics, which had been masked, were visible to the naked eye. Still, the idea that economic performance is seriously deficient on the European Continent—and with respect to both dynamism and inclusion—might surprise many readers of the essentially classical histories of the twentieth century. It was a given—from Max Weber to David Landes and Jared Diamond—that Europe was the discoverer of an economic and social system that propelled it into economic development. And it seemed clear that the Continent was ahead in terms of its scientific achievements. The question was how the Europeans did it.12 In much popular commentary the continental economies are still seen as unexcelled. So there appears to be a paradox to reckon with. The paradox is easily resolved, though. Although the Continent got a jump on other countries in finding economic development of a sort, countries outside the Continent soon adopted similar economic systems and found comparable development, a few by 1800, such as England, some others by 1900, such as the United States. In this latter period most countries experienced further evolution of their economic systems, generally for the better. The rudimentary system of Venice and Florence in the 1300s, Germany’s and northern Europe’s Hansa trading system from 1150 to 1650, and Luther’s mass education system had allowed only a rudimentary capitalism at best. While that early period saw the rise of pioneering entrepreneurs (of the Schumpeterian kind) and perhaps some Hayekian conceptual originators, the capital came mostly from kings, princes, dukes, and barons rather than from a whole industry of rival banks lending the people’s wealth, so the system bore only a rudimentary resemblance to Anglo-Saxon finance-capitalism. Even if the Continent managed to avoid misguided or unfortunate moves causing structural damage to its economies, it was possible that its systems would be surpassed by others. Other economic histories— modernist histories in which national decisions may result from misjudgments (i.e., a wrong model) or from uncertainty (e.g., “unanticipated consequences”)—argue that in the twentieth century the Continent did

12

Edmund S. Phelps and Hans-Werner Sinn

make some very misguided moves with consequences yet to be fully assessed. Friedrich von Hayek, having Britain in mind, warned of losses in political freedoms and in valuable economic freedoms that a country risks from adopting the “socialist” or “totalitarian” thinking that took hold in “Nazi Germany” and “Fascist Italy”—from the “naïve trust in the good intentions of the holders of totalitarian power.”13 Herbert Giersch saw many of the vaunted reforms on the Continent in the twentieth century as taking Germany back to the medieval guild; others have questioned the wisdom of labor union representation in the supervisory boards of corporations.14 More recently various economists have made neoclassical, supply-side criticisms of the Continent’s outsize welfare-state entitlements and resulting payroll taxes, its redistributive taxation, its lifetime unemployment benefits, minimum wage laws, and so forth. These essentially modern critiques by Hayek and Giersch, while stimulating, did not go on to explain how the losses in economic freedoms could cost countries something in terms of economic performance— job satisfaction, employee engagement, unemployment, participation, and productivity.15 Much the same can be said of the latter-day criticisms of the Continent of a neoclassical character. Moreover many European economists counterargue that their corporatism works well: the role of unions brings offsetting benefits in heightened solidarity and personal security, which are valuable in their own right, and they in turn boost participation and productivity as well.16 Furthermore the relational banking works well too, and there is competition in the product market. Many Europeans also contend that their welfare state is harmless because it is financed with fiscal efficiency and because it promotes the health of workers. Existing efforts to resolve empirically some of these questions have been few. So there is a crying need to develop the various theses and test them against data. This volume is an effort at understanding and confronting with data the various constituent ideas in the thesis of structural underperformance characteristics of the continental economies—perhaps the first such effort. Part I, after weighing the premise that there are deficiencies in the relative economic performance on the western Continent, asks whether a perception of such a decline reflects only adverse market conditions and adverse market prospects now afflicting several western continental economies—not any underlying differences in economic institutions, social policy or economic culture. Part II addresses ideas pointing to a weakening in the performance of the institutions of the

Gauging & Explaining Economic Performance in Continental Europe

13

economic system. Part III addresses some of the ideas behind the thesis that the problem lies in the social system imposed on the economy rather than in the economic system itself. Part IV looks into the possibility that the problem with the economic system is not so much its economic institutions as it is the cultural attitudes that have taken root in the Continent’s workplaces. (This was the sequence in the 2006 conference.) 1.2

Market Forces: Demographics, Wealth, and Capital

An interesting idea introduced by Robert Gordon into the Venice conference is that the western continental economies have been struck with relatively high unemployment as a result of large increases in the working-age population, mainly coming from eastern Europe. The impact has been an immediate increase in the absolute number of persons in the labor force even if participation rates did not increase (although the new workers are disproportionately young, the participation rate has been pushed up as well). But since finding a desirable job opening takes time, the interim effect is a rise of the unemployment rate. At the same time, with the entry of foreign workers less skilled initially than the nationals, we should expect hourly productivity to head down toward a lower path. Whether this interpretation, inspired by one subset of the evidence, runs into trouble or gains support by looking at other evidence is hard to judge. Consider the evidence on investment goods purchases. One would think that the influx of new workers would stimulate investment in new plants and equipment. And, the boost to investment expenditure would be short but sharp while the boost to consumption expenditure would be small but sustained. These effects do not appear in the data, however. But perhaps other forces intervened. One explanation is that Europe’s welfare states offer high social replacement incomes for domestic residents.17 These replacement incomes act as lower bounds on wages and prevent the creation of additional jobs for the migrants, such that there is immigration into unemployment. However, as the immigrants are not entitled to welfare, they take the jobs, and the domestic residents who remain the marginal suppliers in the domestic labor market are crowded out into the welfare system. A complementary explanation, one derivable from Phelps’s work in the 1990s, focuses on the impressive pileup in the continental countries

14

Edmund S. Phelps and Hans-Werner Sinn

of private wealth relative to wages. After the “glorious years” of rapid growth (starting in 1955) were over and the continental wealthincome ratio had reached cruising level—sometime around 1980—it was a great deal higher than it had been in 1955. The crucial point, however, is that continental households are phenomenally thrifty. As a result the relatively high ratios on the Continent are easily the highest in the OECD. It can be argued that these high levels of private wealth weigh down heavily on the supply of labor, thus male and female participation rates, and have serious ill-effects on employee behavior as well, which in turn push up the natural rate of unemployment.18 An analysis tying the Continent’s slump to its thrift suggests a straightforward test. That is to say, we should expect to find countries’ labor force participation rates negatively associated with their ratio of private saving to GDP; and their propensities to quit and shirk, thus also the rate of involuntary unemployment, positively associated. In fact, in the OECD economies, which are highly open to foreign capital flows, we do find evidence of such associations—statistically significant evidence (see Zoega, chapter 10 in this volume). If slower ex post productivity growth, even when unaccompanied by slower expected growth, operates to increase the wealth-to-wage ratio, then in narrowing participation rates and pushing up the unemployment rate, we should also expect to see that those continental countries that have suffered relatively slow growth among the OECD members suffer relatively high unemployment and low participation, other things equal. There is some support for these associations. Since the early 1990s, Germany, France, and Spain have had slower productivity growth than Sweden, Ireland, and Iceland; the former countries have also had higher unemployment and somewhat lower participation than the latter countries. Of particular note are the sharp productivity slowdowns that struck several economies in modern memory—Holland from 1995 onward, Germany from 1996, Spain from 1997, and lastly France and Italy from 2001. These contrast mightily with the productivity speedup of Ireland after 1999 and that of the United States, Finland, Korea, and Japan after 2002. The slowdowns were generally followed by increases in unemployment rates—Holland 6.3 percent in 2005 from 4.2 percent in 1998, Germany 10 percent from 8.7 percent, Switzerland 4.1 percent from 3.4 percent, Luxembourg 4.6 percent from 3.1 percent, France 10 percent from 8.7 percent in 2001.

Gauging & Explaining Economic Performance in Continental Europe

15

Finally, we would note that some of these growth slowdowns on the Continent may be the product in turn of deeper market forces. The severity of the overhang of pensions (and medical entitlements) came to be perceived and discussed in several continental countries in the second half of the 1990s. As is well known, the future rise in social insurance claims will be huge in relation to the productive capacity of the economy because the bulge of baby boomers nearing retirement is huge, and because medical science has hugely increased the longevity of retirement-age people. Less well known is that, according to recent estimates by the OECD Secretariat, the largest increases in entitlement spending calculated to occur between 2005 and 2050 are found mostly among the continental countries, not those outside of the Continent: 6 percent of GDP in the Netherlands, 5.5 percent in Canada, 5 percent in Germany, and 4 percent in France (which is the OECD average). In contrast, the increase in the United States is only 2.5 percent, in Italy 2 percent, in Japan 1.5 percent, and the United Kingdom 0 percent.19 1.3

A Dearth of Dynamism

This conference was inspired in part by the conviction that the continental economies’ root problem is a dearth of economic dynamism— loosely, the rate of commercially successful innovation. A country’s dynamism, being slow to change, is not measured by the growth rate over any short- or medium-length span. The level of dynamism is a matter of how fertile the country is in coming up with innovative ideas that have prospects of profitability, how adept it is at identifying and nourishing the ideas with the best prospects, and how prepared it is in evaluating and trying out the new products and methods that are launched onto the market. There is evidence of such a dearth. Italy and France, less so the export champion Germany and the finance champion Britain, appear to possess less dynamism than do the United States, Canada and perhaps a few others. Far fewer firms break into the top ranks in the former. And fewer employees are reported to have jobs with extensive freedom in decision-making, which is essential at companies engaged in novel, thus creative, activity. The dynamism possessed by an economy is a crucial determinant of a country’s economic performance: where there is more entrepreneurial activity, thus more innovation and all the financial and managerial activity it leads to, there are more jobs to fill. Because those added jobs

16

Edmund S. Phelps and Hans-Werner Sinn

are likely to be relatively engaging, satisfying, and fulfilling, participation rises accordingly and productivity climbs to a higher path. One of the conference’s leading conclusions is that the dearth of dynamism found on the Continent (though in some countries more than others) is associated with the peculiar sort of “economic model” there. Is the “economic model” operative in the continental countries a major cause—perhaps the largest cause—of their lackluster performance characteristics? 1.4

The Role of Economic Institutions

A country’s performance presumably depends to a large extent on its economic institutions, and unfortunately, the economic institutions on the Continent do not appear to be good for dynamism. These institutions typically exhibit a balkanized/segmented financial sector favoring insiders, myriad impediments, and penalties placed before outsider entrepreneurs, a consumer sector not venturesome about new products or short of the needed education, union voting (not just advice) in management decisions, and state interventionism. Some studies on what attributes determine which of the advanced economies are least vibrant—least responsive to the stimulus of a technological revolution—have pointed to the strength of inhibiting institutions, such as employment protection legislation and bureaucratic “red tape,” and to the weakness of enabling institutions, such as a wellfunctioning stock market and ample liberal arts education. It seems to many economists that the company law and labor law found on the Continent cannot be healthy for economic dynamism.20 This conference with its session on economic institutions focused instead on understanding what determines which of the advanced economies are least creative in generating new commercial ideas or least adept at selecting the ones with the greatest commercial promise. Of course, this challenge invites study of all the sectors of the economies being compared. However, it is fair to say that this session has paid special attention to entrepreneurship and what may be called “financiership.” Chapter 2 by Roman Frydman, Omar Khan, and Andrzej Rapaczynski addresses the strength of the entrepreneurial spirit in Europe and also the influence of corporate governance and of the financial sector. The authors note the paradox that a far greater proportion in the United States would prefer to be self-employed than in the Netherlands,

Gauging & Explaining Economic Performance in Continental Europe

17

Belgium, Denmark, and France; Spain, Italy, Germany, and Austria are between these extremes. And significantly more Americans do undertake steps to start a business. Yet the number of business owners as a percentage of the economically actively population is actually smaller in the United States than in Italy, Spain, Belgium, and the Netherlands; and not much greater than in Germany, Austria, and France. The authors believe one reason to be that new entrants face more competition in the United States and thus suffer higher failure rates (New York restaurants are a famous example) than start-ups in continental Europe. Another reason, they argue, is that where Europeans open shops or service outlets, Americans “more often venture into more high-tech start-ups offering great potential rewards but also greater chance of failure.” (Thus the new firms in the United States that succeed grow to be far bigger than they were early on, which is not the continental pattern.) While acknowledging that employment protection legislation may play a significant role in explaining why continental employees are less likely to quit to start up a new business (they have more job protection to lose), they explore hypotheses that differences between the Continent and the United States with respect to the ownership structure, the accountability of managers, and level of CEO compensation and the incentives it provides all help explain a greater demand for innovation in the United States than on the Continent. Very possibly this interpretation is part of a larger story: the widespread sense in the United States and Canada that “you can ‘make it’ on your merit” in contrast to the common impression on the Continent that merit is not enough: you can’t get onto a ladder to success without “connections,” without knowing the grand patron. If this is broadly so, high dynamism in an economy not only boosts economic inclusion, as innovative projects create added jobs up and down (especially down) the wage scale; such dynamism also contributes to upward mobility. Chapter 3 by Richard Robb has several themes of which the most provocative is that the size of the venture capital industry is not large in any of the western economies. Moreover the amount of venture capital supplied to the typical recipient is strikingly small. The largest supply of funding for young firms comes from the angel investors of an economy, not the venture capitalists. The right view of these facts appears to be that angel capital (i.e., from friends and family) is the prevailing source of external funds at the start of a young firm’s life; venture capitalists come in with their commercial expertise, many of them having once been entrepreneurs themselves, when the firm has

18

Edmund S. Phelps and Hans-Werner Sinn

established “proof of concept” and begins an operational stage in which it needs more capital and more customers. These findings are complemented in chapter 4 by Anders Hoffmann. While Hoffmann does not see a European deficiency in terms of the number of start-ups, he observes that companies once founded tend to grow slowly in Europe, unable to match the dynamism of young and successful firms in the United States. The lack of a functioning venture capital market in Europe is an obvious candidate for an explanation. Yes, there are stock markets in Europe, but as the equity capital market below the stock market is underdeveloped it is difficult for promising ideas to find proper funding. Robert Shiller in chapter 5 even argues that Europe urgently needs better derivatives markets, because derivatives markets would enable rapid growth by fostering entrepreneurship and risk-taking. Europe, being short of such markets, faces a serious obstacle to capitalist development. While most conference participants seemed to share the view that American financial capitalism has the better ingredients in the policy mix that leads to economic growth, Harold James in chapter 6 presents an alternative viewpoint by praising Europe’s family-owned firms. Such firms may have difficulties in finding enough capital, but they do seem to provide more stability in turbulent times, he argues. According to James, one of the reasons for Europe’s quick recovery after the war was Europe’s entrepreneurial families, and he expects the family firms to help Europe cope with the current forces of globalization. We cannot touch on all aspects discussed in the session on economic institutions. But it would be a pity not to note another dimension— beyond the entrepreneur, his or her employees, and the financier. In what was ultimately to become an influential paper, Richard Nelson and Phelps emphasized that the diffusion of a country’s new method, such as a new fertilizer, or a product, such as a new tractor, depends on the decisions of managers to adopt the novel development; and their willingness to be among the pioneering users will depend on their preparation to make such a decision—their education and their past experience. An obvious corollary, though not explicit in that paper, is that if no managers can confidently be expected to evaluate and, in some cases at any rate, to try the new method or product, entrepreneurs will not be willing to develop innovations—leaving aside the possibility of launching in some overseas markets.21

Gauging & Explaining Economic Performance in Continental Europe

19

In a similar spirit Amar Bhidé argues in chapter 7 that “venturesome” consumers are fundamental to a country if it is to be home to home- or foreign-made innovations. Innovations will not be made in a country (though suppliers of them may be able to find markets in other countries) if there is not a spirit of curiosity and experimentalism on the part of consumers. Thus innovation is the outcome of a multiplepart mechanism involving the conceiver of the new idea, the entrepreneur (often the conceiver or his partner) undertaking its development, the employees providing the essential work for its development, the financiers for needed funds for development and launch, and a market of potentially receptive users of the new product or method—consumers as well as managers. This is a challenge, as Bhidé notes, to the economic policy of what Nelson and Ostroy have dubbed “technofetishism,” which imagines that the state or large industrial labs run by established corporations can be depended on to generate dynamism. They can bring new products and new methods to the market, but they cannot make the market buy them. Innovations, of course, are aided by education and investment in human capital. This is the point that Philippe Aghion and Ioana Marinescu make in chapter 8. They argue that one of the most important sources of growth is quality-improving innovations and that such innovation is most likely to result from postgraduate education, which seems to be underdeveloped in Europe. This view is complemented in chapter 9 by James Heckman and Bas Jacobs on the role of skill formation. These authors also point to the overwhelming importance of skills for economic growth, but unlike Aghion and Marinescu, they argue that skill-forming policies should start very early in a person’s life. They are more effective in this period, and they avoid the usual equity-efficiency trade-off in economic policy decisions because creating more skilled people means both more economic prosperity in general and less inequality at the same time, given that the abundance of skills would reduce the skill premium. Chapter 10 on education by Gylfi Zoega, likewise stresses the dominant role of tertiary education for economic growth. However, Zoega also emphasizes other factors such as unionization of the labor force. He is able to show that European unions have been very detrimental to economic growth, in particular when they were not coordinated. This brings us to another dimension of the European puzzle, namely the role of social institutions.

20

Edmund S. Phelps and Hans-Werner Sinn

1.5 Social Policy: Entitlements of the Welfare State The puzzle posed in chapter 11 by Robert Gordon is a good starting point for the discussion. He wonders why Europe’s income per person never exceeded 75 percent and has since fallen below 70 percent of the US level even though the level of productivity almost reached that of the United States in 1995. How could Europe become reasonably productive and yet perform so poorly in so many respects? Gordon’s proposed answer is that hours per person in Europe have fallen drastically in the past forty years, reflecting long vacations, high unemployment, and low workforce participation. The answer is convincing but begs the question why in turn employment was so low. Was it due to a particular European preference for leisure, as some have argued, or was it due to institutional deficiencies? Economists who lean toward the latter explanation tend to attribute the Continent’s higher unemployment and lower participation, if not also the lower productivity, to the Continent’s social model—in particular, the plethora of social insurance entitlements and the taxes to pay for them. The standard argument, which sees the harm as coming from a high tax rate on payrolls, does not fit the facts, though. If it were true, unemployment should be highest among high-income people whose marginal tax rate is highest. In fact unemployment is concentrated among low-income people in most countries of the Continent. Germany is a particularly good example for this, as it is the world’s champion in terms of the unemployment rate of the unskilled according to OECD statistics. (Compare table 1.4, second column.) Jeffrey Sachs in chapter 12 goes further, arguing that neither the indictment of the welfare state by the neoclassical Chicago school nor the indictment of corporatists and socialists by the Austrian school is weighty enough to warrant a trial. Sachs reaffirms his previously expressed views that both European corporatism and European welfarism are strengths of the continental economies. The explanation given in chapter 13 by Hans-Werner Sinn is that the European welfare state effectively imposes implicit minimum wages by offering fixed wage replacement incomes in addition to normal unemployment benefits that relate to people’s previous earnings. The fixed wage replacement incomes act as lower bounds to the wage distribution pushing this distribution up from below like a hanging harmonica. Via a substitution chain, all wages up to medium ranges

Gauging & Explaining Economic Performance in Continental Europe

21

are pushed upward creating unemployment, but clearly the effect is strongest for the lowest incomes, which are usually associated with the unskilled. Sinn shows that this lower bound to the wage distribution not only implies unemployment and low participation, in general, but also pathological reactions of the economy that laymen tend to interpret as virtues. One such implication is that the Heckscher–Ohlin mechanism implying sector shifts from labor-intensive to capital- and skillintensive sectors, when trade with low-wage countries is opened, is too strong. Too much capital and skilled labor is migrating into the skilland capital-intensive export sectors, creating mass unemployment among the unskilled, a high level of value added in exports, and sluggish aggregate growth at the same time. And as the skill and capital intensive sectors include the downstream segments of vertical production chains, there is also too much vertical segmentation in trade, a phenomenon that Sinn calls the “bazaar effect.” The bazaar effect makes export quantities increase even faster than value added earned in exports. The puzzle that Germany is the champion in commodity exports while it suffers from high unemployment and has had the second-lowest growth performance since 1995 among all European countries next to Italy might be resolved by this explanation. In chapter 14 Hian Teck Hoon points out another flaw in the view that high payroll taxes may have caused low European labor force participation. When the tax rate on payrolls is increased, the consequent reduction of after-tax wage rates is not necessarily an enduring disincentive to work, he argues, as the reduced earnings will bring reduced saving and thus falling private wealth. Once private wealth has fallen to its former ratio to after-tax wages, people may be just as motivated to work as before. An indictment of entitlements, then, has to focus on the huge “social wealth” that the welfare state creates at the stroke of the pen. Yet statistical tests for the effects of welfare spending on the unemployment rate and the activity rate (employment in percentage of the working age population) have yielded erratic results. Whatever the final judgment may be on the employment effect of welfare entitlements on unemployment and on the activity rate, it is hard to see that scaling down entitlements would be transformative for economic performance. (Indeed some economists see increased wealth, social plus private, as raising the population’s willingness to weather market shocks and helping entrepreneurs to finance innovation.)

22

Edmund S. Phelps and Hans-Werner Sinn

Yet times change. Our theoretical models usually postulate an economy with closed borders to labor flows. In recent years, though, France and Italy have suffered substantial outward migration of young people to countries where tax rates and social benefits are lower. When France began adding benefits upon benefits to its welfare state, it was assumed that the workforce in France had nowhere else to go. That appears to be no longer so. Still, most working-age people have ties to their country that are difficult to break. 1.6

The Role of Economic Culture

The other part of the economic model, a part possibly no less important than the institutional part, consists of various elements of the country’s economic culture—attitudes toward work, wealth, change, and the rest and how widespread those attitudes are. Some cultural traits in a country could have direct effects on performance—on top of their indirect effects through the institutions they foster. Values and attitudes are analogous to institutions—some are impeding, others enabling. They are as much a part of the “economy” and possibly as important for how well it functions as the institutions are. Clearly, any study of the sources of poor performance on the Continent that omits that part of the system can yield results only of unknown reliability. Chapter 15 by Edmund Phelps reports evidence from University of Michigan “values surveys” that working-age people in the Continent’s Big Three differ somewhat from those in the United States and the other comparators in the number of them expressing various “values” in the workplace. The values that might have an impact on dynamism are of special interest here. Relatively few persons in the Big Three report they want jobs offering opportunities for achievement (42 percent in France and 54 percent in Italy, an average of 73 percent in Canada and United States), chances for initiative in the job (38 percent in France and 47 percent in Italy, an average of 53 percent in Canada and the United States), and even interesting work (59 percent in France and Italy, an average of 71.5 percent in Canada and United Kingdom). Relatively few are keen on taking responsibility or having freedom (57 percent in Germany and 58 percent in France as against 61 percent in the United States and 65 percent in Canada), and relatively few are happy about taking orders (Italy 1.03—of a possible 3.0— and Germany 1.13 as against 1.34 in Canada and 1.47 in the United States).

Gauging & Explaining Economic Performance in Continental Europe

23

Presumably in countries where that spirit is weak, an entrepreneurial type contemplating a start-up might be scared off by the prospect of having employees with little zest for any of those experiences. And there might be few entrepreneurial types to begin with. Fortunately, we do not have to take for granted—to presume—that a spirit of the new, a liking for challenges, the enjoyment of problem-solving, and so forth, have impacts on a country’s dynamism and thus on its economic performance. Phelps’s paper reports findings that intercountry differences in each of the several performance indicators are significantly explained by some of the intercountry differences in cultural values— not necessarily the same ones for every indicator, which is as it should be. (Nearly all those cultural attitudes have a statistically significant influence on some or most of the indicators.) The weakness of these workplace values on the Continent is not the only impediment of a cultural nature to a revival of dynamism there.22 There is also the ruling economic “philosophy.” There is the solidarist aim of protecting the “social partners”—communities and regions, business owners, organized labor, and the professions—from disruptive market forces; also the consensualist aim of blocking business initiatives that lack the consent of the “stakeholders”—those with a stake besides the owners, such as employees, customer and rival companies. There is an intellectual current of elevating community and society over individual engagement and personal growth, which springs from anti-materialist and egalitarian strains in Western culture. There is the “scientism” that holds that state-directed research is the key to higher productivity. There is the tradition of hierarchical organization in continental countries. Last, there is a strain of anticommercialism. (This attitude was conspicuously displayed by royalty. Louis XIV told his son “we don’t make the money, we marry into it.” Snobbish commoners might have aspired to the same lofty attitude toward industry and labor.) An economic system built on this philosophy might induce the workplace values recorded in the Michigan surveys. Or these workplace attitudes might predispose policy to embrace the social philosophy just described. Either way, workers are led to accept—uncomplainingly— jobs lacking freedom, responsibility, and interestingness. Then the state does not see a popular demand for an improvement in economic policy; so the absence of freedom, responsibility, and interestingness goes on. If that is right, economic culture and economic philosophy are of some importance in determining economic performance.

24

Edmund S. Phelps and Hans-Werner Sinn

Appendix Table 1.A1 Subjective indicators of economic performance: Job satisfaction and employee engagement

Country

Job satisfaction

Job satisfaction index

Employee engagement

Employee engagement index

Austria

8.03

0.86

2.03

0.26

Belgium

7.79

0.70

2.19

0.40

Canada

7.89

0.76

2.70

0.85

Denmark

8.24

1.00

2.76

0.90

Finland

7.56

0.54

2.51

0.68

France

6.76

0.00

1.74

0.00

Germany

6.98

0.15

1.79

0.04

Iceland

7.87

0.75

2.75

0.89

Ireland

7.81

0.71

2.74

0.88

Italy

7.26

0.34

2.03

0.26

Japan

7.66

0.61

2.20

0.41

Netherlands

7.48

0.49

2.16

0.37

Norway

7.88

0.76

2.55

0.72

Portugal

7.42

0.45

2.65

0.81

Spain

7.02

0.18

2.31

0.50

Sweden

7.93

0.79

2.63

0.79

United Kingdom

7.42

0.45

2.80

0.94

United States

7.84

0.73

2.87

1.00

Note: Job satisfaction is measured on a scale from 1 to 10, 10 being the highest, and employee engagement is on a scale from 1 to 3, 3 highest; both are averaged for 1990 to 1993 (Human Beliefs and Values, Inglehart et al., Ann Arbor: University of Michigan Press, 1997). The job satisfaction index and the employee engagement index scores are mapping the job satisfaction and employee engagement measures onto the interval [0, 1].

Notes 1. In Georgics, a poem on Roman farming and culture, Virgil characterized the pater familias who lives from managing his land and his time and who needs knowledge to do it well. 2. We choose 1995 as the starting year because it comes after the dust has settled in the postcommunist countries and before the disruptions of the Internet revolution. 3. See Donald Wittman 1995, The Myth of Democratic Failure: Why Political Institutions are Efficient, American Politics and Political Economy series (Chicago: University of Chicago Press).

Gauging & Explaining Economic Performance in Continental Europe

25

4. See Hans-Werner Sinn 1967, The selection principle and market failure in systems competition, Journal of Public Economics 66: 247–74. See also Hans-Werner Sinn 2003, The New Systems Competition, Yrjö Jahnsson Lectures (Oxford: Basil Blackwell). 5. See Thrainn Eggertsson 2005, Imperfect Institutions: Possibilities and Limits of Reform (Ann Arbor: University of Michigan Press). 6. It may be that average tenure in jobs is longer on the Continent than in the United States and other comparator countries, but the longer tenure surely reflects the lower hiring rates, which are in turn attributable to the higher costs of terminating a labor contract under so-called employee protection legislation. There is no reason to infer from the higher average tenure on the Continent a greater degree of employee engagement and job satisfaction. 7. An excoriating account of employee life in a French corporation, EdF, is Corinne Maier 2004, Bonjour paresse (Paris: Editions Michalon). 8. Of course, the latter country might have owed its greater total factor productivity to its having sacrificed consumption to build more or better economic institutions, rather than more or better tangible capital—and we do not “factor out” institutions to obtain some net productivity estimate though we do factor out tangible capital. Nevertheless, many of the institutional advantages we observe in some countries resulted from the historical accident of making right choices rather than wrong ones in the building of institutions. 9. There are other problems resulting from differences in the way countries generate their data. In France, for example, wage scales exclude its least prepared from work, thus raising measured productivity. 10. Conceivably the explanation for that finding, assuming it will stand up, is that when a country invests a higher share of domestic output, the resulting increase in the level of labor productivity in absolute terms and relative to the level in the other high-productivity countries do not lead to increased consumption if the country’s capital–output ratio is in the neighborhood of the golden rule state. 11. There may be interrelationships among the performance indicators above so that the evidence is less broad than might be supposed. Weak job satisfaction and employee engagement on the Continent could be a proximate cause—though very possibly not the underlying cause—of the poorer participation and higher unemployment rates. High unemployment on the Continent could have led to a mismatch of worker to job, thus weakening job satisfaction and employee engagement. In any case, our task is to find the underlying cause(s) of the entire syndrome of Continental complaints: the high unemployment, middling participation rates, lagging productivity, and deficiencies in employee engagement and job satisfaction. 12. Weber and Landes both presume that Europe chanced upon institutions and cultural traits that fitted them for economic success, the former pointing to Protestantism in Germany and Landes to “European exceptionalism.” See Max Weber 1904–1905, “Die protestantische Ethik und der Geist des Kapitalismus,” Archiv für Sozialwissenschaft und Sozialpolitik, XX-XXI (English trans. Talcott Parsons, The Protestant Ethic and the Spirit of Capitalism, London, Allen and Unwin, 1930); David S. Landes 1998, The Wealth and Poverty of Nations (New York: Norton). A quite different thesis is argued in Diamond 1998, Guns, Germs and Steel (New York: Norton). In an interesting recent work, Becker and Wößmann argue that the German success can primarily be traced to the fact that the Protestant regions decreed general schooling as early as in the sixteenth century, as Luther wanted

26

Edmund S. Phelps and Hans-Werner Sinn

the people to read the Bible themselves rather than relying on the interpretations of the Catholic church. See Sascha Becker and Ludger Wößmann 2007, Was Weber wrong? A human capital theory of Protestant economic history, CESifo Working Paper 1987, 2007, and Sascha Becker 2007, Protestantism, human capital and economic success, (Habilitation thesis, Faculty of Economics, Ludwig Maximilians University Munich, submitted September 25). 13. Friedrich von Hayek 1944, The Road to Serfdom, London: Routledge (quotation from the Preface to the 1976 reprint edition). Other readers, depending on the chapter they focus on, see the book as an indictment of socialist planning, redistribution, social security programs, and an oversize public sector. 14. Herbert Giersch 1993, Openness and incentives, in Openness for Prosperity (Cambridge: MIT Press, ch. 9). 15. The tacit argument in Road appears to be that under the “corporativism” of Nazi Germany and Fascist Italy the state arrogates to itself many of the decisions on innovation and on investment generally without having the wealth of private knowledge possessed by participants. 16. For two recent contrasting views of Europe’s postwar performance see Barry Eichengreen 2007, The European Economy since 1945: Coordinated Capitalism and Beyond (Princeton: Princeton University Press), and Hans-Werner Sinn 2007, Can Germany be Saved? The Malaise of the World’s First Welfare State (Cambridge: MIT Press). 17. Hans-Werner Sinn 2005, Migration and social replacement incomes: How to protect low-income workers in the industrialized countries against the forces of globalization and market integration, International Tax and Public Finance 12: 375–93. 18. Edmund Phelps 1994, Structural Slumps: The Modern Equilibrium Theory of Unemployment, Interest and Assets (Cambridge: Harvard University Press). 19. Pension tensions, Financial Times, table, p. 11, Monday November 28, 2005. 20. Edmund Phelps and Gylfi Zoega 2001, Structural booms: Productivity expectations and asset valuations, Economic Policy 32 (April): 85–126. 21. Richard R. Nelson and Edmund S. Phelps 1966, Investment in humans, technological diffusion and economic growth, American Economic Review: Papers and Proceedings 56 (May): 67–75. 22. An early discussion is contained in Phelps 2006, The Continent’s high unemployment: Possible institutional causes and some evidence, in M. Werding, ed., Structural Unemployment in Western Europe: Reasons and Remedies, CESifo Seminar Series. Cambridge: MIT Press, pp. 53–74.

2

Entrepreneurship in Europe and the United States: Security, Finance, and Accountability Roman Frydman, Omar Khan, and Andrzej Rapaczynski

2.1

Introduction

Growth patterns in Europe since the 1970s present a worrying trend. In the postwar years Europe grew much faster than the United States, but it started from very low levels (45 percent of the US growth rate), and Europe’s very fast growth was in large part a catch up after World War II destruction. Economic growth was facilitated by the fact that even though the war had devastated the capital stock of Europe and killed millions of people, the remaining human capital, in terms of education level and job training, was much higher than that in any developing country. Athough the long years of spectacular growth ended around 1975, Europeans, with the partial exception of Britain, now looked very optimistically into the future and were more than ever determined to push ahead with the buildup of the most extensive welfare-state system in the world. So, just as the growth rates started to decelerate, the rate of growth of the expanding welfare benefits—medical care, state-funded pay-as-you-go retirement, extended vacations and maternity leaves, dramatic expansion of higher education, and housing subsidies— rapidly accelerated. All the while the economic system was assumed to be primed for further growth and not in need of any fundamental reforms. Around the same time the mood in the United States became much more despondent. The failure in Vietnam, the collapse of the Bretton Woods system built on US economic and financial dominance, the military catch-up of the Soviet Union to a co-equal position as a superpower, the first oil crisis, and the perceived loss of competitiveness of US manufacturing—all these made the United States look like an empire in the throes of decline. In response, beginning with deregula-

28

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

tion in the late 1970s, and, more intensively, with the reforms of the 1980s, the United States went through a serious retooling of its economy and a rapid change in its attitude to income redistribution, resulting in a shake-up of corporate structures and governance, a change in the tax system, and a release of both more market forces and greater entrepreneurship. Whether or not as a result of these changes, the United States has grown consistently faster than Europe since the early 1980s, and the difference, measured in GDP terms, has become progressively greater (see figure 2.1). What do we make of this disparity between the United States and Europe? The most worrisome possibility is that we are dealing with a secular trend. Thirty years is a long time, and it may be supposed that there is a systemic difference between the two regions and that unless Europe reforms, perhaps dramatically, the trend may very well continue indefinitely and relegate Europe to a second-rate economic status. It is possible, of course, to say that GDP growth is only one dimension of overall social prosperity, along with such things as social stability, equality, social cohesion, the amount of available leisure time, quality of the environment, or even various “cultural” factors, such as high levels of humanistic education and cultural consumption that do not easily translate into economic terms. In this broader context it may be conjectured that the Europeans have simply traded lower levels of GDP growth for higher quality of life as measured by other metrics. Whether this is a sustainable choice in the long run, given the inter4 3.5 3 2.5 2 1.5 1 0.5 0

EU-15 growth 1970–1980

1980–1990

1990–1995

EU-25 growth 1995–2000

US growth 2000–2005

Figure 2.1 EU and US average real growth, 1970 to 2005 (constant prices 1995). 2005 data are forecasts (sources: Altomonte and Nava 2005 on the basis of Sapir et al. 2004 and Eurostat data)

Entrepreneurship in Europe and the United States

29

dependence of world economies, is another matter that need not be settled here. But we should also consider another possibility, namely that the last thirty-year period, long as it is, may not really reflect any secular trend but rather be a phase in what is essentially a long-term cyclical development. Economic history may contain two different types of periods: times of innovation, change, and retooling; and periods of consolidation, characterized by more gradual technical and efficiency improvements. In this context the early 1980s may not have been just a watershed when Europe completed its catch-up phase and the United States reformed its economy. The period also involved a broader underlying technological and social change. The development of computerization, biotechnology, rapid advances in medicine, the move of the advanced economies away from manufacturing and toward technology and service-oriented economies, the rapid growth of some third world countries that started to compete effectively in many areas (mostly manufacturing) previously dominated by the advanced countries produced a very different economic environment than in the 1950s, 1960s, and 1970s, one in which a very high premium is placed on the ability to adjust quickly and change rapidly in response to events. But this new environment may also be a phase of a historic cycle: the pace of innovation may slow down again after a while, to be followed by a long period of a more gradual development during which new production techniques will be perfected, the production cost will be incrementally reduced, and a number of local efficiencies will be gradually implemented. (A similar cycle had taken place in the early decades of the twentieth century when the period of enormous change related to the inventions of electricity, internal combustion engine, and the rise of the modern corporation, was followed by a long period of gradual development and consolidation.) Corresponding to the two different types of periods in economic development, there may be two different types of economies, each better suited than the other to deal with the challenges of one or the other phase of the cycle. One type (that of the United States) may be more dynamic, more easily adapting to change and innovation, but also imposing more insecurity on its workforce and providing generally lower levels of social (though not political) stability. The other type (characteristic of continental Europe and Japan) may be better adapted to periods of consolidation and gradual perfection of the basic economic institutions. The greater workforce and institutional (including

30

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

corporate) stability, greater insistence on technological rather than financial and managerial, competence, greater ability of corporations and financial institutions to commit to long-term projects with lesser short-term returns may, among other factors, make European economies do better in times when the pace of innovation slackens and a period of consolidation sets in. After all, just when the wheels of fortune were about to change so dramatically around the early 1980s, most institutional economists were expanding on the alleged superiority of European and Japanese financial and corporate institutions in comparison with their American counterparts. Indeed many observers predicted that the Japanese justin-time production methods and the German and Japanese long-term perspective bank financing would bury the American economy, too much focused on short-term profits and shoddy, inefficient production. It may thus very well be that times are going to change again and Europe’s institutional system may be more compatible with the next phase of economic development. Whether the present lag in Europe’s performance is a secular trend or a phase in a cyclical development has, of course, significant implications for the future. If Europe is long-term failing to grow, it must either effect far-reaching reforms or lose its preeminent position in the world of the future. If Europe is choosing to put more emphasis on broader measures of the quality of life, rather than focusing primarily on GDP growth, the main question is whether this choice is viable in a globalized economy and whether Europe can preserve its way of life in the long term, rather than slip to a second-rate status with respect to most (and not just GDP) indicators. Or, Europe may be going through a phase for which its institutions are less well suited, but which will give way to a new period in which Europe can recover. So Europe may need to take some steps to minimize its losses in the present period, though it may not need to remake itself in the image of the more dynamic American economy. Although it may be important to question Europe’s emphasis on the quality of life rather than change and GDP growth that it engenders, we do not take a position on which of these scenarios is the most likely to describe the challenges to be faced by Europe in the future. Instead, we want to focus on why Europe is not performing well now and why, if objective circumstances do not change and the pace of technological and other developments does not slow down, Europe is not likely to do much better.

Entrepreneurship in Europe and the United States

31

While the reasons for the slow European rate of growth, as compared to the United States, are certainly multifaceted and the phenomenon does not have one explanation, we will focus on an aspect that, in our opinion, is not often discussed, perhaps because it does not fit well within the dominant framework of contemporary economic analysis.1 It concerns the different levels of entrepreneurship across the Atlantic divide. An important, perhaps decisive, element of economic growth comes from a society’s ability to search for and pick new projects with high economic returns but also high amounts of (often undiversified) risk. Moreover, at the frontiers of growth, when creativeness and innovation are the real driving force of growth, and imitative behavior is not sufficient to create new value, the risk involved in many projects is often not known, and perhaps even unknowable, with any degree of precision. The presence or absence of institutions allowing a society to make decisions about such projects and use the large amounts of local and often ineffable skills of various economic actors may be decisive in terms of the chances of economic success. We understand entrepreneurship to be a central factor in this process. In order to grow, especially in a fast-changing economic environment, a society must therefore develop institutions that allow and encourage entrepreneurial decisions. This involves, in particular, three aspects on which we want to focus here and which we think may play a role in explaining why Europe grows more slowly than the United States. First, because all entrepreneurial decisions involve risk, a society must allow and encourage people to make risky decisions—this means that economic actors must expect their returns from riskier projects to include a risk premium that will compensate them for the risks they undertake. But the risk premiums offered by different societies for projects with similar risk characteristics may differ for a variety of reasons: states may tax upside returns at different rates, there may be various nonpecuniary rewards (e.g., prestige accorded to people who make money) that some societies provide to a greater extent than others, and, depending on the base level of security guaranteed by different societies, the opportunity cost of becoming an entrepreneur may differ as well. Different societies may also provide financing for entrepreneurial projects at different cost. In what follows we will present some evidence in support of the proposition that the agents’ incentives to make risky decisions are better in the United States than in Europe and that this may in part account for the growth rate disparity.

32

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Second, people in certain environments may differ in terms of their risk preferences. If the French, for example, are by and large more risk averse than the Americans, this may result in a less risky “portfolio” of potentially valuable projects that will be pursued in the French economy (and hence yield lower overall returns). What shapes the risk preferences of people in various societies is a complex question that we do not purport to unravel. But we hypothesize that the Europeans are indeed somewhat more risk averse than the Americans and that part of this aversion may be an effect of living in an environment in which prolonged disincentives to make risky decisions becomes entrenched in many people’s risk preferences. This means that reforming certain aspects of European economies may be made more difficult by the fact that, even if the disincentives were to be removed, it might take some, perhaps considerable, time for people’s preferences to evolve as well. Third, projects often have to be picked under the circumstances in which it is not objectively known what the probability distribution of their returns might be. No one has become very rich by pursuing projects that everyone knows to be lucrative—widely known opportunities are quickly arbitraged away; indeed there are no such widely known lucrative projects, and economists are right that money does not normally lie in the street. What distinguishes successful entrepreneurs from other wannabes is that they see higher returns or lower risks in a project that others do not consider worthwhile or do not even consider. Moreover this difference of opinion between successful entrepreneurs and their less successful competitors is often not a matter that concerns something that can ever be objectively known. Much like the ability to ride a bicycle, entrepreneurship ability may involve ineffable skills or what some have termed “tacit knowledge” (Polanyi 1962) that cannot be communicated or, even less, explained to others but that turns out to be correct. Indeed, were we able to specify some general rules, objectively reproducible criteria of how entrepreneurs arrive at their insights or what they do, the advantages of entrepreneurship would disappear, to be replaced by a system of rules that could be followed by anyone. This last feature of entrepreneurship—that decisions are based, at least in part, on “hunches” poses a potentially serious problem for the institutional design of a society keen on encouraging and picking up risky entrepreneurial projects. In particular, it is very difficult to accommodate entrepreneurship in any institutional setting in which an aspiring entrepreneur needs to convince others to provide financing

Entrepreneurship in Europe and the United States

33

or other approvals required to proceed with his projects. The very idiosyncrasy and ineffability of entrepreneurial skills means that financial backers will be hard to persuade by an aspiring entrepreneur and that members of decision-making bodies are not likely to allow quirky projects to proceed. It is our hypothesis in what follows that, on the whole, American institutions are better suited to deal with these types of problems than their European counterparts. In particular, we believe that the financing of new ventures in America, as well as the governance systems of American corporations, are superior in this respect to their European counterparts. One more caveat before we proceed. What follows is more of an essay with illustrations than a tightly constructed economic theory. Entrepreneurship is a multidimensional concept; it is not well captured by the prevalent economic theories, which tend to assume away all imperfections of human knowledge and leave no room in their models for real creativity and innovation that lie at the center of entrepreneurial behavior.2 Even to the extent that entrepreneurship could be modeled, it is not our ambition to do so here, as much as propose some suggestions about the various factors that may help account for the disparities of economic performance on the two sides of the Atlantic. 2.2 The Choice of Entrepreneurship 2.2.1 Risk Taking in Europe and America When asked if they prefer being an employee or being self-employed, nearly 70 percent of Americans, as opposed to barely over 40 percent of the French, choose the second option. Table 2.1 presents the results of a Eurbobarometer survey asking the same question in a number of countries, and shows that Americans choose self-employment more often than most other people. The same Eurobarometer study shows that almost 60 percent of EU citizens have never considered setting up a business, as compared with slightly over 40 percent of Americans. Young Americans (defined as individuals under 30) express preference for self-employment more often (66.31 percent) than their counterparts in Germany (59.60 percent), Hungary (54.22 percent), Ireland (50.44 percent), Netherlands (40.29 percent), and Norway (31.68 percent) (Blanchflower and Oswald 1998). Similarly a greater percentage of business school graduates in the United States aspire to entrepreneurial careers than those in Germany (Luthje and Frank 2002) and Spain (Uslay, Teach and Schwartz 2002).3 Nascent entrepreneurs and new business owners in the United States

34

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Table 2.1 Choice of entrepreneurship Country

Percentage of respondents preferring to be self-employed in 2004

United States

67

Portugal

63

Spain

62

Italy

52

Germany

46

Austria

46

France

41

Denmark

37

Belgium

36

Netherlands

35

Finland

23

Sweden

15

Source: Eurobarometer 160 (2004).

are also more confident in their skills and more optimistic about good business opportunities (Kollinger 2005). Given these data, one might expect that a greater percentage of the population would be self-employed in the United States than in Europe. But such is not consistently the case. To be sure, Americans put their money where their mouth is: as table 2.2 shows, significantly more Americans than Europeans undertake some steps to start a business. But the same table also shows that the number of established businesses is comparatively small in the United States, and table 2.3 shows that the number of business owners as a percentage of economically active population is not significantly larger in the United States than in many European countries. Why, then, given the much larger number of start-ups and the greater willingness of Americans to go into business for themselves, is the proportion of business owners in the United States no larger than in Europe? Clearly, part of the reason is likely to be that the greater proportion of Americans going into business makes running a business in the United States more competitive. Indeed data show that the rates of firm entry and exit is higher in the United States than in most European countries (Verhoeven and Becht 1999; Stel and Diephuis 2004) and the net entry or survival rate, defined as the difference between the firm entry and firm exit rate, is much lower (Stel and Diephuis 2004).

Entrepreneurship in Europe and the United States

35

Table 2.2 Percentage of population aged 18 to 64 engaged in business activity Nascent entrepreneurial activity

New business owners

Established business owners

United States

8.80

5.20

4.70

France

4.70

.70

2.30

Norway

4.40

5.20

7.30

Germany

3.10

2.70

4.20

Finland

3.10

1.90

8.60

Austria

3.00

2.40

3.80

Italy

2.90

2.30

6.40

Belgium

2.90

1.20

5.60

Switzerland

2.60

3.70

9.70

Netherlands

2.50

1.90

5.70

Denmark

2.40

2.40

4.40

Spain

2.40

3.40

7.70

Sweden

1.70

2.50

6.30

Country

Source: from Minniti et al. (2005). Notes: Nascent entrepreneurs are defined as individuals who have taken some positive step in the past year toward creating a new business that has not yet paid and wages or salaries for more than three months. A new business owner is defined as an ownermanager of a firm that has paid wages or salaries for more than 3, but less than 42, months. Established business owners are owners of businesses that have paid wages or salaries for more than 42 months.

But greater competitiveness of business environment in the United States is not likely to be the whole explanation of why the proportion of business owners in the United States is not larger than in Europe. Indeed, while the chances of survival on the margin should be lower in a more competitive environment, it would be very strange if, all other things being equal, the absolute number of businesses per unit of the labor force were to remain constant (as if all economic environments could support only a fixed number of businesses). The other part of the explanation is likely to be that Americans not only go into business for themselves more often than the Europeans but also that they are likely to go into a more risky business than their European counterparts. In other words, while Europeans may open shops or service outlets, the Americans may more often venture into more high-tech start-ups or other new businesses offering greater potential rewards, but also a greater chance of failure.

36

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Table 2.3 National business ownership rates Business owners as a percentage of workforce (excluding agriculture) Country

2000

2002

Italy

18.5

18.3

Portugal

13.5

13.7

Spain

12.6

12.9

Belgium

11.7

11.3 10.8

Netherlands

10.9

United States

9.8

9.5

Germany

8.7

8.6

Switzerland

8.7

7.6

Austria

8.3

8.3

Sweden

8.3

8.1

France

8.3

8.1

Finland

8.1

7.9

Norway

6.4

6.5

Denmark

6.1

6.7

Source EIM: Comparative Entrepreneurship Data for International Analysis (Compendia 2000, 2002). Note: Included are owners of incorporated and unincorporated enterprises; excluded are unpaid family workers and salaried workers operating a business as a secondary work activity.

We know, for example, that the turnover rates of American and European businesses become comparable once industry and sectoral composition are controlled for—which seems to indicate that Americans start more businesses in more volatile areas in which the failure rates are greater. We also know that the average size of an American start-up is smaller at the time of entry, but that surviving firms in the United States expand much faster and attain higher average size than their European counterparts (OECD 2003; Bartelsman 2003). This again suggests that the volatility (and hence riskiness) of American start-ups is higher than in Europe. 2.2.2 The Impact of Labor Markets Why are Americans more likely than Europeans to start a business, and why are they likely to choose to start a riskier business (and one that presumably brings higher returns)? In what follows we choose to ignore some possible explanations. In particular, we will not focus on

Entrepreneurship in Europe and the United States

37

explanations related to “cultural” differences between the two environments. For example, being a “self-made man” may confer high social status on a person in the United States (thus effectively raising the returns on entrepreneurial projects), while in a more class-conscious European societies, making (but not inheriting) money may be considered “vulgar” by some influential opinion makers (thus effectively lowering entrepreneurial returns).4 Moreover these attitudes may very well become internalized by entrepreneurs, translating into a different utility value of their endeavors. While the influence of culture may thus be considerable, we nevertheless prefer to avoid explanations that stop at cultural differences that might themselves call for further explanations. Instead, we prefer to treat cultural phenomena as related to other social and economic conditions—even if the embodiment of those conditions in cultural values may often extend their impact long beyond their own persistence. The feature of the socioeconomic environment we focus on first is the impact of labor markets on entrepreneurship. It is often observed that labor markets in Europe are less flexible than in the United States: the protection of job security is significantly stronger, minimum wages are higher, and taxation of wages is higher as well. Labor unions in Europe are given a significantly larger role in running many businesses, and nationwide wage negotiations in some countries leave less room for individual businesses to structure their employment relations. It is usually believed that these factors combine to make labor costs in Europe high and lessen the competitiveness of European economies. This may very well be true, but we want to look at the labor market from a particular point of view of its impact on the incentives for entrepreneurship. Even the standard argument that labor market rigidities pose obstacles for firm growth is likely to apply more strongly to entrepreneurial ventures than to established companies. Start-up companies face tighter capital constraints, and higher labor costs may affect them more strongly than other firms. Because entrepreneurial firms require a risk premium to compensate the owners for the increased risks they undertake, these firms are more sensitive to higher taxes, which lower the upside returns. Most important, entrepreneurial firms grow faster than others, and their rate of growth is harder to predict. They are therefore particularly hampered by the inability to shed workers when risky ventures do not pan out as expected or when firms cannot structure relations with their employees, including their compensation, in ways they consider appropriate.

38

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

But the most important way in which labor market conditions affect incentives for entrepreneurship is in their impact on the decision to become an entrepreneur in the first place. Whether people start a business is most often a decision between the relative attractiveness of employment and striking on one’s own. A related choice faces the people whom an entrepreneur may want to employ in a new or otherwise risky business: to the extent that such a business is more likely to fail, how does a bright and capable employee weigh the decision to take on a potentially more lucrative and interesting job in an entrepreneurial business, as opposed to staying on with the job that the person now holds?5 How does this choice compare across the Atlantic divide? All other things being equal, the decision between entrepreneurship and employment involves weighing the risks and rewards involved in both pursuits. It might be generally said that the path of employment usually offers smaller risks and lower returns, while starting a business is more risky but potentially more rewarding. But at a closer look things are more complicated. European countries are generally characterized by much stricter employment protection than the United States. Figure 2.2 plots the relative restrictiveness of labor legislation in a number of OECD countries and shows that the United States has by far the least restrictive employment legislation for workers with both permanent and temporary contracts. With respect to hiring and firing regulations in particular, the United States is widely considered to be among the nations with the least rigid labor laws worldwide; similarly American employers encounter the least restrictive regulations concerning working hours and job tenure (World Bank 2004). In this sense Europeans have more “job security” than Americans do, and thus, when they opt out of a job and decide to go on their own, all other things being equal, they forgo more than the Americans do, who must consider the possibility that they may be fired anyway, even if they don’t take the risk of going into business for themselves. As a result the relative risk-adjusted returns to entrepreneurship are lower in Europe than in the United States. So much is well understood. But, again all other things being equal, most societies, particularly near the development frontier and in times of rapid change (when predictions concerning future business opportunities are difficult to make), face a trade-off between job security, in the sense of legal protection of incumbent job tenure, and higher levels of unemployment due to the resulting lesser flexibility of the labor

Regular employment

7

Temporary employment

6 5 4 3 2 1

rla he

et

s G er m an Sw y ed en N or wa y Fr an ce Sp ai Po n r tu ga l

m

nd

iu

nd

lg

ly Ita

la Fi n

Be N

Sw

St at es itz er la nd D en m ar k H un ga ry Po la nd Au st ria

0

ni U

39

8

te d

Overall strictness of regulatory framework – 2003

Entrepreneurship in Europe and the United States

Figure 2.2 Restrictiveness of employment protection legislation. The composite OECD indicator is intended to reflect the cost implications of labor protection measures affecting both regular and temporary employment (i.e. “strict” or “restrictive” should be understood as “costly”). The summary values take into account, inter alia, overall difficulty of dismissal, procedural inconveniences and requirements implicated when employers elect to initiate the dismissal process, and notice and severance pay provisions. (source: OECD 2004)

market (which requires greater commitment on the part of the hiring employers and makes them more reluctant to hire in the first place). In other words, while Americans have less incumbent job security, by the same token they also have more “employment security” in the sense of being able to leave and re-enter the job market more easily than their European counterparts. This in turn makes entrepreneurship again less risky for the Americans: not only are they not giving up valuable job security when they start a business; they also have more of a cushion if the business does not work out.6 The effects of what we termed job and employment security are, of course, not independent of each other because employment security (the ability to re-enter) lessens the value of job security and, as we noted already, job security lessens the ability to re-enter. Thus what makes the difference in terms of the relative incentives of Europeans and Americans to engage in entrepreneurial activities is the joint effect of job and employment security. But given the strong sense among the Europeans that job security is extremely important to them7 and the

40

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

much more persistent difficulty of re-entry in Europe, we believe that the combined effect of these two factors on the diminished incentives of the Europeans to enter entrepreneurial occupations (and to choose riskier entrepreneurial projects once they are in business for themselves) is likely to be significant. 2.2.3 Access to Capital and the Effects of Accountability Entrepreneurs often operate under significant capital constraints. Most finance the start-up or expansion of their business by drawing on their personal savings or borrowing from family and friends. But the ability to reach other sources of funding may be of great importance, and will have an impact on the likelihood that a business is started or, once started, turns out to be successful. Even when capital is available, the form in which it is accessible may make a significant difference as well to a new or rapidly growing business. On one end of the spectrum, an entrepreneur may turn to debt finance for the venture, perhaps by opening a line of credit with a commercial bank (Landier 2003). Creditors, however, do not share in the upside of the business but are exposed to the downside; lending to new, high-risk businesses is therefore generally inimical to their business philosophy.8 New ventures do not have proven track records, cannot promise steady returns, and rarely possess the requisite tangible assets to post as collateral for their debt obligations (OECD 2004). Moreover, since start-ups generally require significant amounts cash in the early stages of growth, the correspondingly large debt obligations are likely to be inappropriate from a cash-flow management perspective. The alternative, equity-based financing, is fraught with a different set of problems. Successful investors in start-up and new venture environments are necessarily “active.” Innovative, high-risk start-ups often operate in new markets where publicly available information is scarce. Entrepreneurs often have nonfinancial, reputational incentives, such as the desire to be the first to market or to publish, which may compromise long-term profitability. Financiers must be willing to monitor the new venture, incentivize successful commercialization (Jensen 1993; Sahlman 1990), and be ready to intervene by exercising their broad control rights (Landier 2003; Gompers 1995; Lerner 1995a, b). But investor interference can also harm the business, impose wrong attitudes toward risk, and generally prevent entrepreneurs from being able to act on their own business sense. This last point requires some elaboration. “Money does not lie in the street” and people do not make fortunes by acting according to well-

Entrepreneurship in Europe and the United States

41

recognized rules. The truth of this proposition goes beyond the trite observation that an entrepreneur sees opportunities that others miss. For even when told of an entrepreneurial idea, others tend to reject it, not because it is unproved but because it most often cannot be proved at all. Entrepreneurship is not a science: it does not proceed in accordance with well-understood rules or knowledge that can be objectively demonstrated (Kirzner 1979). Under conditions of change and uncertainty an objective assignment of probabilities to potential outcomes of a project is well-nigh impossible, and an idiosyncratic, subjective “hunch” that turns out to be accurate after the fact is a paradigmatic example of entrepreneurship. But the very feature that allows an entrepreneur to succeed by breaking the mold of routine, rule-governed behavior, makes it very difficult for an entrepreneur to convey inherently subjective “hunches” or “intuitions” to others whose views are within the prevailing consensus. And if an entrepreneur cannot explain ideas in intersubjective terms, it is also extremely difficult for him to account for his business decisions (Frydman et al. 2006). Even if a decision is right, the decision maker might be unable to clear it in advance with financial monitors or to explain it adequately after the fact if things happen to go wrong (Hayek 1948). Entrepreneurial decisions thus involve a large measure of ineffable skills and an often idiosyncratic evaluation of the situation that makes bureaucratic oversight, which relies heavily on rational, rule-governed justification of most actions, unsuitable. Consequently, despite the nurturing and monitoring that new entrepreneurial businesses may need, it is crucial that their access to capital not be conditioned on a control structure of bureaucratic accountability. Indeed, if the control structure of a firm requires that the decision maker justify all choices made, the decision maker is likely to forgo those projects that cannot be explained without difficulty to others, and this fact may severely restrict the opportunity set out of which feasible projects will be chosen. Independently of the particular form of financing used, there is therefore an institutional component that needs to be taken into account when considering the governance effects of the financing of entrepreneurial endeavors. New businesses are both too small and too hard to monitor to be able to access financial markets on their own; they usually need intermediary institutions, such as banks or venture capital funds. But there may be a big difference in how appropriate these institutions may be for picking promising businesses, being able to offer them proper services and advice, and monitoring their performance. Banks, for example, even when they make equity investments, are normally

42

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

institutionally inept at coping with the corporate governance issues facing start-up businesses, and as a general matter, mostly companies with strong reputations or long-standing relationships with the banking industry can hope to obtain bank financing (Jeng and Well 2000; Black and Gilson 1998; Ueda 2004). The specialized intermediary institution developed in response to the financing needs of an entrepreneurial start-up is the venture capital firm, which is much more than a simple conduit for investment funds.9 The venture capitalist’s incentive structure, business acumen, and network of suppliers, customers, and contacts forces entrepreneurs to focus on technical development (Black and Gilson 1998). Venture capital firms play a particularly crucial role in knowledge-based industries by selecting the most promising projects, financing and commercializing technological innovation, and lending their much-needed managerial and technical expertise. (Holmes and Schmitz 1990; Hellmann 2000). They also confer reputational advantages on a new business. But the pivotal role of venture capital with respect to the development of entrepreneurial businesses requires that the venture capital firms be structured in a certain way, most important, that they be run by entrepreneurs and not bureaucrats. This means that the managers of venture capital funds must be principals in their firms and that outside investors commit their funds for a specified period of time, without conditioning their participation on burdensome conditions or otherwise restricting the managers’ ability to make entrepreneurial decisions. When, however, venture capital firms are offshoots of traditional conservative financial institutions, such as banks, these institutions tend to staff them with their own bureaucratic personnel, and the latter are likely to go for safer and less speculative investments, and pass on the chance to participate in the more genuinely entrepreneurial projects. How do Europe and America compare on this score? The European risk capital market is not only smaller and more fragmented than that of the United States, it also focuses largely on buyouts as opposed to early-stage and technology investment10 (European Commission 2000; Black and Gilson 1998; Jeng and Wells 2000). Figures 2.3 and 2.4 reveal that seed and early-stage venture capital investment as a percentage of nominal GDP in Europe is simply anemic, as compared to the United States. Most strikingly, new venture financing in the largest economies on the Continent, particularly France and Germany, is an order of magnitude less than in the United States. Efforts in the European Union

43

0.25 2001

2002

2003

2004

2005

0.2

0.15

0.1

0.05

e

n

op

ai

ly

en

ta

lE

ur

Sp

Ita

ar Sw k e Sw de n itz er la nd Au st ria Fi nl an d N or w ay Ire la N nd et he rla nd s

y

m

an D

en

ce G

er

m

an Fr

C

on

U

tin

ni

U

te

ni

d

te

d

Ki

ng

St

at

do

m

0 es

Seed/start-up funds (mil$) / nominal GDP (bil$)

Entrepreneurship in Europe and the United States

1.2 2001

2002

2003

2004

2005

1

0.8

0.6

0.4

0.2

ro pe

n

Eu

Sp ai

ly Ita

ed en er la nd Au st ria Fi nl an d N or w ay Ire l an N et d he rla nd s

al

itz

en t

Sw

ar k

Sw

m

an y D

en

G

er m

an ce

ng Ki

d te

tin on C

U

ni

Fr

at St te d ni U

do m

0 es

Early-stage funds (mil$) / nominal GDP (bil$)

Figure 2.3 Normalized seed/start-up venture capital investments (source: Thomson VentureXpert; compiled by the authors)

Figure 2.4 Normalized early-stage venture capital funding investments (source: Thomson VentureXpert; compiled by the authors)

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

80% 2000

70%

2001

2002

2003

2004

2005

60% 50% 40% 30% 20% 10%

e

n

op

en

ta

lE

ur

ai

ly

Sp

Ita

y

m

an

en

m er

G

D

ce

m

an

ar Sw k e Sw de n itz er la nd Au st ria Fi nl an d N or w ay Ire l an N et d he rla nd s

d te

tin on C

U

ni

Fr

Ki

ng

St d te ni U

do

es

0% at

Bank and financial investment of total investment

44

Figure 2.5 Venture capital investment by banks and financial institutions (source: Thomson VentureXpert; compiled by the authors)

to increase entrepreneurs’ access to capital—tax and bankruptcy reform, and increased financing models—have been largely thwarted by the economic downturn. On the whole, European economies face continuing fiscal and structural challenges to the promotion of venture capital investments and even further shift away from early-stage and seed investments in favor of late-stage or buyout opportunities (European Commission 2004; Bottazi and Rin 2002). The pool of venture capital investors also varies significantly across the Atlantic: In the United States a sizable portion of venture capital is supplied by pension funds, which are usually patient and passive investors. In Europe, by contrast, banks and financial institutions account for more than 30 percent of new venture funds raised (Jeng and Wells 2000; Bottazi and Rin 2002). Figure 2.5 suggests that despite an overall global withdrawal since the “irrational exuberance” of the late 1990s, banking and financial institutions remain relatively entrenched in the European venture capital market: bank-financed venture capital investments in continental Europe are consistently greater, by 58.2 percent in 2000 and 52.4 percent in 2005, than in the United States. Another difference is the relative prevalence in Europe of “captive funds” managed by venture capital firms deriving more

Entrepreneurship in Europe and the United States

45

than 80 percent of their financing from one source, often a financial institution (Hellmann, Lindsey, and Puri 1999). Venture capital firms in the United States are generally organized as limited partnerships, while in France and Germany they gravitate toward other corporate governance structures, many being subsidiaries of corporations or financial institutions (Lerner 1995a, b). As a result of this and other factors, European venture capitalists eschew managerial or supervisory roles over portfolio companies, retain fewer control rights, and replace entrepreneurs less often, as compared to their American counterparts (Schwienbacher 2002; Lerner and Schoar 2003). The sectoral distribution of European private equity investments tends toward manufacturing, agriculture, and finance; by contrast, as suggested by table 2.4, the vast majority of portfolio companies in the United States are in the biomedical, computer, and telecom sectors, while the manufacturing sector garners only a small percentage of total US venture capital investment. The dominance of bank investment and the lack of supervisory control over new ventures in Europe gives rise to a venture capital market that is inherently less effective as a driver of entrepreneurial growth. Banks and financial-institution-controlled funds tend to direct their seed or early-stage investments to value prospects in low volatility industries. Captive funds are similarly among the least aggressive in their investment strategies (Hellmann, Lindsey, and Puri 1999). Indeed passive financing precludes successful early-stage investments in high-tech, high-risk ventures, which naturally require a hands-on approach, technical expertise, and constant monitoring on the part of the investor. 2.2.4 Capital Markets Financial markets are an important determinant of entrepreneurship not only at the start-up phase but also at the time of exit. The liquidity of a successful business may significantly raise its value to begin with, and thus raise expected returns at the time a potential entrepreneur contemplates going into business. The exact mode of exit can also have significant incentive effects throughout the course of the creation and growth of an entrepreneurial business because different forms of exit allow for different governance arrangements before and after the time when some important principals decide to move on. Exit is important for both types of principals of a successful entrepreneurial business: the entrepreneurs who may want to cash out on their investments, and the venture capitalists whose roles are likely to

2

4

1

2

5

4

6

9

12

14

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

17

17

16

16

15

18

17

21

21

12

United States

Source: Bottazzi and Rin (2002).

Europe

Year

Telecom

13

11

9

7

5

7

4

6

4

6

Europe

Computer

58

56

36

30

27

21

18

30

12

19

United States

43

50

46

51

52

56

61

59

58

56

Europe

7

6

8

8

9

12

9

8

8

13

United States

Manufacturing

Table 2.4 Sectoral distribution of venture capital investment (% of total)

10

7

7

7

6

8

5

6

5

6

Europe

Biomed

6

7

17

27

20

22

23

21

22

25

United States

4

2

3

5

4

4

4

4

3

3

Europe

Electronics

8

6

11

8

7

12

10

7

10

17

United States

16

18

26

24

29

20

24

24

26

27

Europe

Other

4

8

12

11

22

15

23

13

27

14

United States

46 Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Entrepreneurship in Europe and the United States

47

diminish over time and who may come to have better opportunities elsewhere. Noncapital inputs, such as reputation, experience, and external monitoring, are extremely valuable to businesses in the earlystages of their growth. As a business model succeeds, however, the relative value of the nonfinancial contributions of the venture capital firm declines. Exercising an exit option enables a venture capital firm to recycle its nonfinancial contributions to the success of the start-up and reinvest in another early-stage portfolio company. Venture capital exit is also important to the investor community at large, as capital providers learn about the relative success or failure of different fund managers, and have an opportunity to reallocate their capital from less successful managers either to more successful ones or to other investment vehicles (Black and Gilson 1998; Berger and Udell 1998). Exit may be effectuated through sale, initial public offering (IPO), or repurchase of a principal’s stake by the company. But all exit mechanisms are not created equal. In particular, exit by IPO is preferred by the original entrepreneur because, as the venture capitalist is replaced by dispersed shareholders, the entrepreneur regains control of the business—something that could not be effected by a sale since an acquirer retains control even if the entrepreneur is kept in a managerial capacity. IPO is also important to the venture capitalist because it increases liquidity and allows the shareholders to reap additional returns at a time when the relative valuation of publicly traded securities is high (Lerner 1994). Indeed, for these and other reasons, IPO is in fact much more profitable than other forms of exit: studies show that a US firm which eventually goes public yields a 195 percent average return over a 4.2-year average holding period; the same investment in an acquired firm provides an average return of only 40 percent over a 3.7-year average holding period (Venture Economics 1988; Gompers and Lerner 1997; Bygrave and Timmons 1992; Bienz 2004). IPO exit, the linchpin of a successful entrepreneurial environment, is only viable in the presence of a large, vibrant public equity market that has numerous providers of risk capital and permits new firms to issue shares. And here lies yet another difference between Europe and the United States that may be one of the reasons for greater entrepreneurship in America. European stock markets, which are generally less capitalized, have traditionally been also unwelcoming of young companies without established track records, instead predicating listing on sustained positive earnings growth (Pagano, Panetta, and Zingales 1998; Rydqvist and Högholm 1995). The effect of Euro.nm, a consor-

48

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Table 2.5 Overall IPO activity 2001

2002

2003

2004

Country

Total IPOs

Capital raised ($mil)

Total IPOs

Capital raised ($mil)

Total IPOs

Capital raised ($mil)

Total IPOs

Capital raised ($mil)

United States Germany

121 23

39,596 1,384

157 5

40,374 133

123 —

41,901 —

275 6

66,710 1,179

—

—

Sweden

6

123

4

213

France

50

4,637

27

1,599

Italy

17

1,652

7

903

Spain

2

1,526

1

601

Netherlands

2

185

—

—

Norway

5

738

—

—

Austria

5

53

1

11

1

0

Finland Denmark

3

87

—

—

7 3

4

257

470

25

3,576

487

9

2,041

—

—

3

1,791

—

—

1

15

7

167

1

20

4

139

1

16

1 —

35 —

Source: Thomson Financial. Note: IPO is credited to the domicile nation of the financed company; capital raised refers to the proceeds and overallotment amounts sold in the market specified.

tium of European stock markets for innovative firms in high-growth industries similar to NASDAQ, is mixed at best. Euro.nm, which opened in 1997, ceased to exist in December 2000. Even though its member exchanges continue to operate independently, early indications are that both overall and venture-backed IPO activity in Europe is still anemic, and continues to lag significantly behind that of the United States (tables 2.5 and 2.6; Botazzi and Rin 2002; European Commission 2004). But until the market for risk capital is adequately developed, early-stage entrepreneurial projects in Europe are likely to be constrained by the inferior exit opportunities. 2.3

Risk Taking and Entrepreneurship in a Corporate Environment

Entrepreneurship is usually associated with individuals and small businesses: the “mad inventor” in his garage, a Bill Gates getting on the train of the computer revolution, or a couple of students inventing the Google search engine. But entrepreneurship is, of course, a matter of degree and there is a continuum of entrepreneurial business decisions made by firms of all sizes. Although large firms often focus on

Entrepreneurship in Europe and the United States

49

Table 2.6 Venture-backed IPO activity 2001

2002

2003

Country

Total IPOs

Capital raised ($mil)

Total IPOs

Capital raised ($mil)

Total IPOs

United States

2004 Capital raised ($mil)

Total IPOs

Capital raised ($mil)

1,412

22

1,785

20

1,640

22

67

4,983

France

5

116

4

20

1

—

2

104

Germany

6

101

1

3

—

—

1

51

Italy

1

22

—

—

—

—

1

102

Spain

—

—

—

—

—

—

United Kingdom Denmark

8

— 162

4

53

—

1

Finland

—

Norway

2

Sweden

2

21

108

24

—

—

—

1

3

—

—

1

4

2

25

1

42

1

3

—

—

—

7

18 —

— 514 — 64

5 —

43 —

Source: VentureSource. Note: IPO is credited to the domicile nation of the financed company.

more predictable cost efficiencies and routinize many of their business operations,11 it is also of great importance that they have an institutional ability and incentives to make risky decisions under conditions of often radical uncertainty. This is why differences in corporate governance structure and management remuneration systems, which may have far-reaching consequences with respect to firms’ ability to make entrepreneurial decisions, may in part account for the difference in the performance of European and American economies. In what follows we will consider three basic types of reasons why firms may be more or less inclined to make entrepreneurial decisions: (1) risk attitudes of the owners, (2) distortions due to agency problems and managerial risk preferences, and (3) distortions due to the accountability structure in the corporate governance model. 2.3.1 The Impact of Ownership Structure To the extent corporate governance works to align the incentives of the management with the interests of the owners and other stakeholders, the risk profiles of such owners and stakeholders will have significant influence on firm behavior.12 If a controlling owner has a large portion of its capital and/or human assets invested in the enterprise, the owner

50

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

is likely to be risk averse and unwilling to bear the risks of entrepreneurial behavior. The problem is particularly acute in firms that are worker-owned or required to maintain a significant board-level representation of labor interests: the volatility inherent in entrepreneurial strategies often entails rapid expansions and contractions in the labor force, a consequence that is anathema to the current employees interested in stable, long-term employment commitments. State ownership also has serious distorting incentive effects, as the state bureaucrats responsible for monitoring state holdings do not personally participate in the upside of corporate decisions but can suffer serious negative consequences if the firm has to fire workers or otherwise disappoint the bureaucrats’ constituencies. Even among private owners who are not otherwise stakeholders in the firm, risk attitudes are likely to differ depending on the level of ownership concentration. In diffusely owned firms, shareholders are likely to be diversified with respect to their investment, and they can be expected to maximize the value of the firm by ratifying entrepreneurial decisions where the risks seem justified by expected payoffs (Frydman et al. 1999). Conversely, concentrated ownership is likely to involve firm-specific, undiversified investment on the part of owners. To be sure, some institutional investors are capable of spreading the risk of even significant holdings across their investment portfolios (Baysinger 1991), but the diversification of entrepreneurial risk increases overall returns only if the individual stakes are not large relative to the size of the portfolio. Large individual or family owners, in turn, nearly always suffer from a lack of diversification.13 Finally, different institutional owners have different investment objectives. We have mentioned already that the state as an owner is likely to have incentive-distorting effects. But other classes of owners, such as banks and insurance companies, might also discourage managers from projects with highly variable returns or distant payoffs. Other institutions, such as pension funds, are likely to have a longer term horizon and promote entrepreneurship and innovation (Zahra 1996; Kochhar and David 1996). Again, how do Europe and the United States compare on this dimension? Table 2.7 (based on several studies of equity ownership of nonfinancial corporations) and table 2.8 indicate that the shareholder base of European companies is extremely concentrated relative to the United States. While almost one-quarter of US firms report the largest owners’ share to be less than 5 percent, and only 9 percent report

Entrepreneurship in Europe and the United States

51

Table 2.7 Ownership concentration of listed nonfinancial firms (% of total) Largest owners’ share

France

Germany

Italy

Spain

Sweden

>50

55

66

89

49

42

9

30–50

—

—

—

—

31

—

25–30

—

—

9

—

—

—

20–25

—

23

—

49

12

29

15–20

42

—

—

—

—

—

10–15

—

—

—

—

11

10

5–10

—

12

2

—

—

29

2

—

—

2

4

23

<5

United States

Sources: Berglof (1990); Berglof et al. (1997).

Table 2.8 Ownership of ten largest nonfinancial firms

Country

Mean ownership by three largest shareholders

Median ownership by three largest shareholders

Average market capitalization of firms ($bil)

United States

0.20

0.12

71.65

France

0.34

0.24

8.90

Italy

0.58

0.60

3.14

Netherlands

0.39

0.31

6.40

Spain

0.51

0.50

1.27

Austria

0.58

0.51

0.32

Germany

0.48

0.50

8.54

Switzerland

0.41

0.48

9.58

Denmark

0.45

0.40

1.27

Finland

0.37

0.34

1.98

Norway

0.36

0.31

1.11

Sweden

0.28

0.28

6.22

Source: LaPorta et al. (1998).

a majority owner, well over half of all European companies are majority-owned by a single entity. The extremely high levels of concentration persist even in the largest European companies: the mean and median ownership by the three largest shareholders in the ten largest nonfinancial firms in Europe are two or three times larger than in the United States.

52

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Table 2.9 State ownership in Europe and the United States Largest firms

Medium-sized firms

20% stake

10% stake

20% stake

10% stake

Austria

0.70

0.70

0.83

0.83

Norway

0.35

0.40

0.20

0.20

Spain

0.30

0.45

0.20

0.30

Denmark

0.15

0.20

0.20

0.20

Finland

0.35

0.35

0.20

0.20

France

0.15

0.20

0.20

0.20

Germany

0.25

0.30

0.20

0.20

Italy

0.40

0.50

0

0.10

Netherlands

0.05

0.05

0.10

0.10

Sweden

0.10

0.10

0.20

0.20

Switzerland

0

0

0

0

Cont Europe

0.25

—

0.21

—

United States

0

0

0

0

Source: LaPorta et al. (1999). Note: The “large firm” set consists of the twenty largest firms in the country by market capitalization of equity at the end of 1995. The “medium-sized firm” sample consists of ten firms with market capitalization of $500 million or greater at the end of 1995.

The incidence of state ownership, essentially nonexistent in the United States, is also quite high in Europe: as shown in table 2.9, a state entity is the controlling shareholder (defined as above 20 percent ownership) in 20 to 25 percent of all European firms. Finally, as shown in table 2.10, financial intermediaries (very often banks or insurance companies) own significant stakes in many European companies, which suggests a conservative, risk-averse attitude among important shareholders. Moreover, in the case of banks at least, the institutional shareholders in European companies are likely to have debt, in addition to equity, interests in the company, which further increases their conservative bias. In the US companies, by contrast, the interests of financial intermediaries (often pension, hedge, and mutual funds, rather than banks) are small (never rising to 10 percent) and overwhelmingly passive. In sum, the owners of European firms are likely to be much less diversified than their American counterparts. Moreover a significant portion of these owners are risk-averse financial institutions or the state, and labor unions in some countries have a legally imposed voice in running corporations. To the extent that corporate governance mech-

Entrepreneurship in Europe and the United States

53

Table 2.10 Financial institution ownership in Europe and the United States Largest firms

Medium-sized firms

20% stake

10% stake

20% stake

10% stake

Austria

0

0

0

0

Norway

0.05

0.10

0.10

0.10

Spain

0.10

0.15

0.40

0.40

Denmark

0

0.05

0

0

Finland

0.05

0.25

0.10

0.20

France

0.05

0.20

0.20

0.20

Germany

0.15

0.25

0.20

0.30

Italy

0.05

0

0

0

Netherlands

0

0

0

0

Sweden

0.15

0.30

0

0.10

Switzerland

0.05

0.05

0

0

United States

0

0

0

0

Source: LaPorta et al. (1999). Note: The “large firm” set consists of the twenty largest firms in the country by market capitalization of equity at the end of 1995. The “medium-sized firm” sample consists of ten firms with market capitalization of $500 million or greater at the end of 1995.

anisms align the actions of European firms with the interests of their owners, the ownership structure of European companies, as compared to their American counterparts, is likely to dampen the managers’ will and ability to make entrepreneurial decisions. 2.3.2 Accountability and Control We have explained earlier, when discussing the relationship between start-ups and their financial backers, the ability to make entrepreneurial decisions depends on a control structure that leaves the decision maker free to make idiosyncratic and unconventional choices characteristic of entrepreneurial behavior. What was true about new entrepreneurs and venture capitalists is also true about decision makers in large- and medium-sized corporations: if managers of such firms are to maximize the set of projects available to the firm and pick those with the highest risk-adjusted returns, the structure of accountability in the firm must be set up in such a way that the decision maker can be guided by his ineffable skills and “hunches” and does not have to explain or justify his decisions. This does not mean, of course, that the decision maker must not be accountable in any sense, but only that the structure of accountability must involve an alignment of incentives,

54

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

rather than require to obtain ex ante authorization for actions or to justify them ex post if things do not work out. It may be stated as a general proposition that European corporate governance arrangements rely less than their American counterparts on the alignment of managerial incentives with the interests of the shareholders and more on managerial accountability and collective corporate leadership. The American governance arrangements, by contrast, typically install the chief executive (CEO) as a semidictator at the head of an extremely hierarchical organization and try to curb the CEO’s inherent risk aversion (due to a heavily undiversified commitment of the CEO’s human capital to the firm) and CEO opportunism by setting up compensation schemes that offer princely rewards for success and relatively mild punishment for failure. First, as we already observed, most publicly owned firms in America have very diffused ownership that leads to shareholder passivity and confers tremendous powers on corporate insiders. As a by-product of what is often perceived as an agency problem, corporate insiders (defined as the management and directors) have almost unfettered discretion in their choice of projects, and exercise a degree of flexibility functionally approaching that of the classic garage inventor. By contrast, European ownership is usually very concentrated, and most firms have one or very few controlling shareholders—the fact that makes other, more formal corporate governance arrangements of less importance and shifts the actual power to make the more important decisions to the owners, unless the owners are disposed to leave things to the management. We have argued that given the undiversified nature of the owners’ investment and their institutional characteristics, this control structure is likely to result in a less willing approval of entrepreneurial decisions, even if the management might be disposed to pursue riskier projects (though, given the compensation structure of European companies, to be discussed presently, they are not in fact very likely to be entrepreneurial). But even with respect to projects that fit with the risk profile of the owners, the managers are not likely to be able to pursue the more entrepreneurial projects because they will be unable to explain them to their owners and convince them to approve them. In the absence of active controlling shareholders, whatever (nonlegal) restraints are imposed on the CEOs come from supervision by corporate board(s). And here again a comparison of the relationship between boards and management in Europe and the United States reveals that American managers exercise a much greater degree of flex-

Entrepreneurship in Europe and the United States

55

ibility and independence than their European counterparts. American corporate governance involves a single-tier board, which CEOs have historically dominated; managers frequently control the solicitation process, and it is commonplace for a current CEO to serve on the board of directors (often chairing it as well) (Dalton and Kesner 1987). The German model illustrates the opposite end of the spectrum. It consists of a two-tiered board system in which the supervisory board (Aufsichtsrat), appoints a management board (Vorstand) to a five-year term and reviews firm and management performance. Managers are neither permitted to serve on the supervisory board nor to exercise any influence over the proxy solicitation machinery. Aufsichtsrat’s members are often appointed because their institutions control large blocks of shares, and their loyalty is to them rather than to the CEOs, to whom they do not owe their financial or social positions (Roe 1993). Even more important, the Vorstand, which is responsible for day-to-day managerial decisions, is a collective body in which the CEO is more of a primus inter pares than a dictator in the mold of the American CEO. (Other western European corporate governance regimes lie somewhere between these two extremes, but there is an increasing structural convergence toward the German model, particularly in France and Italy; Hopt and Leyens 2004.) The foregoing is not to suggest that the monitoring of management by owners and boards is unimportant; indeed it would be foolhardy to ignore the principal-agent issues that may result in managerial selfdealing and other abuses. Rather, the intimation here is that the costs of strict accountability structures need to be considered relevant for the firm’s dynamism: the more entrepreneurial spirit of American businesses may be another side of a relatively relaxed supervisory structure. 2.3.3 Executive Compensation Compensation structure is the preferred method of aligning managerial incentives with the shareholder interests in the United States, and although the American CEO compensation system has recently come under severe criticism (Bebchuk 2001), it may also generally favor more entrepreneurial behavior than its European counterpart. Although shareholder value might benefit from the pursuit of highrisk, high-return strategies, managers, whose human capital investments are nondiversifiable and largely firm-specific, are rationally averse to such undertakings—the inherent managerial preference is for

56

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

corporate diversification and empire-building, and for short-term efficiency gains even at the expense of long-term returns. Both the structure and the amount of compensation can be used to reverse, at least in part, the inherent managerial tendency toward risk aversion. Moreover, if a compensation scheme properly aligns the incentives of the management with the interests of the shareholders, the management might be given broad discretion with respect to both strategic and tactical decisions, thus avoiding the pitfalls that other forms of accountability may pose for the firm’s ability to pursue entrepreneurial projects. Despite substantial heterogeneity across firms, industries, and countries, executive compensation packages are generally reducible to three basic components: base salary; variable, annual bonuses; and long-term incentives. Accounting profit is the most widely used performance measure to set the level of the variable bonus component. Although accounting measures are simple to understand and managers can easily predict the effect of their strategies on profitability, these measures are inherently backward-looking and subject to manipulation. Since managers and their subordinates compile the data on which compensation is based, they may shift earnings across periods or adjust accruals to meet or exceed performance standards (Healy 1985). More important, executives attentive to short-term earnings measures are likely to shy away from projects that sacrifice current profits but have higher expected values over longer time horizons, such as aggressive research and development investment (Dechow and Sloan 1991). To correct the potentially myopic focus of managers, companies offer long-term incentive plans in the form of either bonuses based on a rolling-average of three- or five-year accounting performance or stock-based incentives, such as restricted stock or stock option plans. Stock option grants are particularly effective in mitigating the effects of managerial risk aversion and in providing managers with powerful incentives to pursue more risky, entrepreneurial projects (Hirshleifer and Suh 1992). Not only are managers permitted to capture the upside potential of entrepreneurship through their ownership stake in the firm, but also the value of the options themselves increases with stockprice volatility, a function of the firm’s risk portfolio created by managerial investment strategies. Two important facts emerge from a comparison of executive compensation practices in Europe and the United States. First, as figure 2.6 demonstrates, American CEOs are paid almost twice as much as their European counterparts, and the significant difference persists after controlling for national discrepancies, such as the tax treatment of base pay

Entrepreneurship in Europe and the United States

57

2,500,000 Long-term incentives Perquisites Company contributions: voluntary Company contributions: compulsory Variable bonus Basic compensation

US dollars

2,000,000

1,500,000

1,000,000

500,000

es

e

St d te ni U

C

on

tin

en

ta

lE

er

ur

at

op

nd la

en Sw itz

n ai

Sw ed

rla he

Sp

s nd

ly Ita N

et

y

ce

an m G er

an Fr

Be

lg

iu

m

0

Figure 2.6 Total remuneration of chief executive officers

and perquisites, purchasing power, and public benefits (Abowd and Bognanno 1995), and firm characteristics, such as size, industry, and market environment (Murphy 2002). It is also worth noting that the pay disparity between the United States and Europe is limited to senior executives; there is no meaningful difference across the Atlantic in the compensation of lower level executives and production employees (Abowd and Bognanno 1995). Indeed, reflecting the much more powerful role of the CEO in America, the difference between the CEO pay and that of the second and third person in command is much more pronounced in American firms than in their more collegially managed European counterparts (Bebchuk 2001). Second, the structure of executive compensation varies significantly across the Atlantic. When the value of each component of compensation is expressed as a percentage of total remuneration (figure 2.7), stock-based incentives—stock options and stock ownership—constitute a much higher fraction of overall compensation in the United States (Abowd and Bognanno 1995; Kaplan 1997). Long-term incentives, mostly in the form of stock option grants, represent almost 50 percent of compensation in the United States, but no more than 20 percent in Europe, and the base salary amounts are relatively smaller in the United States than in Europe.

58

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% es

e

St d te ni U

C

on

tin

en

ta

lE

er

ur

at

op

nd la

en Sw itz

n ai

Sw ed

rla he et N

Sp

s nd

ly Ita

y

ce

an m G er

an Fr

Be

lg

iu

m

0%

Long-term incentives Perquisites Company contributions: voluntary Company contributions: compulsory Variable bonus Basic compensation

Figure 2.7 Components of total CEO remuneration. The data represent estimates of typical pay for national executives as of April 1, 2005, in locally headquartered companies with approximately US$500 million in worldwide annual sales. Basic Compensation consists of annual base salary (including regular payments such as 13th month salary and vacation allowance, if applicable) plus performance-unrelated bonus; Variable Bonus consists of performance-related payments; Compulsory Company Contributions consist of employer contributions/expense for social security, compulsory benefit programs and mandated termination indemnities; Voluntary Company Contributions are typical employer contributions/expenses for private retirement, life insurance, disability, medical and other voluntary employee benefit plans (including executive pensions); Perquisites consist of annual cash value of company cars, club memberships and other perquisites that are typically provided to executives; Long-Term Incentives consist of annual expected value of stock-related awards (e.g., stock options, stock grants) and other awards. (source: Towers Perrin WorldWide Total Remuneration Report 2005–06)

Entrepreneurship in Europe and the United States

59

The reasons for the dramatic differences in the amount and structure of CEO compensation are hotly contested in scholarly literature. Some observers explain it as resulting from self-dealing and rent extraction by insufficiently monitored American managers (Bebchuk 2001), while others trace it to the more favorable tax treatment of option compensation in the United States (Murphy 2002) or the differences in the magnitude or type of demand for chief executives . But even if, as the critics contend, the incentive effects of executive compensation in the United States may be far from optimal, the effects of the higher and more incentive-based compensation of the US CEOs are clearly relevant in assessing their incentives to make entrepreneurial decisions: high incentive-based compensation creates a greater upside for the managers and provides a risk premium for the pursuit of entrepreneurial projects. Concluding Remarks Although entrepreneurship is increasingly recognized as central to the dynamism and growth of modern economies, it is hardly mentioned in systematic comparisons of economic performance. In this chapter we explored the institutional factors that could have facilitated or hindered entrepreneurship in the context of European and US economies since World War II. We argued that various features of the European labor market, corporate governance arrangements, and patterns of project financing have provided an environment more conductive to growth based on employment security and importation of frontier technology. However, US institutions have downplayed employment security and provided more powerful incentives for entrepreneurship and innovation. The chapter suggests that anchoring comparative analysis in entrepreneurship may help explain the differences in economic performance on both sides of the Atlantic. Notes 1. See Phelps (2006). 2. For a proposal of an alternative approach to modeling market outcomes that places imperfect knowledge at the center of economic analysis, see Frydman and Goldberg (2007, 2011). 3. A number of other studies also found a greater levels of entrepreneurial interest among American business students than among their counterparts in Germany (Busenitz 2000; Weihe 1993), Spain (Busenitz 2000), France, Austria (Weihe 1993), Italy, Sweden, and Norway (Busenitz 2000).

60

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

4. Survey evidence and anecdotal experiences support the notion that cultures value self-employment and business ownership to differing degrees. The GEM entrepreneurship study, for example, found that national attitudes toward self-employment varied starkly across the Atlantic divide, and that respondents in the United States believed, in overwhelming numbers as compared to European countries, that starting a new business was a respected occupation and that they would approve if their son or daughter started a business (Levie et al. 1999). 5. We assume here that an overwhelming majority of people starting a businesses (as well as most valuable potential employees of such businesses) hold jobs as employees in other businesses and are not just entering the labor market. Starting a new business (or being a valuable employee) requires some knowledge and experience that, in most cases, can only be gained by prior employment. We therefore assume that most potential entrepreneurs are employed and that therefore their choice is between staying on employed or branching out on their own. To be sure, there may be some people who start a business straight out of school or because they can’t find any other employment, but our “hunch” is that, despite an occasional Bill Gates, these are not usually the most successful entrepreneurs. 6. To be sure, the Europeans have more of a safety net when they are unemployed. But the prospect of better support when unemployed may be of lesser importance than the ability to re-enter the job market, especially for people with entrepreneurial temperament who are likely to find “being on the dole” particularly irksome. 7. There is ample evidence for this proposition. See, for example, Clark (1998). According to the Eurobarometer survey, over 20 percent of European respondents preferred wage employment over entrepreneurship because the former was “not as risky”; only 5 percent in the US sample gave the same response. Over half of the sampled Europeans also agreed with the proposition that “One should not start a business if there is a risk it that it might fail,” as compared with only a third of their US counterparts. 8. A member of the Supervisory Board of a prominent Austrian bank told one of us that “the first criterion of the bank’s funding of a business is that owner must not want to get rich quickly”—perhaps a defining characteristic of a true entrepreneur! 9. The term “venture capital,” as used here, refers to seed, start-up, and expansion financing. It is important to distinguish this form of private equity investing from management and leveraged buyouts. Buyouts typically occur in the later, more mature stages of a firm’s growth cycle. Outside the United States, however, “venture capital” is frequently used to describe private equity investments generally, without regard to earlyversus late-stage investing. 10. In 1999 approximately 13 percent of gross venture capital financing in Europe was directed towards seed and start-up investments, as compared to more than 30 percent in the United States. The relative percentages of venture capital allocated to high-tech industries are even more striking: 26 percent in Europe versus 80 percent in the United States (European Commission 2000). 11. For a different role played by entrepreneurial decisions in the cost and revenue sides of business management, see Frydman et al. (1999, 2006). 12. For the impact of ownership on firm performance generally, see Frydman et al. (1999, 2000, 2006) and Frydman and Rapaczynski (1994). 13. Family and individual owners may also have other characteristics that contribute to a conservative attitude with respect to risk taking. Founders and family owners often

Entrepreneurship in Europe and the United States

61

have “dynastic” concerns. We have no hard evidence of this, but anecdotal evidence points to a greater likelihood that European owners are more likely to be concerned with keeping the business in the family, while American founder-owners are more interested in cashing out and not leaving their children in charge.

Bibliography Abowd, J. M., and M. L. Bognanno. 1995. International differences in executive and managerial compensation. In R. Freeman, and L. Katz, eds., Differences and Changes in Wage Structures, Chicago: University of Chicago Press, 67–103. Altomonte, C., and M. Nava. 2005. Economics and Policies in the Enlarged EU, Cheltenam: Edward Elgar. Baysinger, B. D., R. D. Kosnik, and T. A. Turk. 1991. Effects of board and ownership structure on corporate R&D strategy. Academy of Management Journal 34 (1): 205–14. Bebchuk, L. A., J. M. Fried, and D. I. Walker. 2002. Managerial power and rent extraction in the design of executive compensation. University of Chicago Law Review 69: 751–846. Bottazzi, L., and M. Da Rin. 2002. Venture capital in Europe and the financing of innovative companies. Economic Policy 17 (34): 229. Berglof, E. 1990. Corporate Control and Capital Structure: Essays on Property Rights and Financial Contracts. Stockholm: Stockholm School of Economics. Berglof, E., P. Rey, and A. A. Roell. 1997. Reforming corporate governance: Redirecting the European agenda. Economic Policy 12 (24): 91–123. Bartelsman, E., S. Scarpetta, and F. Schivardi. 2003. Comparative analysis of firm demographics and survival: Micro-level evidence for the OECD countries. Working paper 348. OECD Economics Department, Paris. Berger, A. N., and G. F. Udell. 1998. The economics of small business finance: The roles of private equity and debt markets in the financial growth cycle. Journal of Banking and Finance 22: 613–73. Bienz, C. 2004. A pecking order of venture capital exits: What determines the optimal exit channel for venture capital backed ventures? Unpublished working paper presented at CFS Workshop on Venture Capital and New Markets, March 1, 2004. Black, B. S., and R. J. Gilson. 1998. Venture capital and the structure of capital markets: Banks versus stock markets. Journal of Financial Economics 47 (3): 243–77. Blanchflower, D. D., and A. Oswald. 1998. Entrepreneurship and the youth labor market problem: A report for the OECD. OECD, Paris. Bygrave, W. D., and J. A. Timmons. 1992.Venture Capital at the Crossroads. Boston: Harvard Business School Press. Clark, A. E. 1998. What makes a good job? Evidence from OECD countries. Report prepared for Working Party 7 of the OECD Employment and Unemployment Statistics. OECD, Paris. Dalton, D. R., and I. F. Kesner. 1987. Composition and CEO duality in boards of directors: An international perspective. Journal of International Business Studies 18 (3): 33–42.

62

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Dechow, P. M., and R. G. Sloan. 1991. Executive incentives and the horizon problem. Journal of Accounting and Economics 14 (1): 51–89. European Commission. 2000. Report from the Commission to the Council and the European Parliament on the Implementation of the European Charter for Small Enterprises. Brussels: European Commission. European Commission. 2004. Report from the Commission to the Council and the European Parliament on the Implementation of the European Charter for Small Enterprises. Brussels: European Commission. Frydman, R., and M. D. Goldberg. 2007. Imperfect Knowledge Economics: Exchange Rates and Risk. Princeton: Princeton University Press. Frydman, R., and M. D. Goldberg. 2011. Ibeyond Mechanical Markets: Asset Price Swings, Risk, and the Role of the State. Princeton: Princeton University Press. Frydman, R., C. Gray, M. Hessel, and A. Rapaczynski. 1999. When does privatization work? The impact of private ownership on corporate performance in transition economies. Quarterly Journal of Economics 114: 1153–92. Frydman, R., C. Gray, M. Hessel, and A. Rapaczynski. 2000. The limits of discipline: Ownership and hard budget constraints in the transition economies. Economics of Transition 8 (3): 577–602. Frydman, R., M. Hessel, and A. Rapaczynski. 2006. Why ownership matters? Entrepreneurship and the restructuring of enterprises in central Europe. In M. Fox and M. Heller, eds., Corporate Governance Lessons from Transition Economies. Princeton: Princeton University Press, 194–230. Frydman, R., and A. Rapaczynski. 1994. Privatization in Eastern Europe: Is the State Withering Away? Budapest and Oxford: Central European University Press and Oxford University Press. Gompers, P. A. 1995. Optimal investment, monitoring, and the staging of venture capital. Journal of Finance 50 (5): 1461–89. Gompers. P. A., and J. Lerner. 1997. Risk and reward in private equity investments: The challenge of performance assessment. Journal of Private Equity 1 (2): 5–12. Hayek, F. A. 1948. Individualism and Economic Order. Chicago: University of Chicago Press. Paul M. and P. M. Healy. 1985. The effect of bonus schemes on accounting decisions. Journal of Accounting and Economics 7: 85–107. Hellmann, T., and M. Puri. 2000. The interaction between product market and financing strategy: The role of venture capital. Review of Financial Studies 13: 959–84. Hellmann, T., L. Lindsey, and M. Puri. 1999. Banks in the venture capital market: An empirical analysis. Mimeo. Graduate School of Business, Stanford University. Hellman, T. 2000.Venture capitalists: The coaches of Silicon Valley. In C.-M. Lee et al., eds., The Silicon Valley Edge: A Habitat for Innovation and Entrepreneurship. Stanford: Stanford University Press: 276–94. Hirshleifer, D., and Y. Suh. 1992. Risk, managerial effort, and project choice. Journal of Financial Intermediation 2 (3): 308–45.

Entrepreneurship in Europe and the United States

63

Holmes, T. J., and J. A. Schmitz. 1990. A theory of entrepreneurship and its application to the study of business transfers. Journal of Political Economy 98 (2): 265–94. Hopt, K. J., and P. C. Leyens. 2004. Board models in Europe: Recent developments in internal governance structures in Germany, the United Kingdom, France, and Italy. European Company and Financial Law Review 1 (2): 135–68. Jeng, L. A., and P. C. Wells. 2000. The determinants of venture capital funding: Evidence across countries. Journal of Corporate Finance 6: 241–89. Jensen, M. C. 1993. The modern industrial revolution, exit, and the failure of internal control systems. Journal of Finance 48 (3): 831–80. Kaplan, S. N. 1997. Top executive incentives in Germany, Japan, and the US: A comparison, in executive compensation and shareholder value. Journal of Applied Corporate Finance 9: 86–93. Kirzner, I. M. 1979. Perception, Opportunity, and Profit: Studies in the Theory of Entrepreneurship. Chicago: University of Chicago Press. Knight, F. 1921. Risk, Uncertainty, and Profit. Boston: Houghton Mifflin. Kochhar, R., and D. Parthiban. 1996. Institutional investors and firm innovation: A test of competing hypotheses. Strategic Management Journal 17: 73–84. Kollinger, P., Minniti, M., and C. Schade. 2005. Entrepreneurial overconfidence: Evidence from a C.A.R.T. approach. Discussion paper 465. Berlin: German Institute for Economic Research. La Porta, R., F. Lopez-De-Silanes, and A. Shleifer. 1999. Corporate ownership around the world. Journal of Finance 54 (2): 471–517. Rafael La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. W. Vishny. 1998. Law and finance. Journal of Political Economy 106 (6): 1113–55. Landier, A. 2003. Start-up financing: From banks to venture capital. Working paper. University of Chicago. Lerner, J., and A. Schoar. 2004. Transaction structures in the developing world: Evidence from private equity. Working paper 4468-04. Sloan School of Management, MIT. Lerner, J. 1995a. The European Association of Security Dealers: November 1994. Case study. Harvard Business School. Lerner, J. 1995b. Venture capitalists and the oversight of private firms. Journal of Finance 50 (1): 301–18. Lerner, J. 1994. Venture capitalists and the decision to go public. Journal of Financial Economics 35 (3): 293–316. Levie, J., M. Hay, and P. D. Reynolds. 1999. Global Entrepreneurship Monitor (GEM) UK Executive Report. http://www.gemconsortium.org/. Luthje C., and N. Frank. 2002. Fostering entrepreneurship through university education and training: Lessons from Massachusetts Institute of Technology. Brussels: European Academy of Management. Minniti, M., W. D. Bygrave, and E. Autio. 2005. Global Entrepreneurship Monitor (GEM) Executive Report.

64

Roman Frydman, Omar Khan, and Andrzej Rapaczynski

Murphy, K. J. 1999. Executive compensation. In O. Ashenfelter and D. Card, eds., Handbook of Labor Economics. Amsterdam: Elsevier: 2485–2567. Murphy, K. J. 2002. Explaining executive compensation: Managerial power versus the perceived cost of stock options. University of Chicago Law Review 69: 847. OECD. 2004. Financing Innovative SMEs in a Global Economy. Paris: OECD. OECD. 2003. The Sources of Economic Growth in OECD Countries, Chapter 4. Pagano, M., F. Panetta, and L. Zingales. 1998. Why do companies go public? An empirical analysis. Journal of Finance 53 (1): 27–64. Phelps, E. S. 2006. Toward a model of innovation and performance: Along the lines of Knight, Keynes, Hayek, and M. Polanyi. Paper presented at the Conference on Entrepreneurship and Economic Growth, Max Planck Institute and the Kauffman Foundation, Tegernsee (Munich). Polanyi, M. 1962. Personal Knowledge. Chicago: University of Chicago Press. Roe, M. J. 1993. Some differences in corporate structure in Germany, Japan, and the United States. Yale Law Journal 102 (8): 1927–2003. Rydqvist, K., and K. Högholm. 1995. Going public in the 1980s: Evidence from Sweden. European Financial Management 1: 287–316. Sahlman, W. A. 1990. The structure and governance of venture-capital organizations. Journal of Financial Economics 27 (2): 473–521. Schwienbacher, A. 2002. An empirical analysis of venture capital exits in Europe and in the United States. Working paper. University of California, Berkeley. Ueda, M. 2004. Banks versus venture capital: Project evaluation, screening and expropriation. Journal of Finance 59 (2): 602–21. Uslay, C., R. D. Teach, and R. G. Schwartz. 2002. Student entrepreneurs: A cross-cultural analysis of attitudinal differences. Journal of Research in Marketing and Entrepreneurship 4 (2): 101–18. van Stel, A., S. Wennekers, R. Thurik, P. Reynolds, and G. de Wit. 2003. Explaining Nascent Entrepreneurship across Countries. Zoetermeer: EIM Business and Policy Research. van Stel, A., and B. Diephius. 2004. An Empirical Analysis of Business Dynamics and Growth. Zoetermeer: EIM Business and Policy Research. Venture Economics. 1988. Exiting Venture Capital Investments. Needham, MA: Venture Economics. W.H.J. Verhoeven, W. H. J., and J. A. Becht. 1999. Benchmark Ondernemerschap. Zoetermeer: EIM Business and Policy Research. World Bank. 2004. Doing Business in 2005: Removing Obstacles to Growth. Washington, DC: World Bank. Zahra, S. A. 1996. Governance, ownership, and corporate entrepreneurship: The moderating impact of industry technological opportunities. Academy of Management Journal 39: 1713–35.

3

Europe’s Venture Capital Institutions Are Good Enough Richard Robb

3.1

Introduction

According to popular legend, economic innovation begins when an entrepreneur comes up with a bold experimental idea. The entrepreneur takes the idea a certain distance, frequently with seed capital from angel investors, and then pitches the business plan to venture capitalists (VCs). Eventually the entrepreneur matches up with the VC that offers the best fit in terms of personal chemistry, industry experience, and risk appetite. The entrepreneur deploys the VC’s capital to grow the business, while the VC transmits financial and business knowledge that the relatively inexperienced entrepreneur lacks. If all goes well, the entrepreneur comes back to the VC for subsequent rounds to fund further growth. When the time comes, the firm’s investors exit via an initial public offering or a sale to another company. All along the way, the firm spawns employment and spreads vitality throughout the economy. If we believe this legend, we have a ready explanation and a policy response for the seemingly short supply of innovation in European economies: VCs invest less in Europe than the United States, at least partly due to tax and regulatory burdens that European countries impose on domestic VC funds and cross-border investing. Should policy makers reduce these barriers, VC would become more plentiful and innovation in Europe will catch up with the United States. The European Commission offers advice along these lines: “Venture capital funds face obstacles when investing across borders at the same time when innovative firms with high growth potential in Europe face difficulties to access equity because of the fragmented market. Further development of the venture capital industry in Europe requires actions to take in order to reduce the existing market fragmentation for the PE

66

Richard Robb

and VC industry and the support of various industry players is necessary.”1 This chapter, however, challenges the conventional view of venture capital and argues that the standard policy solutions would have little effect: 1. The data show that institutional venture capital finances very little innovation in the United States, Europe, or anywhere else. Even in the United States, the amounts that professional venture capitalists invest in seed and first round deals are miniscule compared with the size of the economy while few recent VC-backed projects would qualify as bold experiments. In most cases VCs keep entrepreneurs on a short leash. The investee companies hew closely to a business plan with a three-year to five-year exit strategy, and the VC can shut the project down at the first sign of serious trouble. Almost all VC projects in the United States are in the information technology and life sciences sectors, which, as Bhidé (2005) observes, account for less than 20 percent of US GDP. 2. European VCs and international investors have already devised practical structures to cope with tax obstacles that appear on the surface to be formidable. It is unlikely that a friendlier tax regime would make much difference. 3. Integration of the Europe’s financial markets would not change existing behavior. European start-ups obtain a portion of their capital from VC funds in their home country and the remainder comes from overseas funds that co-invest with the local fund. This is similar to US practice where the lead investor in a start-up round typically comes from the entrepreneur’s region of the country. Although several theories could explain this phenomenon, this chapter argues that adverse signaling is most important. An entrepreneur who seeks 100 percent of its capital from a VC that is far away will signal that the entrepreneur has exhausted the VCs in its home region. Foreign VCs will refuse to spend resources on due diligence for a project that has been heavily “shopped” and is unlikely to result in a transaction. We can deduce from items 2 and 3 that the relative lack of VC investing in Europe derives from the low supply of projects that VCs like to fund, rather than the other way around. However, from item 1 we do not need to despair. Despite a few dazzling success stories, institutional VC does not contribute meaningfully to innovation in the United States.

Europe’s Venture Capital Institutions Are Good Enough

67

The chapter concludes by arguing briefly that established companies may be an underappreciated source of dynamism. 3.2

Overview of VC Investments in the United States

Several observations support the claim that institutional VC cannot take much credit for funding innovation in the United States: 1. Investment in early-stage US companies is about 0.02 percent of GDP. 2. Almost all investments are in two industries. 3. Investee companies are more mature than the term “start-up” implies. 4. VC’s reputation for risk-taking during the internet bubble is overblown. 5. Business plans are geared towards early exits. 6. VC’s protective rights limit risk if business departs from plan. 3.2.1 Investment in Early-Stage US Companies Is about 0.02 Percent of GDP According to TheDeal.com database, institutional investors participated in 337 seed capital or first-round fundings for US companies between January 1, 2005, and June 30, 2006. The amount invested was not disclosed in 38 of these transactions. The total amount invested in the 299 deals where size was disclosed was $2.9 billion. Median investment for these deals was $6 million and the average was $9.7 million. This equals 0.02 percent of US GDP over the same interval. If we include later-stage rounds of VC funding, which tend to be much larger, the total rises to $18.85 billion or 0.1 percent of GDP over 18 months. The market size computed from TheDeal.com database are basically consistent with statistics reported by National Venture Capital Association (www.nvca.org). According to the industry trade group, VCs invested a total of $22.2 billion in US companies or 0.19 percent of GDP for the calendar year in all rounds combined. We can reconcile the two figures by noting that the National Venture Capital Association totals include investments by the SBIC, venture arms of corporations and investment banks. The Ernst & Young/VentureOne Venture Capital Report for Q4 2005 defines VC as “all investments made by venture capitalists or venture capital-type investors” and arrives at a similar result, finding that total VC investment in 2005 was $21.6 billion.

68

Richard Robb

Even the figures derived from TheDeal.com may overestimate institutional VC funding of early stage companies. First of all, they include angel investors and corporate investors who participated in a round that included an institutional VC. Second, as discussed below, VCs do not inject the entire funded amount at the time they announce a deal. Funding generally comes in several tranches that are contingent upon the company meeting pre-agreed milestones. 3.2.2 Almost All Investments Are in Two Industries As figure 3.1 shows, 90 percent of early stage VC investments in the United States are in information technology/telecom or life sciences. (Life sciences are split into biotech, medical devices, and health care services.) The next largest category at 9 percent is consumer. Consumer deals, however, are largely technology projects geared toward consumers and could easily be classified as information technology. For instance, the largest funding for a “consumer” company over the sample period was $48.5 million for MovieBeam in what was technically a first round. MovieBeam sells video on demand via set-top boxes that receive broadcasts over conventional airwaves. 3.2.3 Investee Companies Are More Mature Than the Term “Start-up” Implies In 1996 National Venture Capital Association’s annual report stated that 77 percent of companies that received VC funding were at least Industrial 3% Other 0%

Telecom 5% Consumer

9% 42%

Life sciences

Information technology

41%

Figure 3.1 Composition of early-stage US VC deals January 1, 2005, to June 30, 2006. Source: TheDeal.com, VC Deal Database

Europe’s Venture Capital Institutions Are Good Enough

69

three years old (Bhidé 2005: 11). Since then the NVCA discontinued reporting age of deals and merely breaks down transaction volume into “early stage,” “expansion stage,” and “later stage,” where “early stage” is defined as a seed or first round. In Q1 2006, the composition was 16 percent for early stage, 41 percent for expansion stage, and 43 percent for later stage. A conspiracy theorist might speculate that the industry’s trade association found that the advanced age of investee companies conflicts with the image that venture capital would like to project. A great deal can be learned about the character of VC investments by looking at a few deals. The three largest “early-stage” deals in our data set were all remarkably well-established companies. Since focusing on the large deals may bias the sample in favor of mature companies, we can see that three randomly selected deals tell the same story. Of the three transactions in the sidebar, the closest to a start-up is probably A&G Pharmaceuticals, which at least plans to use the money for clinical trials of a new product, although it should be noted that diagnostics require a much less arduous FDA approval process than therapeutics. 3.2.4 VC’s Reputation for Risk-Taking during the Internet Bubble Is Overblown During the so-called Internet bubble, start-up entrepreneurs would frequently claim that they had attained vast wealth by inking a deal with an aggressive VC. For example, an entrepreneur might say it had sold 10 percent of the company for $5 million, implying a valuation of $45 million for the stock that was retained. When journalists repeated these claims, the public formed the impression that VCs behaved recklessly. This impression still lingers today. However, a look at elements of the structure of VC investments will reveal that the entrepreneurs’ claims were delusional. The entrepreneur who obtains VC funding has merely taken one step toward realizing the riches in the implied valuation. First, the entrepreneur must build a company that achieves a “liquidity event” (sale to another company or IPO). Following an IPO, the founder’s shares will be locked up for a period of time, so the company must continue to perform well in the public markets before the founders can exit. By the time all this happens, the shares the founder has retained will almost certainly be diluted down to a far smaller percentage of the company.

70

Richard Robb

Three Largest First-Round VC Investments January 1, 2005, to June 30, 2006 1. Company: Investors: Investment date: Amount: Description:

Webroot, Inc. Accel Partners, Mayfield, Technology Crossover Ventures February 2005 $108 m Webroot has been profitable since it was founded in 1998. The company sells “security products that block and eliminate programs that invade personal computers. . . . The new funding will be used to broaden the international expansion of the company’s anti-spyware business.”

2. Company: Investors: Investment date: Amount: Description:

ITA Software, Inc. Battery Ventures January 2006 $100 m Computer scientists from MIT’s Artificial Intelligence Laboratory founded ITA in 1996. According to the company’s website, “ITA Software’s QPX system is the world’s leading airfare pricing and shopping system. QPX has been proven by some of the most demanding customers in the industry, including Orbitz, Alaska Airlines, Alitalia, America West Airlines, Continental Airlines, Galileo, Cheap Tickets, Kayak, Accovia, NLG, and others.” The company plans to use its first round funding to expand from 175 to 300 people.

3. Company: Investors: Investment date: Amount: Description:

Verus Pharmaceuticals, Inc. 7 venture capital funds June 2005 $78 m Verus was founded in 2002. From a press release on July 6, 2005, the company “acquired exclusive, worldwide rights to Twinject. . . . Twinject, a novel epinephrine auto-injector indicated for the emergency treatment of severe allergic reactions (anaphylaxis), was approved by the US Food and Drug Administration (FDA) in 2003.”

Europe’s Venture Capital Institutions Are Good Enough

First-Round VC Investments: First Three in Alphabetical Order January 1, 2005, to June 30, 2006 1. Company: Investors: Investment date: Amount: Description:

A Place for Mom, Inc. Battery Ventures February 2006 $9.5 m According to the company’s website, “A Place for Mom is the nation’s largest Eldercare Referral Service. . . . A Place for Mom has over 10,000 longterm care communities participating in its referral network. This includes all of the top twenty largest assisted living companies in the United States.” The company will use the money to expand and to provide liquidity to angel investors and employees.

2. Company: Investors:

A&G Pharmaceuticals, Inc. Round was led by New England Partners and included a fund sponsored by Maryland Department of Business and Economic Development September 2005 $2 m Founded in 2000. From the website, “Over the last four years, A&G has become the preferred partner for more than 14 biotech and pharmaceutical organizations in monoclonal antibody development.” Obtained funding to complete clinical trials of breast cancer test kit.

Investment date: Amount: Description:

3. Company: Investors: Investment date: Amount: Description:

Accordent Technologies TVC Ventures March 2006 $4 m Founded in 1999. From the website, “More than 1,500 organizations worldwide, including Unisys, the United Nations, National Institutes of Health, and Oracle, rely on Accordent for their online communications.”

71

72

Richard Robb

Institutional VCs never invest in common shares alongside the founders but rather bargain for preferred stock that confers many rights and mechanisms for protecting those rights. A few such provisions in nearly every preferred stock agreement are: Dividend payments VCs receive an annual preferred return of approximately 8 percent This is usually expressed as a cumulative dividend. The dividends are not paid periodically, but at the time of any exit. Liquidation preference In the event of liquidation, sale, or winding-up of the company, the preferred investors receive a multiple of their investment plus the preferred return before any money is distributed to the common shareholders. Liquidation multiples can range up to 6×, so the VCs in our example would receive the first $30 million plus the preferred return following any liquidity event. Anti-dilution If the business stumbles at any point and the entrepreneur is forced to raise money at a lower implied valuation, anti-dilution provisions quickly shrink the entrepreneur’s share of the company. For example, in the case of full ratcheting anti-dilution in our example, $1 raised at a $25 million valuation would mean that the VC’s preferred shares convert into 20 percent of the company rather than 10 percent as specified in the initial terms. The extra shares come at the expense of the shareholders who entered prior to the VC. Vesting of founders’ shares At the time a VC invests, common shares the entrepreneur had previously owned will start to vest according to a schedule. 3.2.5 Business Plans Are Geared toward Early Exits VC funds are typically structured with five-year investment periods and ten-year final lives. During the investment period the VC manager can draw down capital that its limited partners have committed. Early transactions that generate liquidity during the investment period may be recycled into new investments. After the investment period, the fund manager focuses on realizing liquidity from the portfolio it has assembled. To the extent it succeeds in converting portfolio investments to cash over the predetermined time frame, the manager receives an incentive fee and will be able to raise new funds. The most important ways to ensure timely exits are to pick the right companies and to guide these companies in the desired direction after the investment takes place. Additionally the standard preferred stock agreement contains two provisions that strengthen the VC’s hand:

Europe’s Venture Capital Institutions Are Good Enough

73

Redemption The preferred shareholder can force the company to redeem the preferred shares in three to five years at the purchase price plus the preferred return. The threat of a redemption that would deprive the company of its financial resources can motivate the company to pursue a liquidity event. Registration rights The preferred shareholder can demand that the company register for an IPO at the company’s expense at any time. 3.2.6 VC’s Protective Rights Limit Risk If Business Departs from Plan Funding based on milestones VCs fund preferred stock is in several tranches. In the example above, the VCs may agree to inject $1 million every six months contingent on milestones, such as signing up a certain number of customers or completing phase II trials of a drug. Negative covenants The company must seek the preferred shareholders’ approval if the company wants to change the business relative to the plan, issue new securities, merge, or acquire another business. The VCs can use this leverage to replace the CEO when the company is struggling. Board representation VCs stay closely involved with a company in order to transfer expertise, but also to gather information so that they can enforce their rights. The preferred stock agreement usually guarantees one or more board seats to the VCs. Additionally the preferred shareholders are guaranteed “information rights” such as regular financial reports or other information they reasonably request. How can it be that US VCs are so careful and invest so little in genuine start-ups yet a dozen or more household names follow the arc of entrepreneur-with-an-idea → VC fund → hugely profitable IPO? To indulge in a psychological explanation, the VC success stories are romantic and well known. They include Compaq, Netscape, Symantec, Sun Microsystems, Amazon, Google, and eBay. Frequently their histories are embellished by imaginative foundation myths (e.g., the Pierre Omidyar has confessed that he made up the story about inventing eBay so that his wife could trade Pez dispensers online). Even the name “venture capital” with the same root as “adventure” sounds exciting. We may suppose that for every one we know about, hundreds fall below the popular radar. For every uber-VC investor like John Doerr, thousands toil in relative obscurity. But the data suggest otherwise.

74

Richard Robb

While the public doesn’t care to know about every inverter duty motor that Emerson Electric develops and markets for hundreds of millions, entrepreneurs who beat the odds tell a tale that everyone wants to hear. 3.3

European VC

The estimated size of the institutional VC market in Europe is somewhat smaller than that in the United States. Every single early-stage round in our data set contains at least one European VC, and usually the VC comes from the same country as the investee company. Most of the larger deals include at least one US VC fund. TheDeal.com database lists 47 seed or first-round deals for European companies that closed between January 1, 2005, and June 30, 2006. The median deal size was $5.3 million, only slightly smaller than in the United States. The investee companies were located in 13 countries: Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Netherlands, Norway, Spain, Sweden, Switzerland, and United Kingdom. In a similar pattern to that of the United States, 46 out of 47 deals were classified as information technology, life sciences, or telecom. The Ernst & Young/VentureOne Venture Capital Report pegs Europe’s VC market at ≤3.6 billion (approximately $4.5 billion) for 2005 for all rounds combined. This equates to 0.04 percent of GDP or roughly a quarter of the ratio in the United States. The E&Y report estimates that “early-stage” deals make up 31 percent of the all VC investments, which would bring the early stage total to $1.4 billion or half of the amount in the United States. (TheDeal.com database is not reliable for measuring the market’s size. From a personal conversation, TheDeal. com believes it captures nearly every transaction in the United States and about half in Europe.) The European Private Equity Survey published by the European Venture Capital Association estimates from a survey of VCs that European venture capital investment in 2005 was ≤10.8 billion or 0.11 percent of GDP. This seems implausibly high, since it would mean that Europe’s VC market was approximately the same size as that of the United States.2 Investors’ identities were disclosed in 42 of the 47 European deals over the sample period in TheDeal.com database. As in the United States, the median number of investors in these deals was 2. All 42 such deals contained at least one European investor in the round. Out of the 14 deals larger than $10 million, 9 contained at least one US VC and 7 out of 28 of the deals smaller than $10 million contained at least one

Europe’s Venture Capital Institutions Are Good Enough

75

US VC. In 35 cases out of 42, the round contained at least one VC from the investee’s country. Most of the exceptions included a VC from a neighboring country, such as Affiris, the Austrian pharmaceutical company, that closed a ≤8.5 million round with MIG Funds from Germany. Interestingly it is hard to find a single Italian start-up funded with VC capital in the past several years. It is equally hard to explain the phenomenon by a lack of Italian VC funds with the requisite skills. Consider, for example, CDB WebTech. CDB WebTech is a publicly traded fund that invests in other VC funds and makes direct VC investments. It is chaired by one of Italy’s leading tycoons, Carlo de Benedetti. The other senior managers are Vittorio Moscatelli (CEO), Giuseppe Pasetti (vice president of investments), Paolo Bonetti (CFO), and Cristina Tolentino (senior accountant), all Italians working out of CDB WebTech’s Milan office. Yet CDB WebTech’s direct portfolio does not contain an Italian company. Apax is one of Europe’s two or three most active VCs and has an office in Milan, but only 3 out of its 500 portfolio companies are Italian: Newron Pharmaceuticals (2002), HAL Knowledge Solutions (2003), and Keylan SpA (2002). Another major European VC investor, 3i, has participated in ten leveraged buyouts of mature Italian companies since 1997 but made only one Italian VC investment when it co-invested in Newron along with Apax. Thus we can point to three well-heeled institutions in Italy that are set up to invest in Italian start-ups. The absence of many transactions suggests that Italian entrepreneurs are to blame. Evidently they do not start biotech, information technology, or telecom companies, or if they do, they do not turn to VCs for funding. 3.4

Tax Structures

The literature on entrepreneurship routinely cites tax as a partial explanation for the failure of European venture capital to flourish. For instance, the OECD’s “Going for Growth 2006” lists the four main “policy determinants that influence the supply and demand for venture capital” and reason number one is “[e]xcessive taxation of capital income and capital gains that reduces . . . the willingness of individuals to commit savings to venture funds.” It is understandable why the OECD analysts would arrive at this conclusion. If a VC were to march into a continental European country and mechanically set up a fund, the tax burden would be overwhelming. To take an extreme example,

76

Richard Robb

a German company could pay its own income tax on profits that it earned, the fund that invested in the company could pay tax on dividends received or capital gains on selling the company, and investors in the fund might have to pay a third round of tax on any profits that remained. The manger’s fees could be subject to VAT and then a layer of income tax. Even worse, a foreign venture capitalist opening an office on the ground in Germany to make direct investments in Germany on behalf of a non-German fund might create a “permanent establishment” and trigger German tax on a portion of the fund’s global income. The obvious policy prescription for promoting European VC funds is to ease the tax burden. However, a closer look at existing practice suggests that such policies may not be needed: fund managers have already figured out ways to cope with the status quo. With a bit of structuring, managers routinely invest from overseas or set up offices in European countries without attracting double or triple tax, capital gains tax for nonresident investors, withholding tax, or substantial VAT on management fees. These structures are commonplace, legal, and practical. One example will illustrate the point. The German biotech company Elbion AG, raised $33 million from VCs when it was spun off from a German chemical manufacturer in the largest European VC first round funding of 2005. The round attracted seven institutional investors including five venture capital funds. As shown in table 3.1, two VC fund managers were based in Germany, two came from outside Germany, and one is headquartered in London but sources its German investments from its offices in Munich, Stuttgart, and Frankfurt. None of these investors would subject their funds to tax at the fund level. How do they do it? It is worth reviewing the funds’ techniques in a bit of detail,3 since the policy implications are important. We will consider two countries that more or less span the space of tax regimes: United Kingdom and France. France, in particular, proxies for Spain, Germany, Italy, Portugal, and Austria. 3.4.1 United Kingdom The United Kingdom is the most favorable European jurisdiction for an asset manager. The UK Inland Revenue will not impose tax on a fund managed by a UK resident manager investing in UK or non–UK assets if three tests are satisfied:

Europe’s Venture Capital Institutions Are Good Enough

77

Table 3.1 Investors in Elbion AG first-round funding, January 2005 Investor

Location

Investor type

3i Group plc

Headquartered in London; offices in 14 countries in Europe, United States and Asia; 3 offices in Germany

VC fund manager

AGF Private Equity

France

VC fund manager

BayTech Venture Capital

Germany

VC fund manager

Burrill and Co. LLC

United States (northern California)

VC fund manager

DVC Deutsche Venture Capital GmbH

Germany

VC fund manager

Marubeni Corp.

Headquartered in Japan, with offices in 74 countries

Giant trading company, service provider to Elbion

Quintiles PharmaBio Development Group

Headquartered in United States with offices in 50 countries

Biotech consultant, service provider to Elbion

1. The fund is not a UK resident fund. 2. The investment manager is not a branch of the fund. 3. The fund is not controlled by a UK corporation. The first test can be satisfied by incorporating the fund offshore, such is in the Cayman Islands. While the fund should be controlled at the highest level outside the United Kingdom, the UK asset manager can be delegated broad decision making powers under the asset management contract. The United Kingdom offers an Investment Management Exemption that enables UK resident managers to establish that they are not a branch of the offshore fund. To qualify, the manager must carry out the business of providing management services (not conduct a trade or business), not be entitled to more than 20 percent of the gains and profits of the fund, and not carry out any other business besides investment management. The Investment Management Exemption does not exist in continental European countries and may partly explain the appeal of London as a center for fund management. Finally, the third test will be satisfied as long as the fund is not owned more than 50 percent by a UK corporation.

78

Richard Robb

A UK resident asset manager organized to meet these three tests will have a free hand to invest on behalf of an offshore fund without risk that Inland Revenue will impose tax at the level of the fund. Management fees paid to a UK manager will not be subject to VAT if the UK manager acts as a subadvisor to a non-EU manager that in turn manages the fund. The provisions above are enough to ensure that the fund, itself, will not incur UK tax. If the fund invests in European portfolio companies, such as in France, it needs to take additional measures so that French revenue authorities will not impose withholding or capital gains tax when the fund sells its portfolio investments. We now turn to the structuring that will solve this issue. 3.4.2 French VC for Nonresidents We will consider the case of an offshore fund, say resident in the Cayman Islands, for international investors. These investors will, of course, be responsible for any tax due in their home country, as are foreign or domestic investors in a US VC fund. The key is to show that dividends or capital gains can make their way to the end investors free of tax at the fund level. The most common structure involves interposing a Luxembourg holding company (“Luxco”). Luxco invests in securities issued by the French start-up company (“investee”). As in the United States, the usual form of investment would be preferred stock that is convertible into common equity. Under French domestic rules, capital gains on shares are taxed at 16 percent if derived by nonresidents, but under the French-Luxembourg treaty, capital gains on shares derived by a Luxembourg tax resident are taxable only in Luxembourg. The EU Parent/ Subsidiary Directive ensures that a dividend paid by the investee company would not be subject to French withholding tax provided that Luxco holds at least 20 percent of the investee (15 percent from January 1, 2007) and provided further that Luxco has not been interposed for the only purpose to take the advantage of this exemption. The last condition means that Luxco needs some substance. In particular, all board and shareholders meetings have to be held in Luxembourg. Moreover such capital gains and dividends are exempt in Luxembourg due to Luxembourg’s “participation exemption” for parent companies that participate on more than 25 percent of a subordinate company. The final step is shifting these gains up to the Cayman fund. This is accomplished by issuing “convertible preferred equity certificates” or CPECs

Europe’s Venture Capital Institutions Are Good Enough

79

that the Cayman fund buys. The Cayman fund obtains payments made by reimbursement of principal and accrued interest on the CPECs, and the excess profits in the form of partial or total liquidation of Luxco. (A regular dividend is subject to withholding tax in Luxembourg while liquidation proceeds from either total or partial liquidation are exempt from withholding tax.) Why does Luxco issue a CPEC rather than a security with identical characteristics to the security Luxco buys from the French investee? The key is that CPECs are treated as debt for Luxembourg corporate income tax, withholding tax and capital duty purposes, and so will be exempt. At the same time CPECs are treated as equity from a legal perspective. Importantly this allows Luxco to comply with Luxembourg’s thin capitalization rules that will impose punitive duties if a company’s capital structure comprises less than 15 percent equity. This structure will work for a domestic French fund manager, a foreign manager that sets up on office in France or a fund manager that manages a fund entirely from outside of France. A French office of a non-French manager should not execute or implement transactions but merely advise either of the investing entities (i.e., the Cayman fund or Luxco) or the parent fund manger (offshore from France) so as to avoid creating a taxable French permanent establishment. If the target business is not in the real estate area, typically there will be no French wealth tax issues for nonresident investors in the Cayman fund as shown in figure 3.2 and table 3.2. 3.5

Geographic Matching of VCs and Entrepreneurs

Even though funds can overcome tax obstacles to cross-border investing, European VC nevertheless has a local component, as we saw in section 3.3. Several theories could explain why each round includes at least one European investor, usually from the entrepreneur’s home country: (1) The tax structuring described in the previous section is prohibitively expensive. (2) The VC and entrepreneur need to share a common culture and language; proximity will facilitate due diligence, knowledge transfer, and monitoring. (3) After an entrepreneur solicits an investment from a VC and the VC decides to proceed with due diligence, the home country will embark on a process that imposes costs on both sides. A project that has already been rejected by many of the VC’s rivals is probably a waste of time. Entrepreneurs can conceal past rejections but they cannot hide their country. So an entrepreneur coming

80

Richard Robb

Nonresident investors

Fund (Cayman Limited liability Company)

Asset management relationship

VC (UK/US/Offshore) HQ (Cayman/US/UK)

Provision of advice/ PECs/CPECs recommendations (in return for appropriate advisory fee)

Capital

Lux Co (Luxembourg)

100% owned

Equity/convertible debt/warrants

Local VC house/ representative (France)

Management and other shareholders

Investee/target (France)

Figure 3.2

Key: PECs = preferred equity certificates CPECs = convertible PECs Gray boxes = SPVs

Europe’s Venture Capital Institutions Are Good Enough

81

Table 3.2 Sample French VC fund managers Sample French fund managers

Non-French managers with Paris offices

A Plus Finance

Apax

Access Capital Partners

3i

Ace Management

Quintiles

Acland Capital Investissement Activa Ventures Alven Capital Angel Invest Atria Capital Partners Aurel Leven NextStage Auriga Partners AurInvest Avenir FinanceGestion Astorg Partners

from far away signals that VCs closer to home have already said “no”— the San Francisco Internet start-up seeking funding in Spain, for example, will arouse every suspicion. It would be unlikely to command attention in the absence of a strong explanation about why it has strayed so far from the United States.4 It is important to distinguish between these theories from a policy perspective. If we find that theory 1 explains the localization of VC investing, then opening up the financial markets to ease cross-border investing could have an impact. If theory 2 or 3 is more important, then entrepreneurs will continue to obtain capital from local funds after the market is integrated and little will change. In this section we develop a model to make theory 3 more precise and then test the model informally by analyzing data from the United States. We find that localization persists in the absence of tax or cultural barriers, since West Coast VCs tend to fund West Coast entrepreneurs and East Coast VCs tend to fund East Coast entrepreneurs. Consider entrepreneurs who arrive at domestic VCs sequentially to present their projects. The number of VCs in the entrepreneur’s home country is exogenously fixed at n. For simplicity, a VC can either accept or reject a project, and if accepted, the deal’s terms follow the market standard. The probability an entrepreneur’s project will be accepted by a VC after the VC performs due diligence is the outcome of a random variable Θ that is fixed for each entrepreneur. The entrepreneurs do not

82

Richard Robb

know their own value of θ, but entrepreneurs and VCs do know the probability density function, f(θ). An entrepreneur’s value of θ can be interpreted as the chance a VC will identify the merits of the project; VCs differ with regard to their judgment as well as their ability to contribute to a project through business contacts and active participation. This setup would be identical to an academic who submits a paper to journals—after striking out many times, it is possible that a journal will see virtues that others missed or appoint an editor who can bring the paper up to acceptable standards. The entrepreneurs differ with regard to the costs of presenting to a VC. We assume the jth entrepreneur’s cost of finding a domestic VC and undergoing due diligence is a fixed amount. After being rejected, the entrepreneur has to decide whether to quit or to pay the cost of presenting to the next VC. Denote the jth entrepreneur’s cost by cj with probability distribution function Fc(c). In practice, an entrepreneur’s cost might decline with each subsequent approach to a VC if, for example, she can re-use presentation materials. This sort of efficiency might be offset by higher marginal cost of contacting VCs (the entrepreneur will start by contacting the one who is easiest to find) or fatigue. In any case, a declining cost function would not change the basic result of our model and would needlessly complicate the entrepreneur’s stopping rule. The probability a random entrepreneur will succeed on the first visit to a VC is simply E[Θ]. The probability of success on the second visit is lower, since rejection the first time will convey a lower value of θ. Applying Bayes’s law, we have P[success on the second trial] =

∫ θ (1 − θ ) f (θ )dθ , ∫ (1 − θ ) f (θ )dθ

and

∫ θ (1 − θ ) ∫ (1 − θ )

i −1

P[success on the ith trial] =

i −1

f (θ )dθ f (θ )dθ

≡ pi .

In the special case that Θ is uniformly distributed on the unit interval, the probability of success on the ith trial reduces to pi = 1/(1 + i). Assuming risk neutrality, an entrepreneur quits immediately before the ith trial if cj > piV and cj ≤ pi-1V, where V is the value the entrepreneur places on persuading a VC to fund her project. For i = 2, . . . , n, the probability a randomly selected entrepreneur will choose to undergo

Europe’s Venture Capital Institutions Are Good Enough

83

the ith trial given rejection on trial i − 1 is P[c > piV] = Fc(piV). Without altering any of the analysis, we could allow V to vary across entrepreneurs by defining cj as the ratio of the cost to the entrepreneur’s private value of success. The VC, too, incurs a cost of evaluating a potential investment. VCs cannot observe θ while entrepreneurs are able to conceal whether or not they have undergone due diligence. Thus VCs cannot distinguish between entrepreneurs a priori, so any VC that is set up for business will evaluate each entrepreneur that arrives. (Some projects will be unworthy on the surface and never receive a hearing. We can assume that these are excluded from the analysis, so that f(θ) represents the distribution of entrepreneurs that VCs are willing to consider.) From a VC’s point of view, the probability of success of any evaluation is n

P[VC succeeds] = ∑ P[entrepreneur is arriving for its ith trial] i =1

× P[success|the ith trial] p1 + (1 − p1 )p2 × Fc ( p1V ) + (1 − p1 )(1 − p2 ) p3 × Fc ( p2V ) + …(1 − p1 )(1 − p2 )…(1 − pn−1 ) pn Fc ( pn−1V ) . = 1 + (1 − p1 ) × Fc ( p1V ) + (1 − p1 )(1 − p2 ) × Fc ( p2V ) + …(1 − p1 )(1 − p2 )…(1 − pn−1 )Fc ( pn−1V ) Table 3.3 shows the probability a VC succeeds on each trial taking the special case f(θ) = 1 for 0 ≤ θ ≤ 1. The first column is the total number of VCs in the market. The second column gives the probability of success assuming Fc(pn-1V) so that entrepreneurs never become discouraged. The last column gives the probability of success assuming Fc(piV) = 0.5 for i = 1, . . . , n − 1 (i.e., there is a 50 percent chance the entrepreneur will give up after one rejection and a 50 percent chance an entrepreneur will not give up until he has exhausted all n VCs.) Not surprisingly, the chance a VC will succeed declines as the market grows more crowded. The situation is worse if entrepreneurs are not easily discouraged. Imagine an entrepreneur who has been rejected by all ten domestic VCs and would like to try her luck overseas. We will assume that her cost of going abroad is higher, but not so high that she would be unwilling to incur cj + ε in exchange for a probability θ of success. Consider this trial from the perspective of the foreign VC. The VC can deduce from the entrepreneur’s location that all ten domestic VCs must have rejected the entrepreneur’s project. If f(θ) = 1 for 0 ≤ θ ≤ 1 as in table

84

Richard Robb

Table 3.3 Probability VC succeeds in consummating and investment after engaging in due diligence with f(θ) = 1 for 0 ≤ θ ≤ 1 Probability VC succeeds on each trial Number of VCs in the market

Entrepreneur is never discouraged

50% chance entrepreneur is discouraged after first rejection

1

0.500

0.500

2

0.444

0.467

3

0.409

0.441

4

0.384

0.422

5

0.365

0.406

6

0.350

0.393

7

0.337

0.383

8

0.327

0.374

9

0.318

0.366

10

0.310

0.359

3.3, the chance of success on the eleventh trial is p11 = 1/11 = 9.1 percent. (With more than one VC in the foreign market, p11 will be even lower because there is a chance the entrepreneur was rejected by a foreign VC.) If risk-neutral VCs face the same cost function in each of two countries with free entry, firms will enter the market as long as the probability of success multiplied by the value of projects given success is equal to the cost of due diligence. In the case where entrepreneurs are never discouraged, a domestic VC would have a 31 percent chance of success after evaluating a project while a foreign VC’s chance of success would equal to 9.1 percent. The structure where local entrepreneurs present to local VCs is a stable equilibrium: if the VCs expect entrepreneurs to begin with the home country VCs, then VCs will refuse to evaluate foreign entrepreneurs. Note that the result holds even if ε is a small positive number. As long as the cost differential in greater than zero, the entrepreneur traveling abroad will signal to VCs that she cannot raise money back home. Competing explanations for the localization of VC investing rely on costs or taxes. Perhaps it is much more costly for VCs to identify and monitor entrepreneurs who are far away to due travel or cultural or language barriers while structuring to avoid cross-border tax may expensive and risky. A simple, informal test can distinguish between these theories. The test relies on a natural experiment in the United

Europe’s Venture Capital Institutions Are Good Enough

85

Miss. Valley and Great Plains 1% Rocky Mountains 4%

Deep South 0%

Washington, DC Beltway 5%

Southeast 5% Great Lakes

Pacific Northwest and Hawaii

5% 6%

47%

California

25% NY Metro, New England Figure 3.3 US VC funding by region, all rounds, January 1, 2005, to June 30, 2006. Source: TheDeal. com, VC Deal Database

States. In the United States, entrepreneurs who fund from VCs lie principally on the two coasts. VCs, too, are principally lined up on the coasts. VCs from New York can invest in California companies without encountering material language barriers, risk of withholding tax or most of the costs we associate with cross-border investing. Thus the test relies on examining the extent to which entrepreneurs from the each coast receive start-up funding from VCs outside of their home state or region. If entrepreneurs and VCs match up frequently outside of their home region, this finding should support the cost theory and argue against the signaling theory. The analysis will be restricted to seed and first round investments, since our interest focuses on funding of start-ups. We will only consider deals with at least one professional venture capital investor, so deals that are funded entirely by angels or companies such as Cisco Systems will be excluded. Further we ignore rounds less than $1 million or rounds where the amount invested was not disclosed. The data comes from the TheDeal.com venture capital database for the period from January 1, 2005, to June 30, 2006. First, consider California start-ups. During the 18 months ended June 30, 2006, 106 California-based companies obtained seed or first round funding of at least $1 million from a round that contained at

86

Richard Robb

least one institutional investor. The median number of institutional investors was two. The largest deal size was $78 million and the average was $10.5 million. The results are unambiguous. The 31 largest deals each contained at least one California-based VC and for the few deals that consisted of non-California VCs investing in California companies, we can see a natural connection. For example, the largest deal that did not include a California VC was the $9.2 million round in May 2006 for Imagine Communications. The lead investor was Israel-based Carmel Ventures, which was joined by the East Coast firm, Columbia Capital. This is hardly a case of Imagine wandering off to find VCs in far-flung lands: while Imagine is headquartered in San Diego, it operates its R&D lab in Israel and both of Imagine’s founders are Israeli. The second largest California deal funded by non-California VCs was Lala.com. The VC was Ignition Partners, just over the border in Oregon. Mountain View California-based LignUp.com raised $5.9 million from three Utahbased funds and one Japanese; LignUp’s founder grew up in Utah, lived in Japan, and speaks Japanese. Altogether, 7 out of the 106 deals were funded from managers outside of California. Two out of these seven involved VCs from Washington and Oregon and two more involved VCs in states that border California. If we look at New England and NY Metro Area firms, we find a similar pattern. During the same time period, 50 companies in New England and the NY Metro Area received seed or first-round company funding. Forty-six out of 50 including the 19 largest all contained at least one VC fund located in the same region. Like the California deals, the median number of investors was two. In summary, California entrepreneurs include at least one Californiabased VC in nearly every start-up round while East Coast firms include at least one East Coast VC in nearly every start-up round. This observation, together with the finding that European start-ups always contain at least one European VC in every round, fits with our model. Signaling, rather than cross-border tax, language and cultural barriers may best explain the tendency for entrepreneurs to attract a portion of their funding from VCs in their home country. In this case, integrating Europe’s financial market as the European Commission recommends will not change behavior. Local institutions (including foreign outposts of international funds who are currently free to enter European countries as long as they take care to avoid becoming permanent establishments) will fund the local projects and draw on co-investment from foreign firms to the extent that local funds do not have sufficient capital.

Europe’s Venture Capital Institutions Are Good Enough

3.6

87

Innovation in Established Firms versus VC Backed Start-Ups

As stereotypes may overstate the amount of innovation that takes place in VC-backed start-ups, so may stereotypes understate the innovation at established companies. Schumpeter himself identified the conditions that enable large, profitable firms to gamble on new ideas: “Innovation is, in this case [‘trustified capitalism’], not any more embodied typically in new firms, but goes on, within the big units now existing, largely independently of individual persons. It meets with much less friction, as failure in any particular case loses its dangers . . . taking a long-term view towards investment becomes possible.” Scherer (1998) and many others take the opposite view fearing “abominable no-men” that lurk within giant corporations to oppose creativity. Jovanovic (2001) argues that large firms may have had held advantages during the last two-thirds of the twentieth century, but times are changing too quickly for bureaucracies to keep up and so we have “entered the era of the young firm.” This is, of course, a long-standing debate with scholars on both sides and beyond the scope of this chapter to weigh in comprehensively. But it may be worth briefly considering an industry characterized by radical innovation over the past twenty-five years: international capital markets. A trader from a generation ago would not recognize a modern dealing room or the products banks and dealers offer their customers. For good or for ill, we have seen the growth of derivatives in every flavor with notional principal in excess of $240 trillion.5 Securities and futures trading have moved from clumsy physical exchanges to electronic execution. Book entry has supplanted physical delivery. Electronic confirmation has replaced mountains of paper. Who was been behind these innovations? Big banks and dealers, law firms, futures and stock exchanges, industry groups such as the International Swaps Dealers Association, clearing houses, and the Bank for International Settlements have led the way. With the exception of Bloomberg LP that started up in the early 1980s (and without VC backing) and not counting hedge funds, it is hard to find a significant contributor to financial innovation that fits the traditional definition of an entrepreneurial firm. (Some minor exceptions include a handful of start-ups that provide accounting software, pricing models, trading platforms, and other bits of technology.) That large firms innovate is hardly an original idea. Phelps (2006) observes, “the heavy research and development expenditure in the sector of established firms is circumstantial evidence that many large firms are oriented towards innovation.” While these things are hard to

88

Richard Robb

quantify, entrepreneurial activity within institutionally backed startups is orders of magnitude smaller than the entrepreneurial activity within large firms. To take another example, the amount Boeing gambled to develop the 787 Dreamliner is reportedly $5.8 billion, more than all institutional VCs spend globally on start-ups in first-round financings in a year. This suggests we might learn more about why corporatist Europe lags in terms of innovation by examining the behavior of established companies. The compressed wage scales, job security, and close ties between government and industry indicate that a far larger share of big European firms’ profits derives from rent seeking rather than from problem solving. 3.7

Conclusion

I am seeking to make a modest contribution to the debate on the presumed innovation deficit in European economies relative to the United States. Among possible causes such as high taxes that discourage effort and risk taking, education, culture, government regulation of business and labor, inadequate bankruptcy laws, and underdeveloped financial institutions for channeling capital to start-up companies, I would argue in favor of striking the last item from the list. Europe’s VC markets are perfectly adequate. Through careful structuring, European VCs have set up domestic funds and US investors have figured out efficient ways to invest from overseas. Still the data show fairly little cross-border investing; but this is probably because a start-up that tries to raise money from far away signals that it has exhausted opportunities in its home market and is a bad bet for VCs to spend resources on due diligence. We observe less institutional VC in Europe than the United States because European entrepreneurs generate fewer life sciences and information technology projects that fit with the VCs’ investment strategy. Even if European VC were to catch up fully with the United States, Europe’s economy would not become noticeably more dynamic, since the impact of VC in the United States has been exaggerated. Thus policies to promote institutional European VC, including “full integration of financial markets,” are unlikely to have much of an effect. Notes 1. “Merits and Possibilities of a European structure for venture capital funds” zhttp:// ec.europa.eu/enterprise/entrepreneurship/financing/publications_documents.htm.

Europe’s Venture Capital Institutions Are Good Enough

89

2. The European Commission uncritically cites this measure of industry size in numerous documents including a June 30, 2006 press release: “The risk capital markets in Europe: still too small. . . . VC investments were ≤10.8 billion in the EU versus ≤16.5 billion in the US (EVCA; NVCA).” If this were true, one wonders why the Commission would be alarmed by a differential of ≤5.7 billion p.a. considering the GDP of both economies exceed ≤10 trillion. The US data seems far more trustworthy, since the NVCA data is checked with public sources and portfolio companies while the EVCA data relies on surveys sent to “private equity firms.” Also, the NVCA measures of industry size correspond to The Ernst & Young/VentureOne Venture Capital Report and TheDeal.com micro data, while the EVCA statistics do not. 3. The discussion in this section is not intended to be used for the purpose of providing tax advice. 4. The entrepreneur may try to overcome adverse signaling by arguing that its project fits naturally with a foreign VC, and hence the entrepreneur skipped due diligence in its home country. The entrepreneur might appeal to a compelling personal connection with the VC, a similarity between one of the VC’s portfolio companies and the entrepreneur’s project, or a desire to establish contacts in the VC’s country where it sees a market for its products. We will see a few examples later in this section where this tactic seems to have worked. 5. http://www.isda.org/statistics/pdf/ISDA-Market-Survey-historical-data.pdf.

References Bhidé, A. 2006. How novelty aversion affects financing options. Capitalism and Society 1 (1): article 1. European Commission. 2005. Merits and possibilities of a European structure for venture capital funds. http://ec.europa.eu/enterprise/entrepreneurship/financing/ publications_documents.htm. European Commission. 2006. Best Practices of Public Support for Early-Stage Equity Finance. Directorate-General for Enterprise and Industry, European Commission, September 2005. Brussels. Jovanovic, B. 2001. New technology and the small firm. Small Business Economics 16(1): 53–55. OECD. 2006. Going for growth: Structural policy indicators and priorities in OECD countries. www.oecd.org/document/7/0,2340,en_2649_201185_35995079_1_1_1_1,00.html. Phelps, E. 2006. Capitalism theory. The Center on Capitalism and Society. http://www .earthinstitute.columbia.edu/ccs/theory.html. Scherer, F. 1988. Testimony before the Subcommittee on Monopolies and Commercial Law. Committee on the Judiciary, US House of Representatives, February 24. Schumpeter, J. 1928. The instability of capitalism. Economic Journal 38: 361–86. Reprinted in R. Clemence, ed., Essays on Entrepreneurs, Innovations, Business Cycles and the Evolution of Capitalism (1989). New Brunswick, NJ: Transaction Publishers.

4

Promoting Entrepreneurship: What Are the Real Policy Challenges for the European Union (EU)? Anders N. Hoffmann

4.1

Introduction

The European Union (EU) wants to boost entrepreneurship as part of its strategy to transform its economy and build up its future economic and competitive strength. The Commission has published several reports highlighting the importance of entrepreneurship and identifying EU challenges in this domain. The Green Paper on Entrepreneurship states that the main challenge for the European Union is “to boost the Union’s levels of entrepreneurship, [by] adopting the most appropriate approach for producing more entrepreneurs and for getting more firms to grow” (European Commission 2003; 9). Furthermore the current Action Plan for Promoting Entrepreneurship states: “The EU is not fully exploiting its entrepreneurial potential. It is failing to encourage enough people to become entrepreneurs. . . . Europe, unlike the US, suffers from low expansion rates after startup. . . . Whereas US entrepreneurs appear to test the market and, if successful, expand rapidly, many business ideas in Europe never come to market, as their viability is questioned before they can be tested in the market place” (European Commission 2004: 3–4). The European Council has also addressed the need for more entrepreneurship and discussed possible policy measures on several occasions. The European Council’s latest policy conclusions encourage member states to “strengthen the measures to promote a more entrepreneurial culture and the skills to encourage more people to consider a career as entrepreneur, including through entrepreneurship education and training at the appropriate level of education” (EU Council 2006; 11). This chapter addresses two questions relating to the European Commission and the Council’s recommendations: Have the European Union

92

Anders N. Hoffmann

identified the right challenges? Have the most appropriate policy measures been suggested? The first question will be answered by presenting data on firm creation and firm growth from three sources: (1) the work on firm demography published by EUROSTAT, (2) the work on international comparability of firm creation published by the OECD, and (3) new work based on firm registrar in Statistics Denmark (EUROSTAT 2005; Vale 2006). These data will be supplemented by new analyses on highgrowth firms based on a comprehensive commercial database of business accounts. The second question will be addressed using a regression analysis and a comparative analysis of indicators. The answer to this question will be more speculative than the answer to the first question as the issue is more complex and builds on fewer facts. The comparative analysis is performed within a general theoretical framework for the business environment affecting entrepreneurial performance. The general framework, which builds on Lundström and Stevenson (2005) and Verheul et al. (2003), is taken from Hoffmann (2006) and is quantified by 61 indicators (Hoffmann et al. 2005). Policy actions are very broadly defined as any change in institutions, regulations, or tax and support schemes. A notable amount of literature has addressed the differences between the European Union and the United States. Several analyses stand out as inspiration for the work in this chapter. Audretsch et al. (2002) offers a comprehensive theoretical framework and detailed country analysis in the area of entrepreneurship. Lundström and Stevenson (2001, 2005) carry out a comprehensive review of ten countries’ entrepreneurship policies. Both books have provided input to the theoretical framework utilized in this chapter. The empirical part of this chapter’s discussion has also benefited from a series of OECD reports on the differences between the European Union and the United States in the area of entrepreneurship (OECD 2003b, 2005). Grilo and Thurik (2005) and Van Steel et al. (2006) offer good insight into the determinants of US and EU entrepreneurship and explore various types of empirical analyses. 4.2 What Challenges Does the EU Face in Terms of Entrepreneurship? Entrepreneurship is not easily defined as is not a single event, but rather a process that transforms an idea into a firm. Many people leave

Promoting Entrepreneurship

93

the process before they even start a firm, and most new firms exit due to failure while others survive at, or near, the breakeven point. Only a small minority of new firms turn into high-growth firms, also known as gazelles. The chosen definition of entrepreneurship should be compatible with the macroeconomic policy objective or the policy context (Storey 2002). For example, a definition based on Schumpeter’s work is often used if the policy objective is to promote innovation and growth (Schumpeter 1949), whereas another common definition is based on Knight’s work and is often used if the policy objective is to create jobs through self-employment (Knight 1971). The policy context of this chapter is the European Commission’s objective to boost entrepreneurship (European Commission, 2003) and thus the EU challenges in this domain can be summarized as how best to increase the number of start-ups and to generate more high-growth firms. Consequently this chapter defines entrepreneurship as the entry of new firms and creation of high-growth firms. This is a definition closely linked to Schumpeter’s work on entrepreneurs as innovators. The link between growth and entrepreneurship is substantiated in several papers (Acs et al. 2005, Audretsch and Thurik 2000; Scarpetta et al. 2002; OECD 2003a; Brandt 2004a). Therefore this link is not called into question in this chapter. As the optimal level of new firm entry is unknown, the EU objective to increase start-up rates must be seen in a comparative perspective. Carree et al. (2004) show that, for example, countries having a selfemployment rate deviating from a what the authors define as a “natural” rate, given the level of economic development, suffer in terms of economic performance. They found growth to be actually reduced by both a too high and too low self-employment rate. Likewise Audretsch et al. (2002) show that deviating from an “optimal” share of small firms spread throughout the economy can lower future economic growth. Because the optimal start-up rate depends on the level of economic development, the start-up rates in the European Union and the United States are quite similar. An optimal level of high-growth firms does not exist, however, since more sustainable growth of firms is always better. The number of highgrowth firms in the European Union will be compared to that of the United States.

94

4.3

Anders N. Hoffmann

Counting the Start-Ups

Counting the number of start-ups should ideally be a simple task. However, no agreed international definition exists on what constitutes a start-up firm. In the United Kingdom, for example, value-added tax (VAT) registration is the most commonly used measure of start-up activity, whereas in the United States, it is the hiring of the first employee (Vale 2006). While most countries use VAT registration as the measure of firm creation, this does necessary ensure comparability as the VAT registration thresholds vary from zero up to £60,000 among EU countries. The number of new firms has to be normalized by some measure to allow for a cross-country comparison. Many possible denominators are available depending on purpose of the comparison (Iversen et al. 2005). This chapter focuses on generation of growth through entrepreneurship. Therefore the denominator is equal to the stock of existing firms within a given country in order to measure the dynamics within the business sector. Alternatively, the number of people in the workforce or population could be used as the denominator, although that would more measure the entrepreneurial participation and the industrial structure rather than the competitive pressure from entrepreneurship. The data used in this chapter originates from the EUROSTAT Business Demography Project, which has to some degree been successful in providing comparable data on start-up rates for several European countries (EUROSTAT, 2005). The United States has four main sources of start-up rates that produce very different start-up rates varying from around 10 to around 20 percent. Two of the sources are from the US Census Bureau, one is from US Small Business Administration (SBA), and the fourth (producing the highest start-up rates) is from the Bureau of Labor Statistics. The first three sources build to some extent on the same metadata (hiring of the first employee). Definitions of the start-up and stock (number of existing firms) differ slightly across these three, although the resulting start-up rates are somewhat comparable— between 10 and 12 percent depending on year of comparison. The last source builds on a Quarterly Census of Employment and Wages and produces very different results than the other three, which probably is due to the quarterly collection of data where short-lived firms and false starts play a larger role (Pinkston and Spletzer 2004). Even though definitions applied in the US SBA data are quite similar those of EUROSTAT, in that both use the same unit of measure and

Promoting Entrepreneurship

95

define an existing firm on an “alive during the period basis” (Vale 2006), none of the US figures are directly comparable with the EUROSTAT data. A comparison of the raw data shows that start-up rates are somewhat similar in the United States and European Union (figure 4.1). However, this comparison does have one serious methodological flaw. The US data only include employer firms, meaning businesses with at least one employee, whereas the EUROSTAT database includes all new firms. The EUROSTAT data does provide a breakdown by size class, including a category for zero-employee enterprises. To resolve the methodological issue, one could simply subtract the zero-employee category from the EUROSTAT data, which would reduce the average EU start-up rate to around 5 percent. This solution leaves out all firms that existed in period t − 1 without employees and hired their first employee in period t, as a new firm in EUROSTAT per definition did not exist in t − 1. Many countries’ business registrars allow for a more detailed analysis, and comparable start-up rates can consequently be calculated. 16% 1998 1999 2000 2001

14% 12% 10% 8% 6% 4% 2%

N or wa y

ai n Sp

ar k

n

en m D

m ea

EU

Ita ly Lu xe m bo ur g N et he rla nd s Po rtu ga l Fi nl an d Sw ed U ni en te d Ki ng do m

U

ni te d

St at es

0%

Figure 4.1 Comparison of US and EU business start-up rates. The full EUROSTAT data set includes several other countries, but only those countries for which data was available for at least three of the above years are shown in this figure. (Sources: For United States, US Small Business Administration, Firm Size Data; for European Union, EUROSTAT Business Demography.)

96

Anders N. Hoffmann

In cooperation with Statistics Denmark, it was possible to calculate comparable Danish start-up rates. The numerator equals the number of new firms entering the market in year t with at least one employee, plus the number of existing firms hiring their first employee in year t. A total of 4,324 firms started in 2003 in Denmark with at least one employee according to EUROSTAT data. A total of 7,155 firms, existing in 2002 without employees, hired their first employee in 2003.1 Consequently the population of new Danish employer firms in 2003 consists of 11,479 enterprises (i.e., 4324 + 7155). The denominator equals the stock of employer firms, which is defined as the total population of enterprises with employees at any point during 2003—a total number of 101,584 firms. The Danish start-up rate for employer firms is thus equal to 11.3 percent, which is higher compared to its 9.7 percent start-up rate according to the EUROSTAT data. This conclusion is supported by preliminary calculations of start-up rates in Finland and the United Kingdom, where the actual employer start-up rates are higher than the official EUROSTAT figures. A forthcoming paper will look to expand this analysis to estimate start-up rates for employer firms for several other EU countries (Hoffmann, Nielsen, and Vale). Based on this data, it would appear that the European Commission has misidentified its policy challenge. The Danish case show that the start-up rates in the EU might actually be higher than the US start-up rates if the focus is on employer firms. This work is still preliminary, but it does indicate that the EU countries do not have a start-up problem. The same conclusion was implicitly made in Scarpetta et al. (2002), although the authors did not correct for differences in the registration method across countries. This conclusion contrasts starkly to the results of the Global Entrepreneurship Monitor (GEM) Project (GEM 2005), which may stem to some extent from a common misinterpretation of the GEM data. GEM publicizes a so-called Total Entrepreneurial Activity (TEA) index that measures people engaged in the process of firm creation. TEA is therefore not a measure of start-ups, although it is commonly referred to as such. These limitations are recognized by members of the GEM research team. Reynolds et al. (2005) list, for example, several ways to construct GEM data that are comparable with start-up rates based on business registrations. They construct start-up rates for eight EU countries and the United States. In this constructed data, the large differences between

Promoting Entrepreneurship

97

Europe and the United States, found in the TEA index, disappear. The United States has an 11.5 percent start-up rate based on GEM data compared to a 9.7 percent average among the eight EU countries (Reynolds et al. 2005). There is also a degree of subjectivity in the GEM data as they are based on interviews about people’s attempts and successes as entrepreneurs. Overall, the presented data suggests that the Commission should change its focus on “policy measures [that] seek to boost the Union’s levels of entrepreneurship, . . . [by] producing more entrepreneurs.” The EU has start-up rates of new firms comparable to that of the United States. 4.4

Measuring the Generation of High-Growth Firms

Defining what constitutes a high-growth firm may even be more difficult than defining start-up rates. The theoretical literature offers several definitions inspired by the work of David Birch (1987, 1995). In Birch’s work, a high-growth firm has at least 20 percent growth each year over a five-year period. Other authors define high-growth firms as the 10 percent fastest growing firms in the economy (OECD 2002). This chapter defines high-growth firms as the share of firms with a growth rate (in either employment or turnover) higher than 60 percent over a three-year period (from t to t + 2) and with a growth rate of at least 20 percent each year. The requirement of a positive growth rate of at least 20 percent is based on Birch work on gazelles (Birch 1995). The requirement ensures only firms with constant growth are included and not firms that due to changes in owner structure or other external events have a very high growth in one year and then no growth in the following year. The 60 percent threshold is commonly used, but is not based on any hard evidence. Turnover and employment are both included due to the differences in growth patterns across sectors. Knowledge-intensive manufacturing firms grow both in employment and turnover, whereas service sector firms mainly have high growth in employment (Delmar et al. 2003). The time period is shorter than the original Birch work, as the sample size reduces dramatically in some countries if firms’ performance has to be tracked over many years as compared to three years. The data for calculating the indicators on the share of high-growth firms in a given country is taken from the Bureau Van Dijk (BvD), an

98

Anders N. Hoffmann

electronic publishing firm and documented in Hoffmann and Junge (2006). BvD specializes in cleaning and organising data supplied by national information providers (e.g., Companies House in the United Kingdom, INPI in France, National Bank in Belgium). This database has been used by several other researchers (e.g., see Desai, Gompers, and Lerner 2003). In principal, the database covers all firms. However, Hoffmann and Junge (2006) show that firms with fewer than 15 employees are underrepresented in the data. The sample used in this paper is consequently restricted to firms that belong in the size class of 15 to 200 employees. While this threshold does limit the number of firms in the sample, it also has some advantages. For example, many European countries have labor market regulation that take effect at 15 employees, thereby creating problems in comparing growth in smaller firms. The data clearly shows that EU countries lag behind the United States in producing high-growth firms (figure 4.2). Only Ireland produces similar high-growth rates to those of the United States. While the focus here is on young firms (firms less than five years old in period t), the conclusion holds constant when including all firms. 30% 25%

Turnover

Employment

20% 15% 10% 5%

Ita he ly rla nd s N or wa y Sp ai Sw n U ni ed te d en Ki ng d EU om av er U ag ni te e d St at es et

N

Au st ri Be a lg iu m D en m ar k Fi nl an d Fr an c e G er m an y G re ec e Ire la nd

0%

Figure 4.2 Share of high-growth firms among all new young firms. By Birch’s definition, the data show that the share of gazelles in this database is 3.2 percent for the time period 1999 to 2003, which is very similar to Birch’s 3 percent (1995). (Source: Own calculations based on Hoffmann and Junge, 2006.)

Promoting Entrepreneurship

99

An extensive sensitivity analysis in Hoffmann and Junge shows that the ranking of countries stands up to changes in the 60 percent threshold. The ranking is also robust to corrections for differences in the industry structure across countries. Alternative definitions of high growth and different assumptions about the age range of a young firm have also been tested, but again, with little impact on the ranking of countries or the large difference between the EU and US averages. The only real concern is the representativity of the data. The US rate is based on a very small sample, although tests on larger samples but shorter time periods do suggest that the large difference between the EU and the US growth rates is a reflection of the reality. Overall, the presented data suggests that the Commission has adopted the correct approach when stating that “Policy measures should seek to boost the Union’s levels of entrepreneurship . . . for getting more firms to grow” (EU 2002). This conclusion is more preliminary than the conclusion related to start-up rates because the quality of the underlying data does not meet the normal criteria for national statistical offices, used in the previous section. Nonetheless, the richness of the database does allow for corrections for firm age, size class, business cycle, listed firms, consolidation code, and industry structure, which would not have been possible using official statistics as access to the underlying data is restricted. 4.5 What Are the Key Policy Areas for Stimulating High-Growth Firms? The previous sections clearly show that Europe should focus on getting more firms to grow and not on stimulating more firm creation. The key question is therefore: Which policy areas should the EU include in its policy reform in order to encourage more firms to grow? Culture and religious orientation have been emphasized as the main reasons for cross-country differences; others emphasize simply market size.2 However, as Baumol notes, “The most important policy implication is that the stimulation of productive entrepreneurship is a much more straightforward and feasible undertaking than previously recognized” (Kauffman 2005: 23). Another possible explanation of the large differences between the United States and the European Union could be that governance and ownership structure differ substantially. Many European countries are dominated by family-owned and family-managed businesses, which

100

Anders N. Hoffmann

are less likely to grow than independently owned and managed firms (James 2006). The database allows for some tests of this explanation. The BvD publishes an indicator of independence for each firm. This indicator let firm’s management be defined by how independent they are from owners. A firm with many “small” shareholders are classified as independent. Unfortunately, this variable is not available for all countries, but calculation on the German data showed no difference in proportion of high-growth firms between independent firms and controlled firms. The proportion of high-growth firms in independent firms was 2.94 (4.54) for employees (turnover) and 2.84 (4.65) for firms with one large shareholder. Similar results are obtained for other countries with available data. Consequently differences in culture, region, and ownership structure might not be the key reason for the large differences between the United States and the European Union in the share of high-growth firms. Differences in policies might be explaining the large differences in entrepreneurship performance. A three-step comparative methodology is used in this chapter to identify key policy areas driving entrepreneurship performance. First, the entrepreneurial business environment is defined and quantified. Second, the empirical links between the indicators measuring the business environment and the indicators measuring high growth are tested. Third, if the correlation between business environment and highgrowth firms is significantly positive, then key policy areas for enhancing entrepreneurship performance will be identified based on regression and comparative techniques. 4.6 Defining and Quantifying the Entrepreneurial Business Environment The number of new high-growth firms created each year depends on a myriad of underlying factors coupled with the personal attributes of entrepreneurs. No single paradigm or framework exists (Aldrich 2000). This section builds on the eclectic theory developed by Audretsch et al. (2002) and the policy framework developed by Lundström and Stevenson (2002, 2005). Hoffmann (2006) presents the full theoretical framework. This section summarizes the main parts of the framework. Many words and definitions are used in the literature to describe the factors affecting entrepreneurship. The differences between various

Promoting Entrepreneurship

101

studies are often semantic; the essence of the various papers is that a growth-oriented firm is created by a combination of three factors: opportunities, skilled people and capital. Opportunities are the ideas that create genuine value in the minds of other people, and they are essential for starting and growing businesses (European Commission 2002; Davidson 1989). Skills not only entail basic industry knowledge required to succeed in a competitive environment, but also the ability to seize entrepreneurial opportunities (Reynolds, Hay, and Camp 1999; Gavron et al. 1998). Skills include the competencies of the entrepreneur and also access to other competencies within the entrepreneurial infrastructure (Lee et. Al. 2000). Capital is a necessity for firm expansion and growth. Most studies on entrepreneurship highlight capital as one of the most critical factors for success (EU 2003). Capital covers all phases of business life, from access to early seed funds to access to the stock markets. A combination of opportunity, ability, and capital does not necessarily lead to entrepreneurship if costs, such as opportunity cost (e.g., forgone salary and loss of health insurance) and start-up cost, outweigh potential benefits. In this event the opportunity should not be pursued following the rationale of basic economic theory. These incentives reflect the classic market-clearing condition that marginal cost must equal marginal benefit in equilibrium. The incentive structure component in the model represents the various incentives and disincentives that impact the cost-benefit balance of the opportunity. A final component in the model is motivation. Previous work shows that the willingness to pursue entrepreneurial activities relies only partly on the economic factors described above (Davidson 1989). Personal motivation plays a decisive role as it is unique and involves a complex combination of factors, such as personal traits, risk aversion and sociological circumstances determined by the national culture. This model’s understanding of motivation is based on cognitive theory, which has its roots in psychology (Wood and Bandura 1989). Entrepreneurship is also affected by basic macroeconomic conditions. High unemployment, for instance, will increase the share of individuals motivated to become entrepreneurs as a result of job loss. Despite their obvious importance for entrepreneurship, these conditions are excluded in the policy framework as the focus is on differences between the European Union and the United States. The differences in the macroeconomic conditions are assumed to be a less important determinate of entrepreneurship than the differences in the micro-

102

Anders N. Hoffmann

economic structures defined above. The framework is therefore constructed for what is labelled “opportunity-based entrepreneurs” by the GEM project, and not for “necessity driven” entrepreneurs (GEM 2001). Each of the five factors (skills, opportunities, capital, incentives, and motivation) is affected by a series of policy areas. This section focuses on 61 different indicators for the 24 policy areas, which are organized in relation to the factor they affect most based on the author’s qualitative judgement in order to communicate the results in an easily comprehensible manner (figure 4.3). The organization of factors does not play any role in the analytical results as each policy area is analyzed independently of the other areas. A full description of the policy areas is given in appendix B. The 24 areas in the model should cover all policies affecting entrepreneurship, implying that any policy aimed at affecting growth by stimulating entrepreneurship should belong to one or more of the policy areas. While various aggregation and disaggregation of the 24 policy areas can decrease or increase the number of policy areas. This list forms the basis of the Danish entrepreneurship policy (EBST 2005). Not all policy areas can be quantified, but a quality assessment of available indicators has highlighted 61 indicators, which can be used to quantify 18 of the policy areas (appendix B). A quality evaluation of the data suggests that most of the policy areas are measured by accurate indictors, with the exception of taxation where not all aspects were adequately covered (Hoffmann et al. 2005).3 Additional analyses are needed in the area of entrepreneurship education before policies can be suggested. Thematic studies confirm the results of this analysis and show large differences in the emphasis put on entrepreneurship teaching and attitudes in EU and US universities (EBST 2005). The construction of composite indicators is almost an art form in itself and highly controversial. This section uses the methods suggested in the Handbook on Constructing Composite Indicators: Methodology and User Guide published by the OECD and the Joint Research Centre of the European Commission in Ispra (Giovannini et al. 2005). The handbook points to four main problems in constructing composite indicators: (1) selection of indicators, (2) treatment of missing values, (3) normalizations; and (4) weighting—with weighting as the most important problem. To respond to these concerns, the selection and evaluation of the individual 24 indicators are done in Hoffmann et al. (2005). No imputation of missing values is attempted, which implicitly assigns all missing values for a country with a value equal to the

Capital taxes

Procurement regulation

Stock markets

Venture capital

Private demand factors

Wealth and bequest tax

Access to foreign markets

Business angels

Loans

Entry barriers/ deregulation

Technology transfer

Capital

Opportunities

Labor market regulation

Entrepreneurship infrastructure (private)

Bankruptcy legislation

Administrative burdens

Entrepreneurship infrastructure (public)

Initiatives towards specific groups

Business tax and fiscal incentive

Communication about heroes

Entrepreneurial motivation

Personal income tax

Social security discrimination

Motivation/culture

Incentives

Restart possibilities

Entrepreneurship education

Traditional business education

Ability

Figure 4.3 Overview of the main policy areas at the micro level. The darkest policy areas are those that cannot be quantified.

Policy areas affecting entrepreneurial performance

Factors affecting entrepreneurial performance

Total measure of the business environment for entrepreneurship

Promoting Entrepreneurship 103

104

Anders N. Hoffmann

average of all other available indicators for that particular country. Several techniques can be used to normalize indicators, including the standard deviation from the mean, the distance from the mean (where mean = 100), the distance from the best performer (leader = 100) and the distance from the best and the worst performing country (the “minimum–maximum method”). For this study the “minimum– maximum method” has been selected. A sensitivity analysis shows that the ranking of countries is robust to other methods of normalization. No direct solution exists to the selection of weights. Therefore this section will not focus on composite indicators but rather on composite distributions using a new sensitivity technique where weights are assigned randomly to each of the normalized indicators. The calculation is repeated 10,000 times and the weights are drawn from a uniform distribution (from 0 to 1) for each of the indicators. This calculation gives a distribution of possible values for each country in each policy area. This sensitivity technique also addresses to some extent the problem of the selection of indicators. As the randomly assigned weights vary between 0 and 1 for each indicator, the technique also tests indirectly for the robustness of possibly excluding an indicator. 4.7 Linking the Business Environment to the Creation of HighGrowth Firms A simple correlation plot clearly indicates the links between the EU objective to generate more high-growth firms and the quantified business environment. This plot shows a high correlation between business environment indicators and the performance indicator (figure 4.4). About 55 percent of the differences in performance among the countries can be explained by the differences in the business environment. While the correlation is not necessarily a sign of causality, the figure does suggest that most of the policy areas that determine performance have been included in the analysis. The correlation depends critically on the inclusion of the US and Korea because the other countries fall in two groups. The first group of northern European countries seems to require a better business environment for its entrepreneurial performance when compared to the second group of southern European countries. The difference could perhaps be attributed to some of the macroeconomic conditions that are not included in the model or some cultural aspects that are not captured by the indicators.

Promoting Entrepreneurship

105

100 Korea

United States

80

60 United Kingdom Spain

40

Italy Switzerland Belgium

20

France Denmark

Greece

Sweden Finland Norway Germany

Austria

Netherlands

0 0

20

40

60

80

100 R 2 = 0.58

Figure 4.4 Links between performance and business environment. The performance indicator is a simple average of the normalized share of high-growth firms in both turnover and employment in 2001 to 2003. The business environment indicator is the simple average of the normalized values of each of the 18 policy areas that can be quantified.

However, the assumption of equal weights is the main reason for the split into two groups. The southern European countries perform weakly in certain areas, which lowers their average business environment, but they also perform comparably well in some areas to the best countries. For example, Spain has a major weakness in their capital taxes, whereas their bankruptcy legislation is relatively entrepreneurship friendly. The weakness in southern Europe could be in areas less important for performance. The correlation is quite robust to changes in the weights for the various policy areas. A Monte Carlo simulation, where the weights are drawn from a uniform distribution (0–1) shows that the correlations are within a 95 percent confidential interval of the average correlation, which is always significant when different from zero. Thus the weights play a very limited role in determining this correlation. 4.8

Comparing EU and US Business Environments

Many differences exist between the EU and the US business environments. The quantification of the various policy areas allows for a

106

Anders N. Hoffmann

Tech transfer regulation 100

Entrepreneurial motivation Labor market regulation

Entry barriers Access to foreign markets

75

Administrative burdens

50

Loans

25 Bankruptcy legislation

Venture capital 0

Business tax and fiscal incentive

Exit markets

Wealth and bequest tax

Personal income tax

Entrepreneurship infrastructure

Capital taxes

Traditional business education

Restart possibilities

Entrepreneurship education

USA

EU

Figure 4.5 Comparative analysis of the business environment in the European Union and the United States. The EU average covers the EU-15 (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Luxembourg, Portugal, Spain, Sweden, and United Kingdom). (Source: Author’s calculations.)

comparative study that can be summarized in a radar chart (figure 4.5). Compared to the United States, in the European Union five areas— restart possibilities, entrepreneurship education, traditional business education, labor market regulation, and entrepreneurship motivation—are much less conducive to entrepreneurship. In other areas the European Union performs at a par with the United States. For example, the European Union has better access to foreign markets than the United States, capital taxes are quite low in some EU countries and the EU bankruptcy systems are on average inexpensive and efficient. 4.9

The Relative Importance of the Various Policy Areas

While a comparative study provides some input as to what policy areas should be included into the EU reforms, it does not measure the relative

Promoting Entrepreneurship

107

importance of each policy area to promoting entrepreneurship. Such information is needed in order to move beyond simple benchmarking methods and provide policy makers with clear policy priorities. Various fallible ways exist for determining the relative importance of each policy area. Preferably, a multiple-regression based on a panel over several years and countries would be used, but the lack of country coverage and time series for many of the indicators make this difficult. Furthermore the inclusion of all policy areas into a single equation does not provide much insight due to a high degree of multicollinearity among the various policy areas. The simplest method is based on correlation coefficients between the 61 business environment indicators and indicators for high-growth firms. This method uses the raw data and no composite indicators are included in order to keep it simple. A significant correlation suggest that the given business environment indicator is important for growth. Seventeen indicators are significant correlated with the indicators for high growth (appendix A). The significant indicators are related to six policy areas: entrepreneurial motivation, entrepreneurship education, labor market regulation, restart possibilities, entrepreneurial motivation, business tax and fiscal incentive, venture capital, and rersonal income taxes. These six policy areas might therefore be more important for growth than the others. However, this method leaves out a lot available information. The correlation only uses one year of business environment indicators and the richness of the Bureau Van Dijk database is not explored. Therefore more advance regressions techniques are applied below. The regressions are done in two steps. First, the data on high-growth firms is “filtered.” The process controls for possible biases in the database on high-growth firms by filtering out factors that are known to have an impact on growth like sector, age, and size but are unrelated to policy in the short term (e.g., see Evans 1987 or Smallbone et al. 1995). Second, the residuals from the first regression are averaged across all firms within a given country and a given year. This produces two panels, one for the employment data and one for the turnover data, over seventeen countries and three time periods. These two panels are then used to test the importance of the various policy areas. The following “filter” equations (one for employees and one for turnover) are estimated for employees and for turnover using the full database of business accounts (14 million firms):

108

Anders N. Hoffmann

K

dijt = α + ∑ δ k ditk + ε ijt . k =1

The left-hand side is a d ummy variable (equals 1 if firm i in country j and year t is a high-growth firm and zero elsewhere). On the righthand side, k refers to the control variables. The control variables are as follows: 5 age groups (1, 2, 3, 4, and 5 years old), 5 size classes (15–19 employees, 20–49 employees, 50–99 employees, 100–199 employees, and 200+ employees), 17 industry groups (NACE definition), a dummy variable for whether profit and loss accounts are reported consolidated or unconsolidated,4 and a output gap variable5 reflecting differences in business cycles across countries. In the regression for turnover, all the age groups were significant at the 5 percent level. Three of the size classes (15–19, 20–49, and 200+) were significant. Nine of the NACE groups were significant. The output gap and the dummy for consolidated accounts were also significant. Interactions between the various dummy variables were analyzed, but the results were not included into the final regression as only a few of them were significant. Similar results were found in the regressions for employment, although slightly fewer dummy variables were significant. The average of the residuals for all firms within a given year and country was calculated based on the results from the “filter” regressions (figure 4.6). These average residuals represent the difference among the countries with respect to the EU challenge related to the generation of high-growth firms when all of the nonpolicy relevant factors are “filtered out.” These residuals correlated highly (0.9–0.93) with the share of high-growth firms (figure 4.2). This suggests that the different country samples in the database on high-growth firms are unbiased with respect to sector, age and size composition. A few of the European countries do, however, change place in country rankings depending on whether the ranking is based on the residuals or the simple share of high-growth firms. The residuals are then used as the dependent variable in the next regressions. The policy areas are included individually in this second set of regressions as the independent variable (one regression for turnover and one for employment). Data for the policy areas are only available for two time periods, so only two time periods of residuals are included at a time. Large differences exist among policy areas with regard to the time it takes for a change in the policy area to lead to a change in per-

Promoting Entrepreneurship

109

0.2 1999 2000 2001

0.15

0.1

0.05

0

–0.05

nd s N or wa y Sp ai n Sw ed en Sw itz U e r la ni te nd d Ki ng do U ni m te d St at es

r la

Ko re a

he N

et

an y m

ce

er G

nd

Fr an

k ar

la Fi n

m

m

iu

en D

lg Be

Au s

tri

a

–0.1

Figure 4.6 Average (turnover) residual across all firms within a given year. (Source: Author’s calculations.)

formance. The policy areas are consequently regressed on both the first two time periods of 1999 and 2000 and the last two time periods of 2000 and 2001. Ten of 18 policy areas that could be quantified are significantly correlated with some of the residuals. Five areas (venture capital, restart possibilities, personal income tax, and bankruptcy legislation) are always significantly correlated with the residuals, regardless of the time period and the definition of high growth (table 4.1). Two areas (entrepreneurship education and labor markets) are not significantly correlated with high growth in employment in one time period. Two areas (capital taxes and entrepreneurship motivation) were only significantly correlated with growth in turnover and not in employment. One area (business tax and fiscal incentives) was significantly correlated with growth in employment in older firms but not in younger firms (less than five years old). All the policy areas found to be important in the simple calculation in the beginning of this section are among the significant areas. A significant correlation between a given policy area and the residuals suggest that this area is important for the generation of

110

Anders N. Hoffmann

Table 4.1 Summarizing the regression results Significantly correlated with turnover residuals

Significantly correlated with employment residuals

Venture capital

X

X

Exit markets

X*

X

Capital taxes

X

Restart possibilities

X

X

Entrepreneurship education

X

X*

Personal income tax

X

X

Business tax and fiscal incentive

X**

Bankruptcy legislation

X

X

Labor market regulation

X

X*

Entrepreneurial motivation

X

Note: X* not significantly correlated in one of the time periods, but significantly correlated in at the 5 percent level in a two sided test. X** only significantly correlated if all firm age groups are included in the regression, but not if only young firms are included.

high-growth firms. However, these correlations could be spurious. Some significant policy areas may not be directly important for high growth, and similarly policy areas that are insignificantly correlated with the residuals may become significant if other areas are included into the analysis. Multivariant regressions were carried out to include more policy areas but proved to add little value to the analysis due to the problems of multicollinearity and the small size of the panel. For example, entrepreneurial education, personal income tax, and venture capital were all significant, but when the sequence of eliminating insignificant variables was changed slightly, venture capital and restart possibilities also became the significant variables. The multivariant regression also showed that motivation and income tax were potentially significant policy areas together with venture capital. Fixed-effects models were also tested with little success. The policy areas were entered individually, together with country dummies. These regressions proved difficult to do as changes in the business environment are slow to take effect and not all countries change every aspect of their business environment every year. Consequently the use of country dummy variables in this analysis was impossible.

Promoting Entrepreneurship

111

4.10 What Are the Main Policy Challenges for the European Union? The comparative analysis of the business environment in the European Union and the United States showed major differences in some policy areas. The analysis of the relative importance of the various policy areas identified ten policy areas that were shown to be significant for entrepreneurship performance. When combined, these two analyses provides an overview of the main policy challenges for the European Union (table 4.2). When analyzing what policy areas the European Union should focus on, a good starting point may be the policy areas where there is a significant correlation and, at the same time, a large difference between the European Union and the United States (i.e., the shaded upper right corner in table 4.2). However, the relative cost of policy actions should also be taken into consideration before constructing policy reforms. Restart possibilities and labor market regulation are financially affordable policies as they only involve the redesigning of government regulation, yet these policies can be very difficult to implement. Venture capital and exit markets are private market institutions so government does not have a permanent role to play. Governments can stimulate, through timely, well-designed policies, the creation of a private market at a low cost (OECD 2004). Entrepreneurship education at university level Table 4.2 Summarizing EU policy challenges Not significantly correlated

Significantly correlated

Considerable EU and US differences

Tech transfer regulation Traditional business education Administrative burdens

Venture capital Exit markets Restart possibilities Entrepreneurship education Personal income tax Business tax and fiscal incentive Labor market regulation Entrepreneurial motivation

Insignificant EU and US differences

Entry barriers Access to foreign markets Loans Wealth and bequest tax Entrepreneurship Infrastructure

Capital taxes Bankruptcy legislation

Note: A large difference is defined as larger than the average difference.

112

Anders N. Hoffmann

is more expensive. The estimated cost of a US initiative implementing entrepreneurship programs across all disciplines in eight US universities6 was $100 million (Kauffman 2006). Taxation is also highly correlated with high growth, yet changes are often expensive. For example, changes in income taxes are expensive because they benefit everybody in the workforce and not only entrepreneurs. Similarly changes in business taxation and fiscal incentives are expensive. Much of the literature concludes that taxation can have a negative impact on entrepreneurial activity (especially progressive tax systems; Gentry and Hubbard 2000). Other authors argue that there is little empirical evidence linking taxation and entrepreneurship (Parker 2003). Changing the income taxes will probably have much larger effect on the supply of labor, rather than on entrepreneurship. Any tax reform should consequently be analyzed within the context of the general framework of the labor market and based on a more indepth analysis of the cost and benefits of the proposed tax changes. Given the significant policy areas identified for promoting growthoriented entrepreneurship and the cost factor, the EU should aim at improving: •

Restart possibilities

•

Labor market regulation

•

Venture capital

•

Exit markets

•

Entrepreneurship education

Entrepreneurial motivation is also mentioned in most EU papers as an important policy area to boost entrepreneurship. However, the policy instruments to increase entrepreneurial motivation are not clear. In some respects motivation can be seen as endogenous to this model. Motivation is influenced by the success of others and the benefits to one’s self (Wood and Bandura 1989). Improvement in the general business framework can be argued to produce more successful entrepreneurs, which will in turn increase motivation among potential entrepreneurs. For example, the attitude in the United States toward entrepreneurship was negative a half century ago (Acs 2005). During the 1970s and 1980s independent changes in the US economy produced a more entrepreneurial friendly economy, from which several very successful entrepreneurs benefited (Schramm 2004). Today an entrepreneur is regarded positively by American society as the “self-made

Promoting Entrepreneurship

113

man,” and indeed entrepreneurship consequently flourishes in the United States (Hart 2003). Based on this example, the European Union should focus on improving its performance in the other policy areas to help stimulate motivation among potential European entrepreneurs. While this analysis is done at the EU level, it needs to be done at the country level in order to have a real impact. The data and the setup allow easily for this possibility. The country-level analysis will identify challenges both in macro- and in microeconomic structures. Improvements are needed in all areas and not only focus on improvements in one of them. For example, Denmark has a flexible labor market, but still lacks growth in new firms due to a poor performance in the policy areas of restart possibilities and entrepreneurship education at university level, both of which lag significantly behind the United States. A note of caution should also be added. Each European country will have to figure out how to improve its business environment in each policy area based on the unique functioning of its economy. Countries can draw lessons and inspiration from the top-performing countries, but the initiatives have to be tailored made to the national context. This conclusion has been clearly demonstrated in the Danish Policy Report (EBST 2005). 4.11

Conclusion

In the beginning of this paper, it was stated that the European Union is perceived to have an unexploited entrepreneurial potential by failing to encourage enough people to become entrepreneurs and by failing to encourage the limited number of existing entrepreneurs to grow their firms. This chapter concludes that this perception may be inaccurate. The new data presented in this chapter show that the EU countries have start-up rates that are comparable to the start-up rates in the United States. However, that data confirm that there is a lack of highgrowth firms in the European Union. The share of high-growth firms is significantly higher in the United States than in any EU country. Some variation does exist among the European countries, with the United Kingdom and Ireland as the best performing countries. The policy implication emerging from this chapter’s results is clear. The European Union should focus on getting more firms to grow by improving its entrepreneurial business environment for high-growth firms and not on stimulating more entry. Furthermore the analysis suggests that the European Union can stimulate growth by focusing on

114

Anders N. Hoffmann

improving restart possibilities, labor market regulation, access to venture capital, exit markets, and entrepreneurship education at the university level. These critical policy areas are identified through a quantification of all factors of the business environment that affect entrepreneurship and extensive regression analyses. Taxation was also found to play an important role for entrepreneurship, but the costs of changes in the tax system are high. Changes in European tax systems may be needed, but these changes should be based on cost-benefit analysis and be seen in a much broader policy context. The policy areas identified in this chapter are very different from the ones suggested by the European Council, which focused on changing entrepreneurial culture through education. This chapter speculates that culture and motivation might be endogenous and therefore change as a result of the other policy improvements. Regardless, changing culture and motivation will clearly not be enough to boost entrepreneurship in order to meet the EU challenge. More comprehensive reforms are needed to get European firms to grow. While the policy priorities suggested in this chapter provide a good starting point, future work should focus on using more advanced econometric techniques to identify the critical policy areas as more time series become available for the underlying data. More work on the pre-entry of entrepreneurs is also needed. The United States and the European Union might have the same share of nascent entrepreneurs, but the quality of these entrepreneurs may differ. For example, an American entrepreneur may have more pre-start experience than the European counterpart, as the pre-entry stage as hobby entrepreneurship may be easier in the United States. How to ensure coherency in the policy design at the national level is also an important area for further analysis. Appendix A: Data Description This appendix presents a recapitulative table giving a broad overview of the indicators used to define the framework conditions of the business environment. The indicators marked by a * is significant corralled with the performance indicators.

Opportunities

Factors

Share of new enterprises with exports* FORA Access to capital markets IMD, World Competitiveness Yearbook Export credits and insurance IMD, World Competitiveness Yearbook

Access to foreign markets

Barriers to competition—OECD index OECD,Summary Indicators of Product Market Regulation with an Extension to Employment Protection Legislation, p. 25 or 75 http://www.olis.oecd.org/olis/1999doc.nsf/c16431e1b3f24c0ac12569fa005d1d99/5ef586bbe13dd52ac125684a003a8da0/$FILE/00075836.PDF Public ownership—OECD index OECD,Summary Indicators of Product Market Regulation with an Extension to Employment Protection Legislation, p. 25 or 74 http://www.olis.oecd.org/olis/1999doc.nsf/c16431e1b3f24c0ac12569fa005d1d99/5ef586bbe13dd52ac125684a003a8da0/$FILE/00075836.PDF Public involvement in business operation OECD,Summary Indicators of Product Market Regulation with an Extension to Employment Protection Legislation, p. 25 or 74 http://www.olis.oecd.org/olis/1999doc.nsf/c16431e1b3f24c0ac12569fa005d1d99/5ef586bbe13dd52ac125684a003a8da0/$FILE/00075836.PDF

Entry barriers

University/industry research collaboration WEF, Global Competitiveness Report Technological cooperation IMD, World Competitiveness Yearbook

Tech-transfer regulation

Policy area

Indicator source and Internet link

Table 4.A1 Indicators of framework conditions

Promoting Entrepreneurship 115

Capital

Factors

Indicator source and Internet link

Venture capital—early stage OECD, Science, Technology and Industry; Venture Capital: Trends and Policy Recommendations, p. 7 http://www.oecd.org/dataoecd/4/11/28881195.pdf Venture capital—expansion stage* OECD, Science, Technology and Industry; Venture Capital: Trends and Policy Recommendations, p. 7 http://www.oecd.org/dataoecd/4/11/28881195.pdf

Venture capital

Extent of guarantees EU Commission. p. 38 http://europa.eu.int/comm/enterprise/enterprise_policy/analysis/doc/smes_observatory_2003_report2_en.pdf Private credit World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/ Interest rate spread World Bank, Doing Business Cost to create collateral World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/ Legal rights index World Bank, Doing Business Country credit rating IMD, World Competitiveness Yearbook

Loans

Policy area

Table 4.A1 (continued)

116 Anders N. Hoffmann

Revenue from bequest tax OECD 2003, Revenue Statistic Revenue from net wealth tax OECD 2003, Revenue Statistic Top marginal bequest tax rate OECD, Directorate for Science, Technology and Industry; Taxation, SMEs and Entrepreneurship http://www.olis.oecd.org/olis/2002doc.nsf/43bb6130e5e86e5fc12569fa005d004c/2137ebc4eaa738a5c1256c10004e37ec/$FILE/JT00130282.PDF

Wealth and bequest tax

Capitalization of secondary stock markets* OECD, Science, Technology and Industry; Venture Capital: Trends and Policy Recommendations, p. 25 http://www.oecd.org/dataoecd/4/11/28881195.pdf Market capitalization of newly listed companies relative to GDP World Federation of Exchanges, Annual Report and Statistics 2004 http://www.world-exchanges.org/publications/WFE%202004%20Annual%20Report%20and%20Statistics.pdf Capitalization of primary stock market World Bank http://www.worldbank.org/research/projects/finstructure/structure_database.xls Turnover in primary stock market World Bank http://www.worldbank.org/research/projects/finstructure/structure_database.xls Buyouts OECD, Science, Technology and Industry; Venture Capital: Trends and Policy Recommendations, p. 7 http://www.oecd.org/dataoecd/4/11/28881195.pdf

Exit

Promoting Entrepreneurship 117

Capital

Factors

Indicator source and Internet link

Taxation of dividends—top marginal tax rate OECD, Directorate for Science, Technology and Industry: Taxation, SMEs and Entrepreneurship http://www.olis.oecd.org/olis/2002doc.nsf/43bb6130e5e86e5fc12569fa005d004c/2137ebc4eaa738a5c1256c10004e37ec/$FILE/JT00130282.PDF Taxation of dividends—top marginal tax rate for the self-employed* OECD, Directorate for Science, Technology and Industry; Industry issues: Taxation, SMEs and Entrepreneurship http://www.olis.oecd.org/olis/2002doc.nsf/43bb6130e5e86e5fc12569fa005d004c/2137ebc4eaa738a5c1256c10004e37ec/$FILE/JT00130282.PDF Taxation of stock options Eurostat, Competitiveness and Benchmarking Enterprise Policy Results from the 2002 Scoreboard http://europa.eu.int/comm/enterprise/enterprise_policy/better_environment/doc/enterprise_policy_scoreboard_2002_en.pdf Taxation of capital gains on shares—short term OECD, Directorate for Science, Technology and Industry; Taxation, SMEs and Entrepreneurship http://www.olis.oecd.org/olis/2002doc.nsf/43bb6130e5e86e5fc12569fa005d004c/2137ebc4eaa738a5c1256c10004e37ec/$FILE/JT00130282.PDF Taxation of capital gains on shares—long term OECD, Directorate for Science, Technology and Industry: Taxation, SMEs and Entrepreneurship http://www.olis.oecd.org/olis/2002doc.nsf/43bb6130e5e86e5fc12569fa005d004c/2137ebc4eaa738a5c1256c10004e37ec/$FILE/JT00130282.PDF

Capital taxes

Policy area

Table 4.A1 (continued)

118 Anders N. Hoffmann

Government programs Global Entrepreneurship Monitor http://www.gemconsortium.org/

Entrepreneurship infrastructure

Quality of management schools WEF Global Competitiveness Report —

Traditional business education

Entrepreneurship education at primary education* Global Entrepreneurship Monitor http://www.gemconsortium.org/ Entrepreneurship education at higher education* Global Entrepreneurship Monitor http://www.gemconsortium.org/

Entrepreneurship education

Length of time for creditor claims on a bankrupt firm’s assets* OECD, Science, Technology and Industry Outlook; Drivers of Growth: Information; Technology, Innovation and Entrepreneurship http://www1.oecd.org/publications/e-book/9201131e.pdf

Restart possibilities

Promoting Entrepreneurship 119

Abilities

Incentives

Factors

Indicator source and Internet link

Actual cost to close a business World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/ClosingBusiness/CompareAll.aspx Actual time to close a business World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/ClosingBusiness/CompareAll.aspx

Bankruptcy legislation

SME tax rates* OECD, directorate for science, technology, and industry; industry issues taxation; SMEs and entrepreneurship. http://www.olis.oecd.org/olis/2002doc.nsf/43bb6130e5e86e5fc12569fa005d004c/2137ebc4eaa738a5c1256c10004e37ec/$FILE/JT00130282.PDF Taxation of corporate income revenue OECD 2003, Revenue Statistic 1965–2002

Business tax and fiscal incentive

Highest marginal income tax plus social contributions* OECD, Taxing Wages 2011–2002 http://emlab.berkeley.edu/users/webfac/saez/e230b_s04/OECD01_02taxingwages.pdf Average income tax plus social contributions* OECD, Taxing Wages 2011–2002. http://emlab.berkeley.edu/users/webfac/saez/e230b_s04/OECD01_02taxingwages.pdf

Personal income tax

Policy area

Table 4.A1 (continued)

120 Anders N. Hoffmann

Starting a business number of procedures World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/StartingBusiness/CompareAll.aspx Starting a business number of days World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/StartingBusiness/CompareAll.aspx Starting a business cost World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/StartingBusiness/CompareAll.aspx Regulatory and administrative opacity OECD,Summary Indicators of Product Market Regulation with an Extension to Employment Protection Legislation, p. 25 or 75 http://www.olis.oecd.org/olis/1999doc.nsf/c16431e1b3f24c0ac12569fa005d1d99/5ef586bbe13dd52ac125684a003a8da0/$FILE/00075836.PDF Enforcing contracts—number of procedures World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/EnforcingContracts/CompareAll.aspx Enforcing contracts time World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/EnforcingContracts/CompareAll.aspx Enforcing contracts procedure complexity World Bank, Doing Business http://www.doingbusiness.org/ExploreTopics/EnforcingContracts/CompareAll.aspx Starting a business—minimum of capital required World Bank, Doing Business Enforcing contracts—cost (percentage of debts) The World Bank

Administrative burdens

Promoting Entrepreneurship 121

Incentives

Factors

Indicator source and Internet link

Flexibility of hiring World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/HiringFiringWorkers/CompareAll.aspx Flexibility of firing* World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/HiringFiringWorkers/CompareAll.aspx Rigidity of hours* World Bank, Doing Business http://rru.worldbank.org/DoingBusiness/ExploreTopics/HiringFiringWorkers/CompareAll.aspx Number of administrative procedures when recruiting first employee European Union Number of administrative procedures when recruiting additional employee European Union Firing costs* World Bank, Doing Business

Labor market regulation

Policy area

Table 4.A1 (continued)

122 Anders N. Hoffmann

Cultural and social norms* Global Entrepreneurship Monitor Schøtt, Thomas (2005b). Iværksætterkulturen i Danmark og andre lande http://www.sam.sdu.dk/~tsc/CESFOkultur1.doc. Entrepreneurial motivation Global Entrepreneurship Monitor Schøtt, Thomas (2005b). Iværksætterkulturen i Danmark og andre lande http://www.sam.sdu.dk/~tsc/CESFOkultur1.doc. Self-employment preference* Eurobarometer http://europa.eu.int/comm/public_opinion/flash/fl160_en.pdf Wish to own a business Eurobarometer http://europa.eu.int/comm/public_opinion/flash/fl160_en.pdf Desirability of becoming self-employed* Eurobarometer http://europa.eu.int/comm/public_opinion/flash/fl160_en.pdf Attitude toward risk* Eurobarometer http://europa.eu.int/comm/public_opinion/flash/fl160_en.pdf

Motivation

Promoting Entrepreneurship 123

Motivation

124

Anders N. Hoffmann

Appendix B: Quality of the Data The quality of the chapter’s conclusions depends critically on the quality of the underlying data measuring the policy areas. All of the indicators used to quantify the policy areas are from Hoffmann et al. (2005), which includes a quality assessment of each indicator evaluated on the basis three quality dimensions: relevance, accuracy, and availability. These dimensions are taken from the OECD’s Quality Manual for Data Collection (OECD 2003). The quality assessment can be summarized by dividing the significantly correlated areas into four groups. The first group of policy areas (venture capital, labor market regulation, and exit markets) is measured by accurate and relevant indicators. Venture capital is measured by actual venture capital investment as share of GDP in a given year. Labor market regulation is measured by OECD quantifications of difficulties in hiring and firing employees. Exit markets are measured by capitalization of both primary and secondary stock market and by turnover in primary markets. The second group of policy areas (capital taxes, personal income tax, business tax, and fiscal incentive) are measured by accurate indictors, but not all the relevant aspects of the policy area are captured. The tax policy areas are based on comparisons of actual tax rates. These rates only cover part of the taxation system. Large differences exist, for example, in exceptions and aversion across countries. The third group of policy areas (restart possibilities and bankruptcy legislation) is based on relevant indicators, but it is not clear whether they accurately measure the policy areas. All the included indicators are based on expert judgments of the efficiency in handling a given bankruptcy case across OECD countries. Expert judgment can be biased, even though substantial checks are included in the collection of these data by the World Bank (Doing Business 2005). The fourth group of policy areas (entrepreneurship education and entrepreneurial motivation) is based on relevant indicators, but the underlying data are not collected by standardized procedures and they are based on a value judgment. Therefore more detailed analyses are needed in these areas.

Promoting Entrepreneurship

125

Appendix C: Description of the 24 Policy Areas7

Policy Areas Affecting Opportunities Entry Barriers/Deregulation Minimizing government activities and regulation in existing markets creates new business opportunities within established markets, thereby creating a larger demand for potential entrepreneurs while at the same time improving market dynamics. Rolling back government activities (e.g., the liberalization of the telecommunication sector in several European countries in the 1990s) or by deregulating the legal barriers (e.g., relaxing the educational requirements for starting a business in certain sectors) are two ways to improve access to existing markets. Access to Foreign Markets Globalization has opened up for increased international opportunities for entrepreneurs. The decrease in trade barriers and the integration of world markets have made it possible for all types of companies—including new ones—to exploit global opportunities. Even though trade barriers are decreasing due to efforts from international organizations and, as such, are out of the hands of national governments to some extent, national governments can still initiate globalization programs, which help or motivate entrepreneurs to look abroad from the very birth of new firms. Technology Transfer Effective technology or knowledge transfer regulation opens up and speeds up the process of transferring public research into business, thereby effectively creating new opportunities for potential entrepreneurs. This regulation can be enhanced by policies encouraging universities (and other institutions engaged in research and development activities) to facilitate the development of ventures based on publicly funded research. Most important, legislation should develop the legal infrastructure that gives universities ownership of intellectual property developed from publicly funded research as well as the establishment of technology transfer offices that facilitate joint ventures between companies and universities. Private Demand Conditions The willingness of established firms to use new firms as supplies or partners plays a crucial role in the development of entrepreneurship. For example, the success of Silicon Valley compared to the Boston area in the early 1990s has been explained by the more open attitude to co-operation in Silicon Valley. Policies have only a limited impact on private demand, but the public sector can be a role model in their procurement.

126

Anders N. Hoffmann

Procurement Regulation Entrepreneurship friendly procurement regulation increases the amount of government contracts for goods and services awarded to new companies, thereby effectively creating better opportunities for potential entrepreneurs. Procurement regulation in the widest sense—including competitive tendering schemes focused on the purchase of goods, services or science with a potential commercial value—can be made entrepreneurship friendly by encouraging governmental bodies to allocate a specific share of their purchasing to new companies. Policies Affecting Abilities Traditional Business Education Traditional business education, including basic accounting, marketing and finance, are without doubt important abilities not only to run a company, but also to start a company. Differences in the magnitude of business education among developed countries are significant. Some countries include basic business education in the core curriculum in both primary and secondary schools, whereas in other countries it is available only through electives or at dedicated business schools. The former approach obviously ensures that a greater share of the population possess the basic business skills needed to run a company. Policy initiatives could ensure that basic business skills are acquired over a broad range of educations. Entrepreneurship Education In order to strengthen entrepreneurial abilities through education, teaching methods must be refined from primary schools to universities. Activities that go beyond traditional teaching, such as dedicated entrepreneurship centers, internships, teacher and advisor education, and research are necessary for success. Policy initiatives should ensure the supply and quality of entrepreneurship education. Restart Possibilities Serial entrepreneurs are important as they have already proved their ability to establish a business. Yet, failed entrepreneurs are not always able to restart due to legislative barriers. The learning experience from the failure is debated. The possible lessons from failure versus the lessons from success are more a philosophical questions, but it is unquestionable that barriers to re-starting reduce the number of potential entrepreneurs. The policy focus should consequently be on reducing the legislative barriers for serial entrepreneurship. Bankruptcy legislation is particularly important, but also the time and price for restarting a company may be barriers in some countries.

Promoting Entrepreneurship

127

Entrepreneurship Infrastructure (Public and Private) A strong entrepreneurship infrastructure consists of tightly linked regional networks of skilled and specialized advisors with relevant skills and knowledge that assist entrepreneurs, thereby effectively increasing the abilities available to potential entrepreneurs. Advisors can range from lawyers and accountants to experienced entrepreneurs to domain experts at universities. As such nongovernmental involvement is vital to sustaining entrepreneurial networks. Governments can take an important role by initiating and developing the infrastructure. Policy Areas Affecting Capital Loans The supply of debt capital via more traditional credit markets is vital to entrepreneurial activity. Without a large and efficient credit market to supply firms with efficient debt capital, some entrepreneurs will face a financial barrier making it impossible to seize opportunities. Governments can improve domestic credit markets through initiatives to improve access to debt capital in general or to entrepreneurs specifically. The former includes regulation improving the efficiency and competitiveness in credit markets by making debt capital cheaper and more accessible. The latter includes fiscal guaranties for entrepreneurial loans making banks more motivated to help entrepreneurs. Wealth and Bequest Taxation Wealth and bequest taxes impact directly the supply of early stage investment capital. High taxation levels affect negatively the potential supply of liquidity among individuals, which then limits the number and size of investments made by business angels, friends or family. Policy initiatives reducing the wealth and bequest tax rates would enlarge the potential amount of seed and early-stage capital. Business Angels Business angels are typically wealthy individuals who make direct equity investments in the seed stage of companies, and they tend to provide more managerial and business advice through their greater personal involvement than institutional investors do. Although data are scarce, it is believed that total funding by business angels is several times greater than all other forms of private equity finance. Governments in many countries try to cultivate business angels by organising networks and giving special investment tax incentives. Several countries have also tried to improve information flows between angels and potential entrepreneurs that otherwise tend to be informal.

128

Anders N. Hoffmann

Venture Capital Venture capital is an important source of funding for potential highgrowth ventures in need of significant capital for development, growth, and expansion. In order to enlarge the domestic supply of venture capital, governments can either take initiatives to develop national venture funds or improve venture market regulation to grow existing venture markets. The former includes direct investments, and the latter includes relaxing legislation, making it more attractive (or simply possible) for entities, such as pension and insurance funds, to make venture investments. Capital Taxes Capital taxes also have a direct impact on the supply of capital. High taxation levels reduce potential investment rewards, thereby discouraging investments in companies whether new or existing. Policy initiatives reducing capital taxation thus increase financial sources. Some countries also offer special tax incentives for investments in new firms intended to improve the number of business angels. Stock Markets and Buyouts An efficient stock market, a secondary stock market or efficient markets for buyouts are important in order to gather needed capital for the expansion of firms. Furthermore effective exit mechanisms increase the supply of venture capital and also serve as an indirect source to more capital in earlier investment phases. Most countries face the problem of obtaining a critical mass of new firms for a secondary stock market. The development of critical mass must balance two interests. On the one hand, listing requirements and regulations must be simple enough to encourage small businesses to make their initial public offerings through a secondary market. On the other hand, there must be sufficient disclosure, supervision, and enforcement to protect and attract investors. Policies Affecting Incentives Personal Income Taxes High levels of personal income tax reduce the potential financial benefits from starting a business, making it more difficult to reach the cost–benefit equilibrium at which the opportunity becomes worthwhile to pursue. Policy initiatives lowering income taxes are therefore likely to induce a greater number of potential entrepreneurs to engage in entrepreneurial activities. Business Taxes and Fiscal Incentives While corporate taxes do not play a central role for new firms with little or no profit subject to taxation, they will eventually have a significant impact on the profits for high-growth firms. Furthermore, as globaliza-

Promoting Entrepreneurship

129

tion continues to develop, corporate taxation will become a central factor for companies choosing the extent to which they will locate operations abroad. Fiscal incentives can lower entry barriers through financial incentives or support, tax exemptions or rebates, which make more potential entrepreneurs willing to engage in entrepreneurial activity. However, fiscal incentives are a delicate political issue in some countries, and their long-term benefits continue to be questioned. Social Security Discrimination Social security benefits, including health care, pensions, and unemployment benefits, can serve as entry barriers if they are reduced or eliminated as a result of becoming an entrepreneur. Social security policies that put entrepreneurs and wage earners on equal footing in terms of qualifying for benefits can neutralize any discrimination that could otherwise have a negative effect on the amount of potential entrepreneurs pursuing opportunities. Administrative Burdens Administrative burdens are comprised of the amount of time spent collectively to understand and fulfil requirements imposed by governments or other authorities, such as new business registration, filing taxes and financial statements, and understanding which rules and regulations the business is subject to. They can discourage potential entrepreneurs by being overwhelming and difficult to understand as well as being beyond the entrepreneur’s own abilities to fulfill. In countries with substantial administrative burdens, studies show that both job creation and employment settle at lower levels as a result. Policy initiatives to relieve administrative burdens include relaxing the legal demands required to start and run a company. Labor Market Regulation The negative impact of strict labor market regulation, such as high minimum wages and rigid firing regulations are manifold. First, wage employment becomes attractive, thereby increasing the opportunity cost to become an entrepreneur. Second, limitations such as hiring and firing inflexibility can have severe impacts on a corporation trying to grow or to develop a business culture, often through trial and error, that fits with the overall vision and strategy of the company. Finally, high minimum wages means expensive labor and possibly a limiting barrier for a startup. Thus the end result of strict labor legislation is constrained levels of entrepreneurial activity. Bankruptcy Legislation Bankruptcy legislation needs to balance the conflicting risk propensities of creditors and entrepreneurs. Creditors will not provide as much

130

Anders N. Hoffmann

money to entrepreneurial activities if they do not have significant claims to a bankruptee’s assets. However, potential entrepreneurs are less apt to engage in entrepreneurial activity if significant claims are inevitable. The equilibrium, at which the maximum number of potential entrepreneurs can obtain debt capital to engage in entrepreneurial activities, is difficult to both identify and measure, but it is clear that bankruptcy legislation has a strong influence. Governments have a variety of means to relieve the costs of bankruptcy, including debt relief schemes, restructuring, and postponement of debt possibilities. Dept relief schemes can regulate the length, uncertainty, and cost of going bankrupt, thereby altering both direct and indirect costs arising as a result of bankrupt. Reorganization and postponement of debt typically take place prior to bankruptcy, making it possible to alter the business model and, as such, the risk of going bankrupt. Policies Affecting Culture/Motivation Entrepreneurial Motivation Understanding the motivation behind the limited number of entrepreneurs that aim to create high-growth and global enterprises is difficult. It is furthermore a very challenging and slow process of trying to fuel interest in entrepreneurship. Governments can try to enhance preference toward entrepreneurship by implementing entrepreneurship awards and opinion campaigns. Group-Specific Initiatives Awards and opinion campaigns can be targeted toward specific groups, such as women or minority groups, in order to boost the number of entrepreneurs in those groups. Communication about Heroes Elaborating on entrepreneurship history and by communication about and by “heroes” and others help to create a sense of entrepreneurial history, which is important for the evolution of a cohesive entrepreneurship culture. Policy initiatives could ensure the communication of entrepreneurial history and “heroes” in public schools. Policy initiatives could reward “heroes” for communicating their stories in public and acting as role models encouraging others to engage in entrepreneurship.

Notes 1. The author wants to thank Mr. Peter Bogh Nielsen, Statistics Denmark for providing the data. The calculation is based on Danish Structural Business Statistics.

Promoting Entrepreneurship

131

2. For example, Nokia’s former CEO Jorma Ollila often jokes that when Californian entrepreneurs open their garage doors, they had most of the world market waiting in the driveway, whereas the only thing waiting in a Finnish driveway is a meter of snow. 3. The full quality evaluation is found in Hoffmann et al. 2005 but appendix C summarizes the evaluation for the policy areas that are found important in the sections that follow. 4. This depends on whether the firm has income from subsidiaries. 5. The output gap is estimated by the OECD and can be obtained from the Economic Outlook Database. Yearly output gaps exist for all countries except Korea. Korea is estimated for the output gap by the deviation of output growth from average growth in Korea, 1988 to 2004. In the regression the business cycle variable in period t is the average of yearly output gaps from t to t + 2. 6. Florida International University, Howard University, University of Illinois at UrbanaChampaign, University of North Carolina at Chapel Hill, University of Rochester, University of Texas-El Paso, Wake Forest University, and Washington University in St. Louis. 7. The appendix is taken from Hoffmann (2006).

References Acs, Z. 2005. A formulation of entrepreneurship policy. FSF-Nutek Award Winning Series, Entreprenorskapsforum, Sweden. Acs, Z. J., D. Audretsch, P. Braunerhjelm, and B. Carlsson. 2005. Growth and entrepreneurship: An empirial assessment. Working paper 32-2005. Max Planck Institute of Economics, Leipzig. Aldrich, H. 2000. Learning together: National differences in entrepreneurship research. In D. Sextion and H. Landström, eds., The Blackwell Handbook of Entrepreneurship. Oxford, UK: Blackwell Business. Audretsch, D., and R. Thurik. 2000. Linking entrepreneurship to growth. Paper prepared for the OECD Directorate for Science, Technology, and Industry, Paris. Audretsch, D. B., M. A. Carree, A. J. van Stel, and A. R. Thurik. 2002. Impeded industrial restructuring: the growth penalty. Kyklos 55 (1): 81–98. Audretsch, D. B., R. Thurik, I. Verheulm, and S. Wennekers, eds. 2002. Entrepreneurship: Determinants and Policy in a European—U.S. Comparison. Dordrecht: Kluwer Academic. Bandt, N. 2004a. Business dynamics, regulation and performance. STI working paper. OECD, Paris. Birch, D. L. 1987. Job Creation in America: How Our Smallest Companies Put the Most People to Work. New York: Free Press. Birch, D. 1995. Hot Industries. Cambridge, MA: Cognetics. Carree, M., and A. Van Stel. 2004. Business ownership and sectoral growth: An empirical analysis of 21 OECD countries. International Small Business Journal 22 (4): 389–419. Davidson, P. 1999. Entrepreneurship—and after? A study of growth in small firms. Journal of Business Venturing 4: 211–26.

132

Anders N. Hoffmann

Delmar, F., P. Davidson, and W. Gartner. 2003. Arriving at the high-growth firm. Journal of Business Venturing 18: 189–216. Desai, M., P. Gompers, and J. Lerner. 2003. Institutions, capital constraints and entrepreneurial firms’ dynamic: Evidence from Europe. Working paper 10165. NBER, Cambridge, MA. EBST. 2005. Entrepreneurship Index 2005. Copenhagen: National Agency for Enterprise and Construction. Evans, D. 1987. The relationship between firm growth, size, and age: Estimates for 100 manufacturing industries. Journal of Industrial Economics 35 (4): 567–81. European Council. 2006. Press release, Council of the European Union, 6964/06 (Press 65). 2715th Council Meeting on Competitiveness (Internal Market, Industry, and Research). EU Council, Brussels. European Commission. 2002. Entrepreneurship: A Survey of the Literature. Prepared for the European Commission, Enterprise Directorate General, by D. B. Audretsch. http:// europa.eu.int/comm/enterprise/entrepreneurship/green_paper/literature_survey _2002.pdf. European Commission. 2003. Green paper entrepreneurship in Europe. Brussels, 21.01.2003, COM(2003) 27 final. http://ec.europa.eu/enterprise/entrepreneurship/ green_paper/green_paper_final_en.pdf. European Commission. 2004. Action plan: The European agenda for entrepreneurship. Brussels, 11.02.2004 COM(2004) 70 final. http://ec.europa.eu/enterprise/ entrepreneurship/promoting_entrepreneurship/doc/com_70_en.pdf. EUROSTAT. 2005. Business demography in Europe. European Commission, Luxembourg. http://epp.eurostat.ec.europa.eu. Gavron, R., M. H., Cowling, and A. Westhall. 1998. The Entrepreneurial Society, IPPR. London: Central Books. GEM. 2001. Global Entrepreneurship Monitor. P. D. Reynolds, S. M. Camp, W. D. Bygrave, E. Autio, and M. Hay, eds.. Babson College, Wellesley, MA. Gentry, W., and G. Hubbard. 2000. Tax policy and entrepreneurial entry. American Economic Review 90 (2): 283–87. Giovannini, E., A. Hoffmann, M. Nardo, M. Saisana, A. Saltelli, and S. Tarantola. 2005. Handbook on Constructing Composite Indicators: Methodology and User Guide. Paris: OECD. Grilo, I., and R. Thurik. 2005. Latent and actual entrepreneurship in Europe and the US: Some recent developments. International Entrepreneurship and Management Journal: 441–59. James, H. 2006. Europe: Cultural adjustment to a new kind of capitalism? Paper presented at Venice Summer Institute July 2006, Perspectives on the Performance of the Continent’s Economies. Hart, D. eds. 2003. The Emergence of Entrepreneurship Policy—Governance, Start-ups, and Growth in the U.S. Knowledge Economy. Cambridge: Cambridge University Press. Hoffmann, A., P. B. Nellemann, M. Larsen, and N. V. Michelsen. 2005. Indicator manual. FORA, Copenhagen. http://foranet.dk/upload/quality_assessment_of _entrepreneurship_indicators_001.pdf.

Promoting Entrepreneurship

133

Hoffmann, A., and M. Junge. 2006. Comparing the number of high-growth entrepreneurs across 17 countries. FORA working paper. Copenhagen. http://foranet.dk. Hoffmann, A. 2006. A rough guide to entrepreneurship policy. In R. Thurik, D. Audretsch, and I. Grilo, eds., Handbook of Entrepreneurship Policy. Cheltenham, UK: Edward Elgar Press, 2006. Hoffmann, A., P. Nielsen, and S. Vale. 2006. Measuring in start-up rates in EU countries. FORA working paper. FORA, Copenhagen. www.foranet.dk. Iversen, R., Jørgensen, O, and N. Malchow-Møller. 2005. Defining and measuring entrepreneurship. Working paper. Centre for Economics and Business Research. London. http://cebr.dk/upload/2004-a2-14.pdf. Knight, F. [1921] 1971. Risk, Uncertainty and Profit. Chicago: University of Chicago Press. Kauffman, E. M. 2005. Kauffman Thought Book. Kansas City, MO: Kauffman Foundation. Kauffman, E. M. 2006. The Kauffman initiative. www.kauffman.org. Lee, C., W. Miller, M. Hancock, and H. Rowen. 2000. The Silicon Valley Edge: A Habitat for Innovation and Entrepreneurship. Stanford: Stanford University Press. Lundström, A., and L. Stevenson. 2002. On the Road to Entrepreneurship Policy. Stockholm: Swedish Foundation for Small Business Research. Lundström, A., and L. Stevenson. 2005. Entrepreneurship Policy: Theory and Practices. ISEN International Studies in Entrepreneurship. New York: Springer. OECD. 2002. High-Growth SMEs and Employment. Paris. OECD. 2003a. The Sources of Economic Growth in OECD Countries. Paris. OECD. 2003b. Entrepreneurship and Local Economic Development Programme and Policy Recommendations. Paris. OECD. 2004. Venture Capital Trends and Recommendations. Paris. http://oecd.org/ dataoecd/4/11/28881195.pdf. OECD. 2005. micro-policies.

Micro-policies

for

Growth

and

Productivity.

http://oecd.org/sti/

Parker, S. 2003. Does tax evasion affect occupational choice? Oxford Bulletin of Economics and Statistics 63: 379–94. Pinkston, J., and J. R. Spletzer. 2004. Annual measures of gross job gains and gross job losses. Monthly Labor Review, November, US Census Bureau. Washington, DC: GPO. Reynolds, P. D., M. Hay, and M. Camp. 1999. Global Entreprenurship Monnitor. Executive Report. Kansas City, MO: Kauffman Center for Entrepreneurship Leadership. Reynolds, P., N. Bosma, E. Autio, S. Hunt, N. de Bono, I. Servais, P. Lopez-Garcia, and N. Chin. 2005. Global entrepreneurship monitor: Data collection design and implementation, 1998–2003. Small Business Economics 24 (3): 205–31. Scarpetta, S., P. Hemmings, T. Tressel, and J. Woo. 2002. The role of policy and institutions for productivity and firm dynamics: Evidence from micro and industry data. OECD working paper. Paris.

134

Anders N. Hoffmann

Schramm, C. J. 2004. Building entrepreneurial economies. Foreign Affairs 83: 104–15. Schumpeter, J. [1911] 1949. The Theory of Economic Development. Cambridge: Harvard University Press. Smallbone, D., R. Leig, and D. North. 1995. The characteristics and strategies of highgrowth SMEs. International Journal of Entrepreneurial Behaviour and Research 1 (3): 44–62. Storey, D. J. 2002. Methods of evaluating the impact of public policies to support small businesses: The six steps to Heaven. International Journal of Entrepreneurship Education 1: 181–202. Vale, S. 2006. The international comparison of start-up rates. OECD working paper. Statistical Directorate, Paris. Van Stel, A., D. Storey, and R. Thurik. 2006. The effect of business regulations on nascent and actual entrepreneurship. Discussion paper on Entrepreneurship, Growth and Public Policy 2006-04. Max Planck Institute of Economics, Leipzig. Verheul, I., S. Wennekers, D. B. Audretsch, and R. Thurik. 2003. An eclectic theory of entrepreneurship: Policies, institutions and culture. In D. B. Audretsch, R. Thurik, I. Verheul, and S. Wennekers, eds., Entrepreneurship: Determinants and Policy in a European— U.S. Comparison. Dordrecht: Kluwer Academic, 11–83. Wood, P. M., and A. Bandura. 1989. Goal setting and monetary incentives: Motivational tools that can work too well. Compensation and Benefits Review 26 (May–June): 41–49.

5

Innovations to Foster Risk-Taking and Entrepreneurship Robert J. Shiller

5.1

Introduction

Entrepreneurship is a delicate organism. It needs the right environment to flourish. To flourish, it has to incentivize solid ventures, by people who rely on good long-term vision, and not excessive risk-taking or faddish trends, such as have contributed to the recent world financial crisis. Entrepreneurship depends on the animal spirits of its people, and the resulting atmosphere for business ventures.1 There are significant differences across countries and through time in the entrepreneurial environment, and these differences ought to have a lot to do with their economic success. I argue in this chapter that while cultural differences across countries have something to do with the differences in entrepreneurship, there is evidence that these cultural differences may be overstated as determining the degree of entrepreneurship of a country. More important are the factors of economic situation within a country that strengthen or inhibit entrepreneurship. High on the list of such factors is the perception of risk to entrepreneurs and investors in them. Thus, developing new institutions that manage the risk of entrepreneurship is central (indeed risk-taking is what entrepreneurship is all about). There has been great progress in the development of institutions to manage entrepreneurial risk in recent years, the advent of more sophisticated methods of private equity and venture capital. In the future there ought to be further development of risk management institutions that encourage entrepreneurs, and we are in the process of seeing this happen. The further development of risk management contracts that settle on economic indexes will help promote entrepreneurship by allowing entrepreneurs to protect themselves against more and more of the risks of doing business.

136

Robert J. Shiller

Also on the list for promoting the kind of entrepreneurship for promoting economic growth is to develop better institutions for encouraging basic scientific research. Successful basic scientific research is an enterprise much like any business enterprise. We have to construct as well as we can a marketplace where scientific research can compete, and to provide risk management solutions to scientific research insofar as is possible. 5.2 Cultural Differences That Affect Innovation and Entrepreneurship There is a lot of popular talk about differences across countries in their attitudes towards business. For example, it is widely alleged that US North Americans, British, Armenians, Chinese, and Japanese have unusually strong traditions of entrepreneurship that are inhibited by historical forces from time to time but that provide a sound base for these countries to build upon. Nonetheless, with the world economic crisis that started around 2007 to 2008, there are concerns that the entrepreneurial culture in the United States, and possibly elsewhere, has led to excessive, even reckless, risk-taking. There are, of course, many cultural differences between countries. The difficult question is how important are these differences in promoting entrepreneurship. Cultural differences take many forms. For example, it is commonly alleged that some countries are more intellectual and some more practical in their orientation. Countries that have a practical orientation, it is alleged, are more successful in business. In fact, however, promoting highly successful entrepreneurship, the kind that leads to major new forces in the economy, depends on a good balance between the intellectual and practical: a strong interest in deep pursuits coupled with a practical interest in finding applications of these pursuits. Our modern era of rapid economic growth seems to derive from a better recognition of this fact. Until two centuries ago, universities around the world were focused largely on theology and the study of Greek and Latin classics. These were gentlemanly pursuits. Scientists—doing experiments in their laboratories to learn some practical facts about how things work— were often considered too mundane for the university, until modern times. Aristocratic tendencies have provided an obstacle to effective practices that might foster pathbreaking innovations for business. The

Innovations to Foster Risk-Taking and Entrepreneurship

137

French Revolution represented a historic break with this aristocratic past but did not unequivocally succeed in improving the atmosphere for innovation. Some drastic steps were taken: in 1793 all French universities were abolished. They were replaced by professional schools for teachers, doctors, and engineers. But the resulting schools were perhaps a little too practical to be optimal. The French university was eventually restored in 1896. It was not until well into the nineteenth century that a more practical focus, such that would spur expansive and productive scientific research, began in Europe, in the German universities. The principle of Einheit von Forschung und Lehre (“unity of research and education”), championed by Alexander von Humboldt (1767–1835), bringing practical and implementable scientific research into contact with deep intellectual discourse, became the hallmark of the German universities, which, as their strengths came to be appreciated, were imitated around the world, with greater or lesser success depending on the national culture.2 The United States from its beginnings had a general contempt for aristocratic privilege, and at the same time an admiration of practical people. This was long ago noted by Benjamin Franklin (1706–1790). In a 1784 article “Information to those who would remove to America” he wrote (1784: 506): The natural geniuses that have arisen in America have uniformly quitted that country for Europe, where they can be more suitably rewarded. . . . [in America] people do not enquire concerning a stranger, what is he? but what can he do? If he has any useful art he is welcome; and if he exercises it, and behaves well, he will be respected by all that know him; but a mere man of quality, who on that account wants to live upon the public, by some office or salary, will be despised and disregarded.

But, this practical orientation did not make the United States an innovator, not until it learned from Germany about how to restructure its educational system. In one of the earliest imitations of the German university model, in 1861 the Massachusetts Institute of Technology adopted a seal that showed as its motto Mens et Manus (“mind and hand”), a scholar and a laborer with a hammer and anvil standing together. The German university model eventually found an even more conducive environment in the United States than it did in Germany. With their emphasis on the practical, and with no aristocratic traditions that would interfere with the promotion of business-oriented research at the

138

Robert J. Shiller

universities, the United States eventually promoted such research institutions far beyond Germany did. This evidence suggests that the cultural differences make the United States possibly a more fertile ground than Europe for practical businesses, though less fertile for great works of art and treatises on philosophy. There are indisputably cultural differences that relate to business. But there appears to have been relatively little research on how pervasive and how serious they are, and that they really are meaningful in understanding differences in success of economies. A number of researchers have attempted to relate differences in entrepreneurial success across countries with differences in attitudes. David McClelland and his colleagues (1961) devised measures of achievement motivation, and compared these across countries. Geert Hofstede (1980) measured differences across countries in uncertainty avoidance. Fritz Gaenslen (1986) measured individualism in various countries. There are varying opinions about how successful their measures of human attitudes explain differences across countries in entrepreneurial success; my reading is that their successes are only modest.3 When I attended a conference on the US economy in Moscow in 1989, in the last years of the Soviet Union when a transition to a market economy was heatedly discussed at the Institute of World Economics and International Relations (IMEMO), I found that there was much talk about the differences between Russians’ attitudes toward business and entrepreneurship with Western countries. The Soviet economists I met seemed generally to have a poor opinion of their own people’s ability to promote successful business dealings. But, when I questioned them, none of them had any solid evidence about the difference in attitudes between Russians and people of other countries. At this conference I met a young economist, Maxim Boycko of IMEMO, and we, together with the Ukrainian sociologist Vladimir Korobov, decided to do a study comparing attitudes between Russians and people of other countries. Our method was to write questionnaires about business attitudes, translate these into various languages as well as Russian, and try the questionnaires in various countries. We were thus able to compare attitudes toward business across Russia, United States, Ukraine, East Germany, West Germany, and Japan (Shiller, Boycko, and Korobov 1991, 1992). We found a number of differences in attitudes that relate to business across these countries, but not as sharp and significant as the

Innovations to Foster Risk-Taking and Entrepreneurship

139

idle discussions of the Soviet economists would suggest. Our overall conclusion was that the differences across countries were more situational (reflecting the constraints and opportunities in the local economy) than attitudinal (reflecting attitudes toward business). 5.3

Evidence on Differences in Attitudes toward Risk

One important factor that the Russian economists mentioned to me when I visited Moscow in 1989 is that they feared that Russians had lost their risk-taking instinct, after years of living under a communist regime that managed everything for them. But, in my international surveys with Boycko and Korobov, we were unable to find any significant difference between Russia, Ukraine, and the United States in fundamental attitudes toward risk-taking. Our method was to ask about both attitudinal and situational factors related to business in questionnaires that were identical (except for translation into the local language) across countries. Attitudinal factors relate to the individual’s psychology, situational factors are aspects of the individual’s perceived economic environment. Risk aversion, if measured clearly as a taste for risk, is attitudinal. We asked: Imagine you are suddenly ill. Your illness generally does not interfere with your work and does not prevent you from leading a normal life, but you always feel tired, go to bed early, and you are often in a bad mood because of that. The doctors tell you that this tiredness will continue your whole life unless you decide to undergo a surgical operation. (Imagine that you have complete trust in the doctors and you doubt neither their diagnosis, nor the proposed method of treatment.) The operation will completely cure you, but it is risky: there is one chance in four that you will die. Would you decide to undergo this operation? Table 5.1 Responses to question: Would you decide to undergo this [risky surgical] operation? Yes, take risk

No, avoid risk

Survey size

Russia

58

42

105

Ukraine

61

39

97

United States

59

41

119

140

Robert J. Shiller

Table 5.2 Responses to question: Would you be tempted to invest a substantial portion of your savings in it [risky business]? Yes

No

Survey size

West Germany

14

86

132

East Germany

28

72

127

Russia (Moscow)

51

49

122

Russia (Omsk)

42

58

103

Ukraine

44

56

215

United States (New York)

33

67

117

United States (continental)

30

70

122

People in all these countries seem virtually identical in how they would deal with such a personal problem. The evidence here suggests no difference across countries in fundamental risk aversion. An example of a question that probed situational (as well as attitudinal) factors is the following: Suppose that a group of your friends is starting a business that you think is very risky and could fail—but that might also make investors in that business rich. Would you be tempted to invest a substantial portion of your savings in it? Here we see some significant differences across countries. Notably the West Germans were the least likely to want to take on entrepreneurial risks, a conclusion that accords with some popular characterizations of them. But the differences were not always the ones we expected based on conventional stereotypes. The Moscow Russians, not the Americans, were the most likely to say they would take the risks. East Germans (at the time of this questionnaire just emerging from a socialist dictatorship) were more likely to want to take business risks than the West Germans. The differences across countries appear by these measures, as well as other measures in our study, to relate to perceived risks, rather than attitudes toward risks. Why the Russians or East Germans felt the perceived business risks less at the time of the breakup of the Soviet Union is a mystery that we may never be able to unravel. But the general principle here is that perceived risks appear to be a significant barrier to entrepreneurship, and that if we overcome these barriers, there may be no difference in willingness to take business risks.

Innovations to Foster Risk-Taking and Entrepreneurship

5.4

141

Promoting Management of Business Risks: Venture Capital

The development of our financial institutions over the centuries has made it increasingly easy for people to take business risks with a hope of reward and yet not subject themselves to excessive risk. There has been important progress over the last half century in the way entrepreneurship is financed. The world’s first venture capital firm, American Research and Development, was established in 1946 by Massachusetts Institute of Technology president Karl Compton, along with Georges F. Doriot, a professor at the Harvard Business School. Their revolutionary idea was to set up a publicly traded company that invested in a lot of very high risk high tech start-ups, expecting most to fail, but benefiting from the long right tail of the distribution, the enormous success of the small number that succeed. This was a fundamentally different method than that which commercial banks use in their dealings with entrepreneurs. The venture capital firms could offer terms to the firms that they invested in that would protect the entrepreneurs from much of the enormous risk that such ventures entail. In the sixty years since their inceptions, venture capital firms have gradually learned how to manage the risks of their companies and to deal with the agency problems that may be very strong since the success of high-tech ventures may be so hard to judge. They have invented staged financing that keeps venture capitalists on a leash whose length is accurately controlled, they have developed syndication methods that allow the venture capitalist to get second and third opinions about the success of their investments, they learned through time how to participate in the boards of directors of the companies in such a way as use their more general knowledge to help pursue opportunities for the businesses they have invested in, they learned how to award equity or options grants to managers to better align the interests of employees and investors, they developed gradual vesting programs to control the time pattern of managerial loyalty to the firm, they learned techniques to facilitate the weaning of the invested firms from the venture capitalist and placing of them into public markets without generating adverse selection problems (see Gompers and Lerner 2001). The development of private equity and venture capital has been mostly associated with the United States. Though these institutions have been growing in Europe, they have not had as strong an impact

142

Robert J. Shiller

there. The performance of European venture capital firms in 1993 to 2003 has been disappointing, and the performance has been particularly bad for the critically important early-stage venture investments (Machado Rosa and Raade 2006). Some of this disappointment is due perhaps to errors in judging the high-tech bubble of the 1990s, but that memory lingers on and inhibits venture capital even today. Laura Bottazzi and Marco Da Rin (2002) have argued that the quality of European venture capital, overall, is not yet equal to that in the United States. They conclude that European venture capital firms have produced fewer “superstar” companies and, more generally, are not systematically associated with more dynamic companies. Not only is the total quantity of venture capital less, but the venture capital tends to be spread out much more across firms in Europe, much smaller amounts invested per company. They speculate that, despite its rapid growth, European venture capital suffers from a relative shortage of qualified experienced professionals. The European Commission has noted the importance of venture capital, and has as a result transformed the European Investment Fund (EIF) into a major investor in venture capital. The EIF is a public-private partnership owned by the European Investment Bank (62 percent), the European Union (30 percent), and European banks and financial institutions (8 percent). This is a laudable effort to promote venture capital in Europe. But, the total portfolio of the EIF as of 2005 was only *3.2 billion. In contrast, the total amount of venture capital investments in the United States made in 2005 alone was $22.2 billion.4 Creating a quasi-government fund may not be in the true spirit of venture capital, and perhaps other ways of promoting venture capital in Europe should be pursued. 5.5 Promoting Management of Business Risks: Index-Based Derivatives Other developments in capital markets in recent decades have been the rapid development of derivatives markets. The development of the first financial futures and options markets, as well as swaps began in the 1970s and 1980s. We are seeing now the development of index-based derivatives, and the proliferation of these derivatives is something that can help entrepreneurship by promoting hedging against a multitude of specific risks that impinge on the specifics of the venture. Any index of the outcome

Innovations to Foster Risk-Taking and Entrepreneurship

143

of an economic risk can be the basis of settlement of risk management contracts. Any time the uncertainty related to pursuing innovation is reduced, it becomes easier for entrepreneurship to proceed. I will give one example of the effects of the development of a new risk market on entrepreneurship that I am particularly involved in: the development of derivative products for real estate risks, particularly risks to the value of owner-occupied homes. Owner-occupied homes are arguably the biggest asset class of all. In the United States real estate currently owned by households amounts to about $20 trillion. This is larger than the stock market. The risks to this asset class appear substantial. The market value of homes has certainly been volatile in recent years. And yet, there has been virtually no issuance of derivative products that would hedge the risks of these. It is true that in 1991 the London Futures and Options Exchange attempted to create futures markets for these, but that effort quickly failed. It is true that in the United Kingdom derivative products for commercial real estate have begun to proliferate in the past few years, and there are Goldman Sachs covered warrants traded on the London Stock Exchange, but these are still quite small in total value. In the late 1980s my colleague Karl Case and I thought that the obstacle to developing good derivative markets for single-family homes was that there was no good measure of their value. We then developed a weighted repeat-sales methodology and began producing home price indexes that were far better any other available. The indexes we created were specifically designed for contract settlement (see Shiller 1993, 2003) and have been improved over the years and are now produced in collaboration with Fiserv, Inc. (a financial services firm) and Standard & Poor’s (a financial index provider and rating agency). The firm that I co-founded with Allan Weiss, and Sam Masucci, MacroMarkets LLC, teamed up with the Chicago Mercantile Exchange (now the CME Group), Fiserv, Inc., and S&P’s to create futures and options markets for single-family homes on each of ten United States metro areas. These are cash settled based on the S&P/Case Shiller Home Price Indexes. These began trading May 22, 2006 (see http:// www.housing.cme.com), and as of August 2006 the futures open interest totaled $67 million dollars, and including the underlying value represented by the options, the combined contracts stood at over $100 million dollars. Unfortunately, as of 2010 the open interest has plummeted to only a few million dollars. The CME Group has taken some steps to revive the market, and through the work of John Dolan there

144

Robert J. Shiller

is now a new website homepricefutures.com that reports on this market. The problem at this point is that regaining liquidity in this market is hampered by the low level of liquidity that is in this market now. People don’t want to trade in a market unless many others are trading, so a vicious circle holds this market back, for now. The next stage in this enterprise, and others, is to develop real estate index-linked notes (including the ETF-like MacroShares that Weiss and I invented and that my colleagues at MacroMarkets LLC, Terry Loebs, Sam Masucci, Bob Tull, and others, have developed and promoted), as well as swaps and other over-the-counter products. Some significant successes, largely in Europe, have been achieved using the Investment Property Databank (IPD). We do not yet know how successful these new risk management devices will eventually be, but the kinds of interest and explorations that these new markets have stimulated suggest that the ability to manage real estate risk could ultimately create a multitude of new enterprises. We expect to see new kinds of mortgage lenders, who provide downpayment insurance to home buyers worried about the value of their homes, and hedge the risks that they incur in writing the mortgages on the new derivative markets. We expect to see new kinds of homeowners insurance providers, who protect homeowners against risks to the value of their homes and not just against physical damage to their homes, and then hedge the risks that they incur in writing the policies in the new derivative markets. We expect to see a proliferation of homebuilder activities, spurred by such new concepts as a homebuyer price warranty, activities that will then be hedged in the new derivative markets. We even expect someday to see that those individuals who buy and fix up homes for resale (a very popular form of micro entrepreneurship) will get a shot in the arm when it becomes possible for these people to pursue their business in a manner that is hedged against aggregate real estate risk. In all these cases entrepreneurship is facilitated by dealing anew with risks that had inhibited action before. 5.6

Promoting Scientific Enterprise

Scientific enterprise, if it is going to continue to push back the frontiers of knowledge, requires a kind of entrepreneurship that is not dissimilar from that of business. Technological progress is the main driver of economic growth, and this progress is the outcome of enterprise just as much as progress in building automobiles is. Unfortunately, in the

Innovations to Foster Risk-Taking and Entrepreneurship

145

current world economic crisis, the importance of sponsoring scientific research may not be on the front of everyone’s minds, and some scientists are suffering just when they should be the centerpiece of stimulus packages. Major breakthroughs in basic science may require expenditures of many hundreds of millions of euros over a period of many years to achieve an important mission. The projects may involve the coordination of large teams of researchers. Scientific research requires long time periods to come to fruition. Moreover scientific research thrives on competition, just as do businesses. The competition allows the most capable to survive, and attract further funds to allow the research to function. Highly risky ventures need to get funding, and the funding needs to take account of the nature of the risks, to consider the possibility of diversifying risks by investing in many small projects. There is, however, an important obstacle to the development of markets for scientific research: Scientific research, particularly basic science research, functions substantially as a public good. The direct returns to the individual researcher, if any, are often dwarfed by the externalities that the research generates. Researchers have to be willing to share their results freely if science is to progress, and doing so inhibits their ability to profit from it. For this reason the research requires government funding. Government funding is most effective if it can somehow create the semblance of a market for scientific research, allowing scientists to run their enterprises as if they were collecting the profits that accrue to their discoveries. The system that has to be constructed is one that allows researchers to compete and be rewarded for major new ideas. In the nineteenth century Germany led the way in effectively sponsoring scientific research, with the invention there of the graduate school, which was an academic research organization built around the training of young researchers. This gave Germany an intellectual edge for some time. It is significant, for example, that the field of economics was substantially dominated by Germany during that period, and that most of the founders of the American Economic Association were trained in Germany. However, the advantage that Germany had was lost sometime in the twentieth century. Sinn (2003) points out that Germans were prominent on the list of Nobel prize winners in science in the early twentieth century, and that their numbers dropped sharply by the late twentieth century.

146

Robert J. Shiller

The system that is in place in the United States has been a model for much of the world. The National Science Foundation was founded in the United States on the advice of Vannevar Bush in his famous 1947 book Science, the Endless Frontier. The most basic idea in his book was that the funding of science should create a mechanism—I would say a marketplace—for shifting government funds to the most successful scientific researchers. There would be no government scientific agenda beyond rewarding the most fruitful research. How to create such a marketplace for ideas? Bush’s invention was to create committees of the most successful scientists who would anonymously review proposals submitted to them by individual scientists (not scientific organizations). In this environment top scientists are able to let their creative spirits run unrestricted by institutional or political restraints and reward those who pique their imagination and only those. Moreover, Bush stressed, the reward should be relatively longterm and stable funding, and large enough for serious enterprise, today we might say it is like the funding that commercial enterprises routinely receive from their venture capitalists. The US National Science Foundation was highly successful in stimulating important research. Freed from linking their research to government or business agendas, researchers produced quality research that ultimately promoted business interests. The US NSF was imitated by many other countries, and these national science foundations typically follow the same pattern of peer review that the US model does. These have of course had many successes. But many of these national science foundations have a problem inherent in the fact that they are organized by countries much smaller than the United States: it is difficult to maintain anonymity in the peer review process. Knowing the topic of the proposal and the fact that it probably comes from within the country, the peer reviewer of the proposal probably knows the proposer, and cannot make a politically unconstrained judgment. Moreover the proposer is more likely to be able to guess the name of the reviewer. With the loss of anonymity, the process of candid assessment of scientific contribution breaks down. Owen-Smith et al., in a study of life sciences research comparing the United States and Europe, concluded that the tendency for funding sources in Europe to be national rather than European has led to national clusters of specialists in Europe that have tended to evolve in such as way as to deepen already narrow competencies rather than broad exploration. In contrast, in the United States, the regional clusters

Innovations to Foster Risk-Taking and Entrepreneurship

147

tended to develop in more generalist directions, pursuing multiple therapeutic areas, more integrative of “goal-oriented therapeutic research with fundamental biological investigation.” They concluded that the latter kind of research that is more conducive to cooperation with commercial ventures. The solution for Europe is to create a national science foundation of sufficient scale, functioning like the US NSF for all of Europe. The European Unions have done this, but the scale is not large enough. The Council of the European Union reached an agreement, that, as part of the Seventh Framework Program (FP7) a new European Research Council (ERC) will be created. The ERC was then launched in February 2007. But the ERC that was created is too small. The US NSF had a budget of $5.480 billion in fiscal year 2005, of which $4.234 billion went to research and related activities. In contrast, the ERC received a budget of only *7.5 billion, for 2007–13, or, on an annual basis, about one quarter the size of the US NSF budget. Even after the establishment of the ERC, the bulk of scientific research in Europe will be handled by regional national science foundations. For example, the Deutsche Forschungsgemeinschaft (DFG) had a 2005 budget of *1.390 billion, the Max Planck–Gesellschaft in 2003 had a budget of *1.234 million. The Hermann von Helmholtz–Gemeinschaft Deutscher Forschungszentren e.V. has a recent annual budget of *2.2 billion. In Italy the Consiglio Nazionale delle Ricerche (CNR) had a 2002 budget of *980 million. Obviously the national research institutions dwarf the European Research Council. 5.7

Conclusion

It is naturally hard for any country to control the level of entrepreneurship within its borders. Entrepreneurship relies on the “animal spirits” of its inhabitants, which is impossible to legislate. But I have argued that there is a lot involved in the economic environment that can make or break entrepreneurship independent of the natural attitudes of the population. Notably the revolution that has been occurring over recent decades in venture capital and private equity needs to continue. We should not allow the financial crisis of 2007 to 2009 to interfere with further progress in institutions that promote entrepreneurship. These institutions have been and should continue to be innovating rapidly in the provision of risk management, as well as sophisticated incentive packages.

148

Robert J. Shiller

For entrepreneurs this is a fundamental development that needs to be promoted in every country. The development of derivative markets, which has been proceeding at a rapid pace, helps promote business activities at all stages, including the beginning stages of entrepreneurship. Of particular interest going forward is the proliferation of markets for economic indexes that will allow individual entrepreneurs to limit the specific risks that their venture entails. The development of institutions for scientific research that, in effect, create an anonymous marketplace for funds supporting genuine basic research needs to be pursued further. Basic scientific research is the ultimate source of much of our economic growth. All these things are already happening. But progress could be much faster if governments recognized their significance and promoted them more. Notes 1. See Akerlof and Shiller (2009). 2. See Ben-David (1977). 3. Comparing national indicators of attitudes mislead us in evaluating the prospects of entrepreneurial success of nations. Lamont (1992) and Giannetti and Simonson (2003) have found evidence of major variations across regions within European countries in attitudes relating to entrepreneurial spirit. Regional or cultural pockets of entrepreneurial zeal may serve to supply a country with many entrepreneurs if economic conditions favor entrepreneurship. 4. PricewaterhouseCoopers/National Venture Capital Association Money TreeTM Report, December 31, 2005.

References Akerlof, G. A., and R. J. Shiller. 2009. Animal Spirits: How Human Psychology Drives the Economy and Why It Matters for Global Capitalism. Princeton: Princeton University Press. Ben-David, J. 1977. Centers of Learning: Britain, France, Germany and the United States. New York: McGraw-Hill. Bottazzi, L., and M. Da Rin. 2002. Venture capital in Europe and the financing of innovative companies. Economic Policy 17 (34): 229. Franklin, B. 1784. Information to those who would remove to America. The Boston Magazine 1: 505–509 October. Gaenslen, F. 1986. Culture and decision making in China, Japan, Russia and the United States. World Politics 39: 78–103.

Innovations to Foster Risk-Taking and Entrepreneurship

149

Giannetti, M., and A. Simonov. 2003. Does prestige matter more than profits? Evidence from entrepreneurial choices. Stockholm School of Economics. CEPR discussion paper 4157. London. Gompers, P., and J. Lerner. 2001. The venture capital revolution. Journal of Economic Perspectives 15 (2): 145–68. Hofstede, G. 1980. Culture’s Consequences: International Differences in Work-Related Values. Beverly Hills: Sage Publications. Hsu, D. H., and M. Kenney. 2004. Organizing venture capital: The rise and demise of American research and development, 1946–73. Unpublished paper. Wharton. Lamont, M. 1992. Money, Morals, and Manners: The Culture of the French and American Upper-Middle Class. Chicago: University of Chicago Press. Machado, R., C. Dantas, and K. Raade. 2006. Profitability of venture capital investment in Europe and the United States. Economic paper 245. European Commission, Directorate-General for Economic and Financial Affairs, Brussels. McClelland, D. C. 1961. The Achieving Society. Princeton: Van Nostrand. Owen-Smith, J., M. Riccaboni, F. Pammolli, and W. Powell. A comparison of U.S. and European university–industry relations in the life sciences. Management Science. Powell, W. W., and J. Owen-Smith. 1988. Universities and the market for intellectual property in the life sciences. Journal of Policy Analysis and Management 17: 253–77. Rodgers, D. T. 2000. Atlantic Crossings: Social Politics in a Progressive Age: Cambridge, MA: Bellknap Press. Shiller, R. J. 1993. Macro Markets: Creating Institutions for Managing Society’s Largest Economic Risks. Oxford: Oxford University Press. Shiller, R. J. 2003. The New Financial Order: Risk in the 21st Century. Princeton: Princeton University Press. Shiller, R. J., M. Boycko, and V. Korobov. 1992. Hunting for Homo sovieticus: Situational versus attitudinal factors in economic behavior. Brookings Papers on Economic Activity, pp. 127–94. Shiller, R. J., M. Boycko, and V. Korobov. 1991. Popular attitudes towards free markets: The Soviet Union and the United States compared. American Economic Review 81 (3): 385–400. Sinn, H. W. 2003. Ist Deutschland noch zu Retten? Berlin: Ullstein.

6

Europe: Cultural Adjustment to a New Kind of Capitalism? Harold James

6.1

Introduction

Cultural explanations of economic performance are very popular but inherently problematical. Some analysts think that cultures never change, and hence they cannot explain any kind of alteration; others see culture as shifting rapidly, but in response to economic change.1 Such diverging approaches generate considerable confusion, when presuppositions about cultural determinants are translated into attempts to analyze and even into policy prescriptions. In consequence economists are generally wary of cultural explanations. The first of two quite well-known examples of the impact of culture on economic life is Margaret Thatcher who was captivated by the thesis of the American historian Martin Wiener, which described modern British culture as deeply anti-entrepreneurial and responsible for Britain’s lethargic economic performance.2 But then the culture (and as a result the performance?) changed, and now an Anglo-Saxon legal culture is supposed to generate more effective corporate structures. Second is the conventional thinking, until relatively recently, that poor Asian economic progress could be explained in terms of Confucian values; for the past fifteen years exactly the same values are said to cause exceptional dynamism. Cultural explanations just seem to be tailored to re-describe new realities, and it is not a surprise that they as a result arouse suspicions. Perhaps a more helpful way of tracing the impact of culture is to see it embodied in institutions that create specific incentives that affect the interactions of individuals. In particular, we should think about the interplay of three very powerful social constellations: families, states, and markets. Especially in the European context, this relationship has been much misunderstood. Recently a great deal of literature has been

152

Harold James

devoted to demonstrating that there is no simple opposition between state and market, in particular, that a well-functioning market needs a secure institutional framework that can only be provided by wellfunctioning states. With a complete absence of a state, there is no way of enforcing contracts, which are at the heart of the market process. Where states become abusive, arbitrary, or corrupt, the scope for rentseeking increases. Economic agents in the market instead of looking for technical improvements or innovations as a way of expanding their activity, try to capture the state. The proposition that the state requires the market is perhaps more controversial: but it really should not be so. The extreme example of the malfunctioning of communist systems demonstrated how when states become arbitrary and seek to replace markets, they lose legitimacy because they have extended themselves into too many areas in which individuals have strong feelings. A great deal of this discussion about the way in which an efficient and just operation of markets and states can proceed ignores the contribution of the family to the functioning of markets and states. This is surprising, because the family can be understood as providing a link across generations, and with this a perspective on time. The process of building institutions and markets is one that demands a long-term, not a short-term perspective, and it is a process that becomes dysfunctional if there is no long-term outlook. Economists are often looking for what they call intergenerational equity and for mechanisms for making intergenerational transfers that have the most acceptance and legitimacy. Children are an investment in the future, and societies that do not adequately reproduce themselves have great problems in other areas that demand equity and justice. Children are often the major rationale for individuals to accept sacrifices. Especially in largely immigrant societies such as the United States, there is a powerful tradition of enduring poverty in the belief that the next generation will be better off. In current discussions about the effects of technology and global competition on middle-aged workers who have accumulated a particular and no longer needed set of skills, such arguments are often the only honest consolation for “globalization losers.” My presentation focuses on the diagnosis of Europe’s contemporary malaise as well as on the variety of Europe’s responses to apparent stagnation. It is striking how words such as “capitalism,” “market,” “enterprise,” or “globalization,” which generally evoke positive associations in North America and Asia, are negatively charged in western

Cultural Adjustment to a New Kind of Capitalism?

153

Europe (but less so in central Europe) as well as in Latin America. It is not that Europe is a globalization loser: with powerful export economies, a tradition of innovation, and overwhelmingly prosperous societies, Europe might be thought of primarily as a winner. But Europeans are overwhelmingly worried. First, I examine the political and public policy response to globalization, and why the current European reaction is so problematic (the domain of political culture). A political culture derived from Bismarck that mixed an idealization of the nation-state with a sense of insecurity that could only be countered by welfarism allowed the capture of the political system by particular interest. Then I look at what alternative sources of resilience may be offered by social structures, traditions and beliefs (and in particular, by the family). The analysis thus moves to a broader concept of culture. 6.2

The European Malaise

The Lisbon agenda for transforming the European Union into a modern, technologically driven, knowledge-based economy is generally recognized to have failed. Most modern European states have become large, bureaucratized, inflexible, and incapable of responding to the demands for change in a globalized world. They are not good at spreading technologies or ideas across national frontiers. At the same time these modern states are subject to great, and increasing, expectations about what they might provide: how they should make their citizens not only prosperous, but also happy. Popular and political responses to globalization depend on many chance variables, such as the level of skill of a particular population (more skilled people are likely to be beneficiaries of globalization rather than losers), or the extent of the manufacturing sector (rich countries with a large share of their population employed in manufacturing are likely to find the job losses that accompany changing trade patterns unpleasant and unacceptable). But above all, the reaction to globalization is a function of the size of the state, and the assumptions about how politics operate that flow from the size of the state. One currently prevailing wisdom is that the main beneficiaries of globalization are large states, such as India or China, with a strong geopolitical position, as well as a nuclear potential. They may be, we are often told, about to rewrite the rules of the global order. In fact much of this vision of a Chinese- or Indian-dominated world lies in the future: needless to say,

154

Harold James

China and India are not as successful at globalizing and developing their economies as the real champions like Singapore, South Korea, Taiwan, and Chile. Both giants have a great deal of entrepreneurial vigor, but they also encounter enormous problems in reforming the public sector. Large states in reality have a greater problem in responding to globalization, whereas the small states have characteristics that may make them the natural winners of globalization. They are better at the public policy adjustments, freeing up labor markets, establishing a solid framework for competition, and facilitating cross-border takeovers and mergers. The urgency of such a program has been underlined in much recent writing.3 Small states are also more likely to be successful in defending crucial aspects of the welfare state, while at the same time promoting flexibility, a point made recently by André Sapir who developed a typology of different European reactions in which there was a specific “Nordic model” of capitalism.4 The most impressive and extensive economic and political reforms, allowing an opening up to the global economy, and a dismantling of domestic distortions in the past twenty years, have occurred in Chile, Ireland, and New Zealand. In each case, besides regulatory reform and extensive privatization, the share of public spending in GDP was reduced by over ten percentage points: and the reduction contributed greatly to making a more flexible economy that could grow more quickly. By contrast, the characteristic pathology of large states’ resistance to globalization is particularly evident on the European continent. The political and commentating classes (though not all the business elites) in France, Germany, and Italy (and perhaps also Russia) feel that their nation’s days of economic strength—and perhaps also of political glory—lie in the past, and that the globalized world is becoming hostile and uncomfortable. They think of industrial policy in terms of national champions, from steel to automobiles but also even to yoghurt producers; and they treat any relocation of production following from changing competitive positions as a blow to national prestige. The result is that the former great powers become depressive, backward-looking, nostalgic, and bewildered. The semi-large states in Europe (Poland and Spain) may face a similar political economy problem, with exaggerated expectations of what is politically do-able. The most obvious “sick cases” of the modern European Union are three of the four large countries, Germany, France, and Italy. In each

Cultural Adjustment to a New Kind of Capitalism?

155

the malaise has produced a political blockage, with very narrow election and referendum results, and right and left wing parties that are both bitterly internally divided about liberalization and globalization. The new divisions on the left are within the SPD, the French socialist movement, and the Olive Tree coalition: one part in each group is opposed to globalization, and to the European Union (the disputes came out especially in the course of the French referendum). An analogous dispute in Britain in the 1990s ended with the victory of a reformist wing of the Labour Party, but Blairism has become a term of abuse for most European socialists. Disputes between modernizers and reactionaries are mirrored on the right, where large business interests like the European Union, while traditionally minded voters and small businesses are worried about immigration, the effects of Basel II on lending to small businesses, and competition from Asian producers. The Northern League, or France’s National Front are clearly antiglobalization parties; but such sentiments also exist within parts of the CDU/CSU. The post-1949 Federal Republic was used to being regarded as the great economic success story of Europe, but since the postunification recession of the early 1990s, it never really recovered, and growth was very sluggish. As Hans Werner Sinn has repeatedly pointed out, Germany proved fundamentally incapable of dealing with the aftermath of the East German revolution and the unification of 1990.5 Chances for reform were repeatedly cast away, first by Helmut Kohl’s Center-Right coalition, then by Gerhard Schröder’s Red-Green Coalition, but also unfortunately by the Grand Coalition under Angela Merkel, and then by the new right-center coalition under Mrs. Merkel. The Schröder coalition fell apart prematurely because of the strain caused by a debate about the implementation of the 2010 reform agenda. Frau Merkel has drawn the lessons that prudence is the paramount political virtue. The promotion of new visions of national identity and national pride, such as the recent television campaign Du bist Deutschland looks like a classic response to economic failure, rather than as a really effective way of boosting economic performance.6 And there is still a cultural obsession with decline: one example may suffice, the work of the German sculptor Magdalena Jetelová who won the 2006 Lovis-Corinth-Preis with her broken down and irregularly angled classical temple entitled Prozesse und Phänomene des Niedergangs, des Abschieds und der Perspektivlosigkeit (“processes and phenomena of decline, farewell and the absence of a vision”).

156

Harold James

At the same time as Germany became more a great power in the geopolitical sense, and its politicians talked about normalizing their national existence, it became less like an economic superpower. The consensus society that was at the center of the Wirtschaftswunder looks inflexible, and creates obstacles to change. Germany is surrounded by small states that all seem to be prospering more in the new geopolitical and geoeconomic environment: the rapidly growing central European formerly communist states, Poland and the Czech Republic, but also the other small neighbors on whom Germans traditionally looked down but that are not necessarily low-wage producers: Denmark, the Netherlands, Austria, and Switzerland. After years of German condescension on Austrians, there is even a new and paradoxical slogan, that Austria is “the better Germany.” Recently released statistics showing a faster growth rate in Switzerland than in Germany produced a new round of German breast-beating. Many French writers such as Nicolas Baverez, the author of La France qui tombe, also express a vividly felt sense of national decline. France believes that the rules of the global economy work against French national interests. Policy makers deduce that they need to make up new rules, and plan for a world order that is more to their liking. Even proliberalization politicians tend to take up the radical rhetoric of globalization critics. The former socialist prime minister, Lionel Jospin, eloquently argued that “the need to take control in adapting to reality places a special responsibility on the state. . . . We cannot let supposedly natural laws guide the evolution of our societies. This would be an abdication of our political responsibilities. On the contrary, we must seek to govern the forces that are at work in economic globalization.”7 Jacques Chirac considers “ultraliberalism” as “the new communism,” an enemy requiring complete ideological countermobilization.8 In a widely quoted account, he is supposed to have recognized after the collapse of the Juppé government and its reform program in 1997: “France is an ungovernable country. She no longer wants us, or our reforms. I have drawn the consequences.”9 Italy is even more obviously sick. The centrist political establishment in the 1990s tried to implement reform by selling it as a necessary accompaniment of a process of European integration. The currency union reduced state borrowing costs. But it did not alter microeconomic behavior, and especially not the behavior of labor markets; and both the Berlusconi coalition and (apparently) the new government lack a solid enough political basis to introduce effective reforms.

Cultural Adjustment to a New Kind of Capitalism?

157

Governments of whatever political complexion are vulnerable to holdups by small coalition partners: the Liga Nord for the governments of the right, and now the anti-reform and fundamentalist Rifondazione Comunista for the left coalition. Coalitions can only be held together by the manipulation of old and odd cultural images: for Berlusconi, the image involved the demonization of the “left” and “marxism”; for the left, the only uniting vision is based on disdain for Berlusconi’s showmanship. The economic outcome is an increasing overvaluation and uncompetitive wage costs that could no longer be corrected by a parity alteration of the lira. Some members of the Berlusconi government began to make comments suggesting that Italy would be better off without the currency union. For post–Soviet Russia, the sense of faded imperial glory (and consequently of the particular duty of the state) is even stronger. The newly victorious capitalism looks as if it was simply a strategy to undermine soviet strength and power. Even though there has been a substantial economic recovery since the crisis of 1998, capitalism is widely unpopular, not least because it has produced very wide discrepancies in income and wealth. The political clown Vladimir Zhirinovsky expressed the post-soviet feeling well when he asked, rhetorically: “Why should we inflict pain and suffering on ourselves? We should inflict pain and suffering on other people.”10 By contrast, such an externalization of the costs of adjustment is not a very practical possibility as an expression of political sentiment by the inhabitants of small states. Why is the nostalgic vision of past empire both so powerful and so paralyzing politically? There is the interplay between the possibilities for action and control that a large state offers, and a powerful series of historical arguments and associations that still continue to resonate. Obviously a larger state can inevitably do more to control the economy. In a small state setting, an imposition of a dense network of controls is likely to lead to the loss of mobile factors of production, whereas in a large state it is harder for labor or capital to escape. Small states are thus likely to be quicker in liberalizing their labor markets, and reducing rates of taxation on capital. The difference is also noticeable when it comes to fiscal discipline. Large states can borrow freely on capital markets, and they are subject to constant political pressures that lead to high fiscal deficits while the markets absorb their debt. Since the 1990s, fiscal discipline in the world as a whole, and also in the European Union countries, has improved. But since the late 1990s deficits have increased dramatically in Japan (moving from a central

158

Harold James

government fiscal balance of −3.8 percent in 1998 to −7.0 percent in 2005) and the United States (where the equivalent move is from +0.5 percent to −3.4 percent). The EU countries were constrained at first by the stability and growth pact, but since 2001 there has been a dramatic deterioration in France and Italy, and a milder one in Germany. In the early twenty-first century, government deficits in the large states have soared. Smaller states (and of course, poorer states) are forced to a more rapid adjustment. There are also powerful political economy arguments about why smaller countries should be better at adapting.11 In a globalized world, small states are able more easily to see where there is a comparative advantage and to devote resources to that sector without a big competitive push and pull from other sectors for favors from the state. Thus, to take perhaps the best known examples, Taiwan can pursue the development of electronic goods, Finland mobile telephony, and Switzerland financial services. These all have developed into global leadership positions, and are acutely kept competitive, not in a national but in an international framework. The difference in approach between small and large states is even more noticeable, and more destructive, when it comes to international governance. Large states try to make international rules, and often build their domestic legitimacy on their claims to be able to shape a larger world: they think in terms of mondialisation maîtrisée. Instead of accepting the international system roughly as it is, with all its imperfections, they think that they can use their weight to alter the rules—to their favor. This is true in tussles over the commercial system. Aggressive trade policy is or can be used as an instrument of policy and can create new opportunities for the assertion of power and the development of economic muscle. The current trade order thus reflects the ability of the United States and to some extent also the European Union to impose its vision on the world. Political power can also be applied to serve national interests in issues of corporate governance or the design of the international monetary order. In the making of big state policies, all the aspirations and resentments engendered by a previous wave of globalization are still alive. In the late nineteenth century, faced with an earlier wave of integration of goods, capital, and labor markets, many observers came to the conclusion that a strong state was needed to protect citizens against the consequences of globalization. The European welfare state was invented by Bismarck in Germany primarily as an exercise in domestic politics,

Cultural Adjustment to a New Kind of Capitalism?

159

and Bismarck’s legacy still continues to shape expectations of what the nation-state is and what it can and should do. Bismarck was the master of interest politics; he understood how to assemble political coalitions around economic issues of workers, farmers, and businessmen inclined to demand tariff and quota protection. The smokescreen behind which modern interest politics was created came from the idea or a geopolitical necessity of the state. Some German politicians still present this rationale of why a large state needs to act. When challenged by the apparent success and prosperity of smaller neighbors, they will still explain that the large states have global responsibilities, and that smaller states have simply abdicated their role. Small states are indeed less likely to wrap their politics in a Bismarckian way, with geopolitical activism cloaking domestic interest politics. A word of warning: the association of smallness and reform willingness is not an inevitable relation, and there are plenty of small states, from Greece to Cuba, with abysmal records as reformers. Traditionally smaller states have emphasized their greater level of social solidarity. Hence there is less of a need to put together complex distributional coalitions. More important, small states realize much more readily that they cannot go it alone. They are less likely to fall for the myths and temptations of empire. As a result small states are likely to be more open to trade and more resistant to the dangers of protectionism. They are also more likely to have open capital markets. And immigration is less likely to be disruptive, since the economic aspects of cross-border migration (which are largely beneficial) will outweigh the temptation to see immigration in terms of a clash of values or of civilizations. The best modern analyses tend to argue for ways on institutionalizing the small-scale vision on a general European scale, for instance, by the promotion of regionalism where the region is a focus of loyalty and a facilitator of adaptation but cannot provide an alternative model of an overall economic and political framework. Growth thus often takes place in a decidedly regional, rather than national, framework: in Bavaria, or Lombardy, or Catalonia. The best strategies for growth pick up what is beneficial in the small state tradition of adjusting to the world and seek to harness the paradoxical power of the petite. Small-state or regional approaches are obviously not the only ways of providing a mechanism for adjustment as well as a source of identity in a world that changes with bewildering rapidity. The literature on

160

Harold James

states rightly focuses on the macroeconomic framework required for stability, on sound fiscal and monetary policies. But there is also a need for appropriate microeconomic incentives. These can be generated by a hands-off approach to regulation: but in general, state activism here means the imposition of obstacles, and for positive incentives we need to look elsewhere. 6.3

The Culture of Family Firms

The political roots of Europe’s peculiar pathology of the state are worth further examination. The logic of a powerful state did not simply have its origins in Bismarckian power politics. One of the most powerful drivers of the idea of a strong state and state planning in the second half of the twentieth century was the belief that the nonorganized market had been inefficient, and in particular that growth had been restricted because of the prevalence of many family businesses. The state could plan markets in a rational way (while family capitalists could not). This interpretation was given a boost by the interwar Great Depression, as well as by a concomitant wave of writing about the characteristics of scientific management. The concept of family capitalism is often used, particularly in the United States, to explain deviation from some notion of an ideal-typical American path, of the kind best described by Alfred Chandler.12 Firms in this tradition should move smoothly from an entrepreneurial to a managerial type of organization, with the movement to the multidivisional firm being taken as the key indicator that such a transition to business modernity has been achieved. The dominance of family business has historically been used, notably by David Landes, to explain poor French economic performance until the second half of the twentieth century, when the deficiencies of such organizations were overcome by a transition to planning in which technically trained business elites replaced dynastic control.13 Charles Kindleberger repeated the accusation that the French family firm “sinned against economic efficiency.”14 Many analysts immediately after the Second World War believed that the dominance of family firms accounted for inadequate investment: one consequence was inadequate capacity for production (e.g., specialty steels) required for France’s military security. Recently the under-capitalization of many German Mittelstand firms has been a prominent issue as banks, stripped of former guarantees from the state, scramble to meet the capital adequacy requirements of Basel II and

Cultural Adjustment to a New Kind of Capitalism?

161

reduce their credit to the Mittelstand sector. The financial limitations that keep family firms small are also generally held to explain why Italian modernization remained incomplete and polarized in a dualism between a large and until recently state-dominated industry and myriad small producers.15 More recently the Economist quoted the Italian economist Francesca Bettio of Siena, that the family is at the root of all Italian ills. “It is responsible for the fact that most Italian companies are small and privately owned; it has contributed to a low female participation rate in the workforce; and it is at least partly to blame for low social and labour mobility.”16 The idea of the family firm is thus at the heart of a debate over the costs of the divergence of a continental European model of capitalism from the “Anglo-Saxon” one. According to this view, path dependence locked continental economies on a suboptimal institutional track. This model was also used to explain failure elsewhere. A perhaps idiosyncratic interpretation of the business experience reflects interwar criticism of the limited investment horizons of British firms. Alfred Chandler in Scale and Scope portrays Britain as still being in thrall to “personal capitalism” in which family relations played a big part. The term refers to a style of management, and not exclusively to the actual structure of ownership. William Lazonick uses a similar diagnosis for the ills of British capitalism.17 One variant of the gloomy depiction of family business holds that such firms may play a useful role in resolving the problems of trust that arise in smallscale local economies, but that large family firms still carry a heavy price.18 The penalty of the family is thus far heavier for larger size enterprises. Such an interpretation escapes from the simple mechanism of Chandler’s account of family firms being a childlike stage in the path to the mature managerial enterprise. It allows interpretations in which economies thrive on the interplay of a dualism between a small family-based sector and large modern enterprise. The real historical picture is almost completely the reverse of the polemical caricature of family firms given by these twentieth-century critical analysts. Rather than being responsible for backwardness, an enhanced importance of family firms was a response to uncertainty, and especially the arbitrariness of the state. In the European past, state inadequacy was compensated by the resilience of other institutions— notably the family. Family firms increased in importance at times of great political upheaval because they were able to provide a basis of trust in a setting when legal norms were in rapid flux: all over

162

Harold James

continental Europe at the time of the French Revolution and the Napoleonic wars, but again also in the aftermath of twentieth century conflicts. German families, for instance, were able to reconstruct businesses after Allied detrustification and decartelization, while new entrepreneurs created or recreated the dynamic Mittelstand enterprises that were at the heart of the German success story. In Italy, a special new bank, Mediobanca, allowed Italian family firms such as Pirelli, Fiat, or Falck to reestablish themselves after 1945. The German and Italian postwar miracles were thus largely driven by family firms, and family business offered a bridge over a political chasm. By contrast, in the United States and the United Kingdom a more politically continuous and stable history meant a reduced demand or need for family firms. The superiority of family firms because of their association with higher levels of trust also characterizes some very dynamic parts of the world, notably in India and China, where the family firm plays a vital role in promoting innovation and entrepreneurship in a politically nontransparent and potentially unstable setting. These are large states that are inefficiently managed, and innovation flows not from state initiatives but from alternative forms of social cooperation, in particular from extended families. These cooperative arrangements can provide a longer term and stable framework for expectations and calculations, and for investment decisions; they can convince both customers and their workforce of their stability, of being in business for the long run. Actually family businesses are quite common in most parts of the world, precisely because of their high degree of institutional resilience in the face of high levels of uncertainty. Over three-quarters of registered companies in the industrialized world are family businesses, and—especially in continental Europe—they include some very large companies. Historians of the subject are often surprised by the extent that the phenomenon has survived into present day business existence.19 According to one recent calculation, 17 of the largest hundred companies in Germany are in family hands, 26 in France, and 43 in Italy.20 France and Italy still consider themselves to be the “champions of family capitalism.”21 In France, at the beginning of the twenty-first century 33.8 percent of the total market value of listed corporate assets was controlled by just fifteen families (and 22.0 percent by five families). For Italy, the equivalent figures are 21.9 percent and 16.8 percent, and for Germany, 25.0 percent and 15.7 percent. By contrast, in the United Kingdom, the equivalent figures are just 6.6 percent and 4.1

Cultural Adjustment to a New Kind of Capitalism?

163

percent.22 That is not to say that there are no long-lived family firms in the United States and Canada—the often eccentric Duponts and the highly secretive Cargills would be obvious examples. But many of the most North American outstanding family dynasties are relatively recent. By the end of the twentieth century there were new opportunities for all types of business, and hence a real and novel entrepreneurial challenge. Firms are much more unstable, and frequently change their locus of activity (Mannesmann, Siemens). Well-established corporate behemoths (ICI, Marconi) can collapse as a result of wrong strategies. Globalization can be seen as a new threat, with disruptions to established markets, massive legal changes. As in the past, family ties can provide a locus of stability. There are also new possibilities arising from: 1. The relative retreat of the state in European economies. 2. The greater activity of capital markets, which freed family businesses from their traditional fear of long-term involvement with and hence dependence on banks. Banks are no longer the major source of business or industrial finance. 3. The internationalization of business activity, which added new possibility for opening up markets, as well as potentially new sources of capital. All three of these considerations enable the family firm to play a greater role than in the past. The most dynamic examples are cross-national and have a capacity for rapid innovation: the iconic Italian examples of fashion in textiles and clothing, Ermenigildo Zegna and Armani, or the German Hugo Boss; in automobiles, Porsche and the Quandtowned BMW; in publishing, Bertelsmann as well as less specific conglomerates Wendel or Haniel. There is certainly no evidence for the model developed by David Landes and Alfred Chandler that dominated most analysis of this subject in the later twentieth century, in which there is a one-way street called modernization that involves a transition from personalized capitalism to the multiple-division publicly quoted corporation managed by a technocratic managerial elite for whom the question of ownership plays little or no role. The twentieth-century rise of the brand, and the perception of an individual relationship required by a good brand, actually favors the family business. Ermenigildo Zegna is an

164

Harold James

embodiment of the virtues of his company in a way that chief executives of big corporations (e.g., Hewlett Packard’s Carla Fiorina and her successor Mark Hurd) sometimes aspired to but in practice almost never could be. This topic is the subject of my recent book, which attacks the Chandler–Landes hypothesis.23 The case studies (one French, one German, and one Italian, chosen to be as nearly comparable as possible over the boundaries of specific national economic cultures) actually point in a different direction. Falck started out as a joint stock corporation, the Acciairie e Ferriere Lombarde, in which Giorgio Enrico Falck was a gifted technician with little capital of his own. By the 1930s he had turned it into a family business. From the 1870s to the 1950s the major vehicles of the Wendel (France) and Haniel (Germany) interests were joint stock companies, Wendel SA and the Société des Petit-Fils de François de Wendel, and the Gutehoffnungshütte AG. Their fortunes started to revive with a turn after the 1960s or 1970s to different enterprise forms, non-quoted holding companies (in Germany, the form of the GmbH offered particular advantages). By the early twentyfirst century there is little surprise when even major companies want to “go private.” The examples of the retreat from the publicly quoted joint stock company are not necessarily to be understood as “history reversing itself,” but rather of a rational response to better developed capital markets. Family ownership has the advantage of being visible and identifiable, in contrast to the anonymous capitalism of large numbers of individual investors or the facelessness of institutional investors. If ownership is an important or even the defining feature of the capitalist process, it may be desirable that it is transparent. The greater difficulties that arise when disposing of ownership in consequence offer a guarantee of continuity, and make property part of a stakeholding and relatively permanent pattern of institutional arrangements, in which there are higher levels of commitment. This means that it may be easier to motivate managers and workers than in a setting when they do not know whether tomorrow the (faceless) owners will walk away. Families in business recently responded to this sort of analysis by developing a concept of “professional ownership.” The family and its long-term vision thus offered a striking and reassuring alternative to the emphasis on “shareholder value” that had been so fashionable in the 1990s, and had been linked with the “Americanization” of business conduct. At the beginning of the twenty-first century much European debate

Cultural Adjustment to a New Kind of Capitalism?

165

focused on whether “Anglo-Saxon capitalism” was not even a greater threat to national self-understandings than family capitalism had been. From a different angle, the CEO of IBM, Samuel Palmisano, recently announced the end of the traditional hierarchical multinational corporation, because of its impedance of information flows, its absence of rootedness, and its tendency to generate antiglobalization backlashes.24 With the collapse of ideas about a “new economy,” the end of the dot.com bubble, and the revelations of corporate corruption and scandal, some European businesses looked for a new model of responsible conduct. Family interests reasserted themselves in continental Europe, especially against managers and executives who seemed to have become too Americanized. The Mohn family ousted the leading manager of Bertelsmann, Thomas Middelhoff, who had wanted to turn a family company into an internationalized or even a denationalized enterprise. In France Jean-Marie Messier was similarly deposed from Vivendi Universal, and the Lagardère dynasty tried to regain control of Vivendi. Unlike Messier, the Lagardères were seen as deeply French. At the funeral of Jean-Luc Lagardère in 2003, the wife of the president of the Republic, the prime minister, and five ministers attended. The left wing newspaper Le Monde published a striking defense of a national form of capitalism: “at a time when the stock exchange keeps on falling, family capitalism, privileging long-term patrimonial interest over short-term operations, seems to have all the virtues.”25 The head of the modern Wendel company (and incidentally long-term president of the French employers’ association Medef), Ernest-Antoine Seillière, recently formulated his distinctive vision of a new French capitalism in a very striking way: “There is no capitalism that is not family. The anomaly of capitalism is the [stock] market.”26 6.4

Conclusion

Can the culture of family capitalism rescue Europe from a deep-seated economic political malaise? Does it offer an alternative way of regenerating growth and economic dynamism? There has powerful potential. But it would be wrong to end on an excessively optimistic note. First, family firms are still the easy targets of blame: as in the contemporary discussion of what is wrong with Italy, or Germany, or in the complaints of the management of the steelmaker Arcelor in resisting a hostile takeover bid from Mittal.

166

Harold James

Second, family firms also reflect strains and tensions in the family, and they are not immune from general trends in society. A consequence of the expansion of the role of the state in Europe, and of the fact that for most of the second half of the twentieth century the etatist economy really seemed to run well, has been a devaluation of the family and a crisis of family values. It is this that might be interpreted as a break with a history in which family firms—small and large—and family values more generally played a crucial role in the economic development of modern Europe. Acknowledgment The author would like gratefully to acknowledge support from the Marie Curie Program MEIF-CT-2006-41674. Notes 1. See Eric L. Jones, Cultures Merging: A Historical and Economic Critique of Culture (Princeton: Princeton University Press, 2006). 2. Martin J. Wiener, English Culture and the Decline of the Industrial Spirit, 1850–1980 (Cambridge: Cambridge University Press, 1981). 3. See, for instance, Jordi Gual, ed., Building a Dynamic Europe: The Key Policy Debates (Cambridge: Cambridge University Press, 2004). 4. André Sapir, Globalisation and the reform of European social model (2005 paper). www.bruegel.org/Repositories/Documents/Publications/working_papers/EN_ SapirPaper080905.pdf. 5. See Hans-Werner Sinn and Gerlinde Sinn, Kaltstart : Volkswirtschaftliche Aspekte der deutschen Vereinigung (Tubingen : J.C.B. Mohr, 1991); Hans-Werner Sinn, Ist Deutschland noch zu retten? (Berlin: Ullstein, 2005). 6. See Henrik Müller, Wirtschaftsfaktor Patriotismus. Vaterlandsliebe in Zeiten der Globalisierung (Eichborn, 2006). 7. Philip H. Gordon and Sophie Meunier, The French Challenge: Adapting to Globalization (Washington, DC: Brookings, 2001: 100). Gouverner des forces qui sont à l’oeuvre dans la mondialisation, Le Monde, June 27, 2000. 8. Quoted in: EU puts reforms on hold to avert French poll disaster, Agence France Presse, March 23, 2005; also in: London’s allure, Financial Times, October 27, 2005. 9. Franz-Olivier Giesbert, La Tragédie du président: Scènes de la vie politique 1986–2006 (Paris : Flammarion, 2006); quoted in: John Thornhill, How Chirac failed France , Financial Times, May 21, 2006. 10. Quoted in: Chrystia Freeland, Blood, sweat and tears—for others, Financial Times, December 9, 1993.

Cultural Adjustment to a New Kind of Capitalism?

167

11. See Alberto Alesina and Enrico Spolaore, The Size of Nations (Cambridge: MIT Press, 2003). 12. Alfred D. Chandler Jr., The Visible Hand: The Managerial Revolution in American Business (Cambridge, MA: Belknap Press, 1977). 13. David S. Landes, French entrepreneurship and industrial growth in the nineteenth century (Journal of Economic History 9: 45–61, 1949). See also Daniel Henri, Capitalisme familial et gestion industrielle au XIXe siècle (Revue Française de Gestion 70 (September): 141–50, 1988; Maurice Lévy-Leboyer, Le patronat français a-t-il été malthusien? (Le Mouvement Social 88: 3–50, 1974). 14. Charles P. Kindleberger, Economic Growth in France and Britain, 1851–1950 (Oxford: Oxford University Press, 1964: 115). 15. Robert Pavan, Strategy and Structure of the Italian Enterprise, Ann Arbor: University Microfilms, 1973. 16. Economist, November 26, 2005, Survey of Italy, p. 6. The concept of Rhineland capitalism as a special path was popularized by Michel Albert, Capitalism versus Capitalism, (London: Whurr, 1993; original French edition 1991). 17. Alfred D. Chandler Jr., Scale and Scope: The Dynamics of Industrial Capitalism (Cambridge, MA: Belknap Press, 1990: esp. 239–40). This interpretation was criticized by Roy Church, The family firm in industrial comparison: International perspectives on hypothesis and history, in Geoffrey Jones and Mary Rose, eds., Family Capitalism (London: Frank Cass, 1993: 17–43); William Lazonick, The cotton industry, in Bernard Elbaum and William Lazonick, eds., The Decline of the British Economy (Oxford: Oxford University Press, 1986: 18–50). 18. Peter Payne, Family business in Britain: An historical and analytical survey, in Akio Okochi, Shigeaki Yasuoka, eds., Family Business in the Era of Industrial Growth: Its Ownership and Management: Proceedings of the Fuji Conference (Tokyo: University of Tokyo Press, 1984). 19. See, for instance, the session of the International Economic History in Budapest, 1982: Leslie Hannah, ed., From Family Firm to Professional Management: Structure and Performance of Business Enterprise (Budapest: Akadémiai Kiadó, 1982). 20. The complex evolution of family affairs, Financial Times, February 3, 2003, p. 6. 21. Frédéric Lemaître, La France: championne du capitalisme familial, Le Monde, April 18, 2003. 22. Mara Faccio and Larry H. P. Lang, The ultimate ownership of Western European corporations, Journal of Financial Economics 65: 393, 2002. 23. Harold James, Family Capitalism, (Cambridge: Harvard University Press, 2006). 24. Samuel J. Palmisano, The globally integrated enterprise, Foreign Affairs, May–June 2006, pp. 127–36 (quote from p. 134). 25. Frédéric Lemaître, La France, championne du capitalisme familial: “A l’heure où la Bourse n’en finit pas de chuter, le capitalisme familial, censé priviligié les intérêts patrimoniaux à long terme par rapport aux opérations à court terme, paraît même paré de toutes les vertus.” 26. Le Monde, July 30, 2002.

7

Venturesome Consumption, Innovation, and Globalization Amar Bhidé

7.1

Introduction

The “techno-fetishism and techno-nationalism” described by Ostry and Nelson in 1995 has drawn strength over the last decade from concerns in the West about globalization. That mind-set incorporates two related tendencies. One is the focus on the upstream development of new products and technologies while glossing over their downstream consumption and use. The other is the belief that national prosperity requires upstream international leadership in upstream activities—“our” scientists, engineers, entrepreneurs, and firms have to be better than every other country’s—they must write more papers, file more patents, and successfully launch more products. Otherwise, competition from low-wage countries like China and India will erode living standards in the West especially as they upgrade their economies to engage in more innovative activities. In this chapter I claim that these two tendencies misconstrue the nature and role of innovation as well as the implications of globalization. I argue that the willingness and ability of individuals and firms to acquire and use new products and technologies is as important as—and in small countries more important than—the development of such products and technologies. Moreover, unlike many individuals and organizations, nations do not have to outperform “competitors” in order to prosper. Notwithstanding the rhetoric about the competitive advantages of nations—a transplant from the domain of interfirm rivalry that has displaced references to old-fashioned comparative advantages—countries are not locked into zero-sum trade. An innovation originating in one country does not impoverish other countries. Rather, it tends to improve standards of living in all countries that have the downstream capacity to acquire and implement the innovation.

170

Amar Bhidé

My interest in the connection between consumption and innovation (and the policy implication thereof) dates back to 1982. As an employee of the consulting firm McKinsey & Co., I worked on a study to help the European Union promote the information technology industry. The focus of the study was entirely on what the European Union could do to help the producers of IT equipment through grants, subsidies, and tax breaks. My efforts to broaden the scope to include the behavior and needs of the users of IT were futile. I was the lowest ranking consultant on the team, and the clients on the EU side had no interest. A Harvard Business Review article (Bhidé 1983) about the importance of the nature of the demand for innovative products that I then wrote had similarly negligible impact. My perspective has subsequently been informed, over the last eighteen years, by my studies of new and emerging businesses that for convenience we may call “entrepreneurial” firms. Along with numerous research associates and students, I have examined, in varying depth and detail, more than five hundred such firms. These studies, including notably, a recent study of more than a hundred US-based venture capital backed firms, suggest that few entrepreneurial ventures—including those characterized as high tech—undertake cutting edge, “upstream” R&D. Rather, they combine (often not in particularly radical ways) and distribute innovations generated by upstream individuals and firms; to borrow terminology from the computer industry, they play the role of “system integrators” or “value-added resellers.” Accordingly for their success—and their much vaunted contributions to productivity— entrepreneurial firms require not just an ample supply of innovative inputs; they also require venturesome and resourceful customers. Many such customers are not in the high-tech sector but are willing to take a chance on innovative products and services. Moreover entrepreneurial firms do not combine and “add value” just to domestic innovations; in an era of growing cross-border flows of ideas and knowledge, the sources of their innovative inputs are increasingly global. Therefore an increased supply of innovative inputs from abroad is a boon, to the entrepreneurial firms, their customers, and to the broader economy. Although I have derived my perspective mainly through an inductive process from my field research, a review of the literature shows that many of the sources are not novel. I will mention these as I go along. Here I will start with especially noteworthy works that have contributed to my discoveries or confirm discoveries that I have independently made.

Venturesome Consumption, Innovation, and Globalization

171

The close relationship between technology adoption and economic development has been examined by several economic historians. Among these are Morrison (1966), Rosenberg and Birdzell (1986)—who argue that the West grew rich first because the resistance to adopting new technologies was weaker there—and Mokyr (1990). By way of more contemporary examples I argue that technology adoption continues to play a critical role. In my argument I also repeat Carter and Williams’s (1964) warning of four decades ago, that “it is easy to impede growth by excessive research, by having too high a percentage of scientific manpower engaged in adding to the stock of knowledge and too small a percentage engaged in using it. This is the case in Britain today.” Similarly, in 1986, Paul David wrote that innovation had become a “cherished child, doted upon by all concerned with maintaining competitiveness . . . whereas diffusion has fallen into the woeful role of Cinderella, a drudge like creature who tends to be overlooked when the summons arrives to attend the Technology Policy Ball.”1 My argument includes country and firm level differences based on their “absorptive capacity” for innovations. The term “absorptive capacity” has been used in the economic development literature since at least the early 1960s to refer to the limited capacity of backward countries to put new investments (and the innovations they may embody) into productive use. Cohen and Levinthal (1989) applied the term “absorptive capacity” to refer to the ability of individual firms to effectively absorb new technologies, and that use of the term has now become commonplace. Although their definition is broad, Cohen and Levinthal and subsequent researchers nevertheless focus mainly on high-tech firms, examining, for instance, how their internal R&D efforts help firms use research produced in university labs. In examining how absorptive capacities matter, I pay more attention to organizations that don’t have any formal R&D efforts and to individual consumers. I also suggest that the use of innovations, like their generation, has unruly entrepreneurial aspects, such as willingness to deal with Knightian certainty. Firms’ venturesome consumption of innovations, like venturesome production, falls outside neoclassical models and (unlike R&D spending) eludes objective measurement. Last, the idea of an “innovation system” comprising many related but different components (instead of a single innovator or a swarm of similar innovators) provides here a useful expositional device. The system as a construct popularized by Richard Nelson (1993) and other

172

Amar Bhidé

scholars accurately reflects how modern innovation really works. Researchers, however, tend to focus on the upstream elements of the system and their linkages, for instance, between university researchers and commercial R&D labs. I would include in such a system the users of innovations who are far removed from university labs and have no internal research programs. 7.2

A Preoccupation with Leadership

Back in the 1980s when the United States was concerned about the “productivity slowdown” and the threat supposedly posed by Japanese manufacturing preeminence, David (1986) observed that “success” in the United States was “equated with ‘leadership’ . . . with pioneering on the technological frontiers. To be an assiduous ‘follower’ seems somehow to have acquiesced in defeat, abandoning adventure for the haven of routine.” Concerns about Japan have receded in the last twenty years, but the preoccupation with technological leadership remains well nourished by rapid growth in China and India. In “America’s Technology Future at Risk” Prestowitz (2006a) writes: “American wealth, economic growth and national security have long been based on technological leadership. . . . [T]he United States has always focused on new technology as the main engine of economic welfare. For more than half a century America’s broad technological leadership has been unchallenged.” Prestowitz is especially concerned about the US position in the telecommunications industry, which “has long been an indispensable element of America’s technological leadership and economic success.” Now, however, the United States is “well on its way to surrendering leadership in advanced telecom products and services.” Prestowitz points to several other alarming developments: In 2005 the United States had a “$55 billion trade deficit in Advanced Technology Products.” Venture capitalists are “pressing the start-up firms they finance to move R&D to Asia. . . . Many telecom and technology companies [have] cut vital R&D spending by 10 to 40 percent. At the same time, government R&D spending in these areas has also fallen by over 30 percent.” “Foreign companies make up the majority of the top ten recipients of US patents each year and the United States has fallen behind the EU and lost ground to Asian countries in the publication of scientific articles. The United States is awarding fewer Bachelor of Science degrees

Venturesome Consumption, Innovation, and Globalization

173

than it did in 1985 and far fewer than Japan, the EU, China, India, and even Korea.” Using different arguments from Prestowitz, the economist Richard Freeman (a former businessman and trade negotiator in the Reagan administration who now runs a think-tank he founded) nevertheless reaches similar conclusions. Freeman’s (2005) NBER working paper, like Prestowitz’s article, asserts the significance of leadership: “leader” or “leadership” appears in the title and five times on just the first page of the paper. Shorn of their qualifying clauses and sentences, Freeman’s concerns can be said to be as follows: “Leadership in science and technology gives the US its comparative advantage” and “in a knowledge based economy, contributes substantially to economic success.” Unfortunately for the United States, its “global economic leadership” is under threat. “Changes in the global job market” are “eroding US dominance in science and engineering.” Freeman predicts that “the erosion will continue into the foreseeable future.” By “increasing the number of scientists and engineers, highly populous low income countries such as China and India can compete with the United States in technically advanced industries” and “undo the traditional ‘North–South’ pattern of trade in which advanced countries dominate high tech while developing countries specialize in less skilled manufacturing.” The evidence that Freeman offers to support his claim that US dominance in science and engineering is and will continue to erode includes the fact that in 1970 “over half of science and engineering doctorates were granted by US institutions of higher education” but that since then, the US share has steadily declined. As shown in table 7.1, European countries produced 7 percent fewer PhDs than the United States in 1975. By 2001, EU institutions granted 54 percent more PhDs, and by 2010, the number of PhDs will probably rise to nearly twice the number granted by US institutions. Japanese institutions granted just 11 percent of the PhDs granted by US institutions in 1975. By 2001 that percentage had more than doubled, to 29 percent. China granted virtually no doctorates in 1975. By 2001, it was granting nearly a third as many as the United States and by 2010, it was expected to grant more doctorates than the United States. Overall, according to Freeman’s projections, the US share of world’s science and engineering doctorates is likely to fall to about 15 percent in 2010. Like Prestowitz, Freeman points to a fall in US shares of scientific publications, patents, and bachelor of engineering degrees and to the

174

Amar Bhidé

Table 7.1 Ratio of number of science and engineering PhDs from universities in selected countries outside the United States to those from the United States

United States

All EU countries

France, Germany, and United Kingdom

1975

1.00

0.93

0.64

0.11

na

1989

1.00

1.22

0.84

0.16

0.05

2001

1.00

1.54

1.07

0.29

0.32

2003

1.00

1.62

0.49

2010

1.00

1.92

1.26

Japan

China

Source: Freeman (2005).

expansion in Asian R&D establishments. “Data on publications and citations by country of investigator show that the US predominance has already begun to drop,” writes Freeman. “In spring 2004, the front page of The New York Times reported a fall in the US share of papers in physics journals while Nature reported a rise in the share of papers in China. The NSF records a drop in the US share of scientific papers from 38 percent in 1988 to 31 percent in 2001 and a drop in the US share of citations from 52 percent in 1992 to 44 percent. The share of papers counted in the Chemical Abstract Service fell from 73 percent in 1980 to 40 percent in 2003.” “Many high-tech companies,” continues Freeman, “have begun to locate major research installations outside the United States.” In 2004, the CEO of Cisco declared that “Cisco is a Chinese company” when he announced that the firm was setting up its newest R&D facility in China. One of Microsoft’s major research facilities is in Beijing. OECD data shows a large increase in United States outward R&D investment from 1994 to 2000. . . . As of mid-2004, the Chinese government registered over 600 multinational research facilities in the country, many from large US multinationals. By contrast, in 1997 China registered less than 50 multinational corporation research centers.” The concerns of Prestowitz, Freeman, and others have apparently resonated with the Bush administration. Noting the “uncertainty” engendered by “new competitors, like India and China” President Bush, in his 2006 State of the Union Speech, announced the “American Competitiveness Initiative” that would (according to a White House Press release) “help the United States remain a world leader in science

Venturesome Consumption, Innovation, and Globalization

175

and technology.” The initiative included proposals to double the federal commitment to “critical basic research programs in the physical sciences”; make permanent the research and development tax credit (to encourage “bolder private-sector initiatives in technology”); and support universities that “provide world-class education and research opportunities.” Policy makers in other advanced countries also apparently subscribe to the thesis and employ similar arguments and rhetoric. In the United Kingdom, the Chancellor of the Exchequer Gordon Brown’s 2006 Budget Statement noted that China and India had “4 million graduates a year to Britain’s 400,000” as well as more computer scientists and engineers. “Every advanced industrial country knows that falling behind in science” he said “means falling behind in commerce and prosperity.” He proposed that the government “do more to support the dynamism and enterprise of business . . . start[ing] with the importance of Britain leading in scientific invention and discovery.” Measures announced in the budget included increasing expenditures on scientific discovery, simplifying allocation of research funding for universities and expanding the scope of R&D credits. The “Lisbon Agenda” of the European Union commits to raise research spending in the member countries to at least 3 percent of GDP. According to a European Commission website devoted to the agenda, “the EU invests less of its GDP in research and development than its main competitors”—just 1.96 percent of its GDP compared to 2.59 percent for the United States, 3.12 percent for Japan, and 2.91 percent for Korea. And Europe “does not have enough scientists and researchers—5.3 per 1,000 workforce compared to 9 per 1,000 in the United States and 9.7 per thousand in Japan.” The proposed solutions to the shortage of scientists include creating a “European Institute of Technology” and making “science a more attractive career option.” The European Union and Messrs. Prestowitz and Freeman can’t both be right about whether or not the United States is “behind” Europe. But putting that aside, why does scientific and technological leadership matter to a country or region in the first place? For Prestowitz, the historical correlation of US leadership with prosperity makes it self-evident that any erosion of leadership must impair standards of living. The economist Freeman relies on models of North–South (or rich country–poor country) trade to reach this conclusion. Freeman’s reasoning may be paraphrased as follows:

176

Amar Bhidé

1. In highly simplified classical or neoclassical models, trade always benefits both parties, but when complications like first-mover advantages or increasing returns to scale are introduced, the models show that gains to one country may come at the expense of another. 2. Trade models also predict that technological advances in a country may help (or hurt) its trading partners depending on the sector in which they occur. In particular, “a country benefits when a trading partner or potential trading partner improves technology in a sector in which the country does not compete but loses when a country improves its technology in country’s export sector. It is good for Alaska if El Salvador improves its technology for banana production but bad for Nicaragua.” 3. North—South trade is mutually beneficial when “the South competes with the North for production of older products through low wages but is unable to compete in the newest technology.” If, however, the South starts competing with the North in “the high-tech vanguard sectors that innovate new products and processes” the South gains at the expense of the North. Thus as China increases its supply of scientific and engineering workers and competes with the United States for upstream innovations instead of just trinkets and toys, the United States begins to lose rather than benefit from trade. (“The loss of comparative advantage can substantially harm an advanced country,” writes Freeman.) Freeman acknowledges that there are models which show that “under some circumstances the loss of technological advantage could benefit the advanced country.” But Freeman dismisses such a scenario as “more of a theoretical curiosum than a realistic representation of the current economic world.” In Freeman’s judgment, the “loss of technological superiority overall is likely to be disastrous for US workers and firms.” Trends like the “multinational movement of R&D facilities to developing countries are harbingers” of the difficult “adjustment problems” that await US workers. But recall Freeman’s evidence about the loss of US shares in scientific publications, citations and patent counts. Look again at the data in table 7.1 on the ratios of scientific and engineering (S&E) PhDs. Toward the end of his article Freeman says that the increase in S&E workers in Europe and Japan is “recent,” whereas the table points to a trend that has been in place for more than two decades. Richard Nelson and Gavin Wright wrote about “The Rise and Fall of American Technologi-

Venturesome Consumption, Innovation, and Globalization

177

Table 7.2 Ratio of PPP adjusted per capita GDP in selected countries and regions to United States per capita GDP

All EU countries

France, Germany, and United Kingdom

Japan

China

$19,830

0.74

0.76

0.72

0.03

$28,090

0.72

0.75

0.80

0.06

1995

$30,165

0.74

0.77

0.83

0.09

2001

$33,983

0.75

0.77

0.77

0.12

2003

$35,373

0.73

0.75

0.74

0.14

United States (GDP per capita, constant $2,000) 1975 1989

Source: World Development Indicators Online.

cal Leadership” in 1992. Table 7.2 shows there has been no decline in US per capita incomes in either absolute or relative terms. PPP-adjusted per capita income in EU countries was about 75 percent of US per capita income in 1975, and the gap has remained more or less at that level since then.2 Japanese per capita incomes reached 80 percent of the per capita incomes in the United States by 1989; after that, relative incomes in Japan have actually fallen a bit. According to “convergence theories,” European and Japanese incomes should have been catching up with US incomes anyhow; with substantially increases in their share of PhDs, scientific articles, and the like, why didn’t Europe and Japan roar ahead? Why, instead, was the growth rate in output per hour over 1995 to 2003 in Europe was just half that in the United States (Gordon 2004)? Japan’s reconstruction and export led boom for nearly four decades after the Second World War also doesn’t sit well with Freeman’s account (see table 7.3). As is well known, Japan grew at miraculous rates as it moved, to use the language of the trade models, from the South to the North, and the composition of its exports changed from low-end trinkets to cutting-edge goods. US per capita income and productivity that started at a much higher base did grow more slowly than Japan’s but grow they did. Indeed prosperity increased in most Western countries, all of which could not possibly have been leaders in science and technology. What accounts for the gap between the “technical leadership is a must” assertion, on the one hand, and many decades of experience, on the other? Are Freeman et al. measuring leadership inappropriately?

178

Amar Bhidé

Table 7.3 Annualized growth rates of PPP adjusted per capita GDP

United States

All EU countries

France, Germany, and United Kingdom

Japan

China

1975–1989

2.52

2.36

2.39

3.25

7.15

1989–2003

1.66

1.76

1.66

1.10

8.00

1989–1995

1.20

1.46

1.62

1.80

9.75

1995–2003

2.01

1.84

1.68

0.56

7.80

Source: World Development Indicators Online.

Or does the problem lie with the models? In the sections that follow, I argue the latter. Although the mathematics behind the models may be sound and industrious researchers may even provide evidence that is consistent with their predictions, to reach the conclusions suggested by Freeman is like walking over a bridge whose design ignores the force of gravity. In the sections that follow I suggest that the North–South models invoked by Freeman omit or mischaracterize vital features of the modern innovation system, especially the role of the users. 7.3

Nondestructive Creation

Just as a devout Hindu might begin a journey with a prayer to the Lord Ganesh, it is obligatory to start a discussion on modern innovation by invoking Schumpeter. The thousands of pages he wrote over more than four decades contained sharp, unequivocal claims as well as tangles of contradictions: Elster (1993) describes Schumpeter as an “elusive” writer who could contradict himself in the course of a single paragraph. Nevertheless, as Rosenberg (1976) puts it, “his model has become the accepted one for all innovative activity.” Even the North– South trade theories that are neoclassical rather than Schumpeterian incorporate the key elements. I do not question Schumpeter’s overall thesis—that innovation drives long term growth—but in my view, the model (or at least the common conceptions thereof) has elements that are inconsistent with the realities of the modern innovation system. First, consider Schumpeter’s (1961) assertion that “a perennial gale of creative destruction” is an “essential fact about capitalism. It is what capitalism consists in and what every capitalist concern has got to live with.” Destruction is the price of innovation: the automobile must

Venturesome Consumption, Innovation, and Globalization

179

displace the buggy makers and mass merchandisers must put the country store out of business. Destructive creation is also central to the North–South models where if the South starts to innovate, incomes in the North are reduced, because for Southern innovators to win, its Northern competitors have to lose. Many innovations do not in fact displace existing products and services because they create and satisfy entirely new wants.3 Indeed, this nondestructive creation represents a critical complement to creative destruction for the following reason: Long-run economic growth of course requires productivity growth. But productivity growth doesn’t just come from improved efficiency—using fewer resources to satisfy our current wants. The creation and satisfaction of new wants can also increase per capita output. For instance, an artist may increase her productivity by developing new techniques that speed up her output of paintings. Alternatively, she may develop a new oeuvre that commands higher prices. She may produce exactly the same number of canvases as before, but provided her work sells at higher prices, her economic output and productivity increases. Moreover the new oeuvre may serve as a substitute for more traditional paintings, so innovator’s productivity gain comes at the expense of the productivity of artists’ who face reduced demand. But it doesn’t have to: the new oeuvre may appeal to completely new sensibilities and find a place on walls that otherwise would have remained bare. Economies cannot in fact sustain increases in productivity and living standards simply through increasing efficiencies in the satisfaction of existing wants. In the short run, increased efficiencies reduce costs and as costs decline, people consume more of the good or service. But eventually the law of diminishing utilities sets in. Sated consumers refuse to buy more even if prices continue to decline. After that, further increases in efficiencies reduce the demand for labor.4 Rather, it is the entrepreneurial activity of creating and satisfying new wants that keeps the system humming. It employs the labor and purchasing power released by increased efficiencies in the satisfaction of old wants. It also creates incentives for continued increases in efficiencies even after demand for old wants has been fully satisfied: producers who satisfy old wants have to keep economizing on their use of labor because they must compete for employees (and share of consumers’ wallets) with innovators who satisfy new wants.

180

Amar Bhidé

The historical record shows that the great prosperity brought about by the Industrial and more recent digital revolutions has both destructive and nondestructive roots. Assuming that this historical pattern is maintained, if the countries of the South start innovating, some of their innovations have the potential to destroy existing businesses in the North. Or they may simply increase the basket of goods we consume and absorb the labor and purchasing power that is constantly being released by improved efficiencies in the production of old goods. 7.4

Massively Multiplayer Game

According to Schumpeter—and, implicitly, in the North–South models —noteworthy innovations are carried out in one shot, by a single innovator. This precludes the development of innovations through the efforts of multiple players located in different places and contributing at different times. But, as we will see in the section below what’s precluded by the Schumpeterian and North–South models is in fact a central feature of modern innovation. 7.4.1 An Evolutionary Process According to Schumpeter, the economically significant innovations that disturbed the “circular flow” were large and spontaneous rather than small and adaptive. They so displaced the “equilibrium point” that “the new one [could] not be reached from the old one by infinitesimal steps. Add successively as many mail coaches as you please, you will never get a railway thereby.” Schumpeter also distinguished such innovations (“carrying out of new combinations of the means of production”) from their antecedent inventions. “The making of the invention and the carrying out of the corresponding innovation,” he wrote, “are, economically and sociologically, two entirely different things.” (Schumpeter 1939) Inventions are “economically irrelevant” as long as they are not carried out into practice (Schumpeter 1934). In his 1976 book, Perspectives on Technology, Rosenberg makes a persuasive case for an incremental process, involving ongoing rather than one-off inventive activity. Innovations don’t appear, “fully grown” and ready for commercial exploitation. Writes Rosenberg: “To date the invention of the fluorescent lamp in 1859, the gyro-compass in 1852, the cotton picker in 1889, the zipper in 1891, radar in 1922, the jet engine in 1929, or xerography in 1937 is to select years in which significant steps were indeed made. But in none of these years was the product

Venturesome Consumption, Innovation, and Globalization

181

concerned even remotely near a state of technical feasibility.” Solving the problems that remain after the initial conceptualization takes “protracted inventive activity.” Or, an invention may be technically feasible, but its economic superiority over existing techniques may require many improvements in its “performance characteristics, often in inconspicuous and unspectacular ways.” Early diesel engines for instance were too heavy for economic operation, and early jet engines had unacceptably low performance characteristics until the development of materials that could withstand high pressures and temperatures (Rosenberg 1976: 72–73). Innovations of the digital revolution, introduced in the years after Rosenberg’s 1976 book, also involve continuous rather than one-shot innovation. Consider the evolution of microcomputers after 1975—the year in which the pioneering Altair was introduced. The current generation of laptops and desktops has come a very long way from the Altair. Altair aficionados derived less practical use from their machines than did the turn-of-the-twentieth-century automobile buffs. Lacking basic input or output devices (e.g., keyboards and printers), Altairs could not even scare horses. Numerous innovations turned this oddity into a ubiquitous artifact. Some of these innovations—the mouse, graphical user interfaces, and electronic spreadsheets—represented conceptual breakthroughs. Others (e.g., word-processing software) were borrowed from mainframes and minicomputers.5 7.4.2 Specialization Schumpeter placed the individual entrepreneur at the center of the innovative process in his early work, but he later claimed that the large corporation would inevitably usurp the entrepreneur’s role. His 1911 book, The Theory of Capitalist Development, credited capitalist innovation to entrepreneurs with the “dream and will to found a private kingdom” and the “will to conquer.” The 1934 work, Capitalism, Socialism and Democracy, placed kingdoms ahead of conquerors. In creating the giant enterprise, Schumpeter now declared, entrepreneurs had eliminated their own function. The “perfectly bureaucratized giant industrial unit” could automatically discover and undertake the “objective possibilities” for innovation. It had “come to be the most powerful engine of progress.” Schumpeter’s claim is said to be one of the most extensively tested in the field of economics but with inconclusive results (Acs and Audretsch 1991).6 In my view, the Schumpeterian hypothesis and the

182

Amar Bhidé

empirical research it spawned misses the point. Different types of organizations produce different and often complementary innovations; therefore comparisons of their contributions can mislead. The distinctive contributions of large and small companies have in fact been discussed by many authors, among them Arrow (1982), Winter (1984), and Acs and Audretsch (1991).7 In my previous work (Bhidé 2000, 2006) I have discussed the comparative advantages of firms not just based on their size but also according to the source of their financing: the public markets, professional venture capitalists, angel investors, and self-financing by entrepreneurs in undertaking initiatives with different capital requirements and novelty.8 Moreover the broad categories of firms and financiers that I have mentioned include heterogeneous organizations.9 The innovation system resembles a rain forest rather than a tree farm or, to use a contemporary metaphor from cyberspace, a massively multiplayer online game rather than solitaire or chess. It contains a multitude of species (or characters) differentiated along many dimensions, rather than say just by their size or shape, or whether or not they have wings.10 Their interactions can be predatory or symbiotic, accidental or deliberate, extended or brief. And, as we will see in a later section, the interactions can extend past national borders. 7.5

Venturesome Consumption

In the North–South trade models—as in all mainstream economic theories—the users of new technologies are at once passive and omniscient. They play no role in the development of innovations, but once innovations appear, users know exactly whether they should buy the offering and what they should pay. In Schumpeter’s model too, the innovator is the star, while those who then imitate or modify have secondary parts. Consumers don’t appear in the cast at all. In fact both the neoclassical and Schumpeterian models fail to do justice to the role of users. In a system where innovations are carried out by a multitude of players, except for the end consumers, the producers of innovations are also users of “upstream” or “adjacent” innovations. Moreover, as I will discuss next, users—including those at the end of the line—often play a venturesome or entrepreneurial role in leading or participating in the development efforts, bearing “unmeasurable and unquantifiable” risks and in resourceful problem solving. Therefore, contrary to the upstream centric view, the willingness and

Venturesome Consumption, Innovation, and Globalization

183

ability of users to undertake a venturesome part plays a critical role in determining the ultimate value of innovations. 7.5.1 Contributions to Development MIT’s Eric Von Hippel has been the leading proponent of the view that innovation often starts with users, particularly the so-called lead users rather than the manufacturers of the product or service. In a 1988 book, which built on his research dating back to the mid-1970s, Von Hippel reported that users had developed about 80 percent of the most important innovations in scientific instruments and most of the major innovations in semiconductor processing. In his 2005 book, Von Hippel writes that “a growing body of empirical work shows that users are the first to develop many and perhaps most new industrial and consumer products.” In the ventures that I have studied over the years, user led innovation does not appear to be the norm, except in the very broad sense that most innovators do put themselves in the shoes of users.11 My recent study of venture-backed businesses does, however, suggest an important role for users, even if they don’t initiate or take the lead in developing new products and services. In interview after interview, we were told of the importance of what in the idiom of the industry is known as “customer engagement” with a few potential purchasers— the so-called alpha or beta users. These users engage in a dialogue with the development team that helps determine the attributes of the product or service that is ultimately sold. For instance, developers may start with the core component of a solution to an important problem faced by potential customers but, in the course of the dialogue with users, learn about complementary functions that must be added to the core to make it work. Or developers may conceive of a product with many functions and then learn that some features add more to the cost of the product than they do to its value. Similarly customer dialogue could help design an effective user interface; as the success of Google’s search engine and Apple’s iPod shows, the so-called look and feel of a product may be as important to its utility as the technical features that lie “under the hood.” For many of our interviewees, an amorphous agglomeration of the many things learned from their interactions with customers that are incorporated into their products, rather than their core idea, was the most valuable source of their intellectual property. Interestingly the ownership of patents does not seem to be the norm in our sample or

184

Amar Bhidé

in the VC-backed companies studied by Cockburn and Wagner (2006). Our interviewees claimed that expense of filing patents was not worth it because even if the costs of filing were not very high, the costs of enforcing them would be prohibitive, so there was no point in incurring any filing costs. These claims in conjunction with the nature of the development process suggests that the companies might not have been in a position to file strong patents: the core idea was not particularly novel or obvious, while the truly valuable intellectual property, developed together with customers and embedded in the optimal combination of features, user interfaces, and so on, was too amorphous to be captured in a patent.12 7.5.2 Bearing “Unmeasurable and Unquantifiable” Risks According to Knight (1921) the essence of entrepreneurship involves responsibility for uncertainty—facing unmeasurable and unquantifiable risks rather than betting on situations (as in a casino) where the odds have been well established by many prior trials. But it is not just the producers of an innovation who must confront Knightian uncertainty—purchasers (who may have no role in its development) also cannot form objective estimates of their risks and returns. One source of uncertainty lies in whether or not the innovation actually does what it is supposed to do. A product that works in the lab or with a few beta sites may not work for all users because of some unexpected difference in the conditions of its implementation. In addition a product that works fine at the outset may fail later on. An innovation, like a theory, can never be proved to be good—at any moment we can only observe the absence of evidence of unsoundness. Repeated use of a product may surface hidden defects that cause malfunctions, increase operating costs, or pose health and safety hazards to the user or the environment. Unanticipated technical failures injure both the developers and the users of the innovation. Indeed users may face greater exposure. In many products and services, failures can cost users many times their purchase price. Defects in a word-processing or email package that costs just a few hundred dollars can wipe out many years of invaluable files and correspondence. Or even where the data are not lost, the costs of transferring the files to a new software package—and learning how to use the package will tend to be substantial. Similarly a defective battery in a laptop can start a fire that burns down a house—this actually happened to a friend. Tires that wear badly can have fatal conse-

Venturesome Consumption, Innovation, and Globalization

185

quences. A security hole in its servers can cripple an online brokerage, and the belated discovery of the hazards of asbestos can lead to tens of billions of dollars of removal costs. Consumers’ investments in products that work perfectly well for them may also be impaired if they fail to attract a critical mass of other users. As that becomes evident, vendors (and providers of complementary add-ons) will abandon the product and stop providing critical maintenance, upgrades, and spare parts. Or the vendors will go out of business. This has been commonly seen in the IT industry. Customers have been left stranded when upgrades and new releases fail to have “backward compatibility,” and when new technology makes products obsolete. As mentioned, many innovations aim to satisfy new wants rather than just provide a substitute for existing products. Customers then face Knightian uncertainty about whether and by how much the value they derive from the innovation will exceed its price. In the schema of neoclassical economics, consumers have a gigantic, well-specified utility function for all goods, extant as well as un-invented. Therefore when an innovation that creates a new demand (or a new combination of demands) appears, wise consumers consult their utility functions as they would a tax table to learn exactly what that may be worth to them. As it happens, there is, to my knowledge, absolutely no empirical basis for such utility-maximizing assumption. The evidence from behavioral researchers like George Lowenstein instead points in exactly the opposite direction: people don’t have a clue about the value of things they have not experienced before. When researchers ask subjects how much they would pay for some novel experience, such as kissing their favorite movie star, they receive whimsical responses, anchored to some irrelevant piece of data just planted in the subjects’ minds by the researcher, such as social security numbers. One interpretation of these behavioral experiments is that people are irrational; the other is that they simply don’t know, and blurt the first thing that comes to mind to earn their $5 for participating in the experiment. Behavioral research has been criticized for experiments where the subjects, unlike actors in the real world have no stake in the outcome, but in this instance the experiments really do seem to correspond closely to reality. It is highly improbable, for instance, that anyone who wears glasses or uses contact lenses has a firm grasp of the economic value of (successful) corrective laser surgery or someone who has a conventional TV set has a good measure of the additional value of

186

Amar Bhidé

switching to the sharper images provided by a digital product. Indeed I am skeptical that people who actually have laser surgery or buy a digital TV set ever quantify the value. Before or after, the enhanced utility is as much a shot in the dark as the value of the pleasure Lowenstein’s subjects anticipate from kissing movie stars. I personally have not seriously considered either laser surgery or buying a digital TV set, but I have been enticed by the latest in personal computer hardware and software for more than two decades. I have no idea of the value of any of any of the numerous upgrades I have experienced (or for that matter, a good estimate of the time and opportunity costs I have incurred in the course of these upgrades.) Similarly, although I have worried about—and periodically endured—the consequences of technical defects and abandonment of favorite programs by vendors, I have never actually made any effort to quantify the probability distributions. I cannot even imagine being able to enumerate all the dire possibilities. Similarly people who have corrective eye surgery may ask about the probability that something might go wrong, so that the operation won’t give them 20/20 vision. But what basis could they possibly have for evaluating the consequences twenty or thirty years later? I suspect that most don’t even try. Organizations who purchase expensive systems do often expend many worker-years of effort to evaluate the costs and benefits. For example, as of this writing, Columbia Business School is in the process of acquiring a new “courseware” platform. A committee has been formed, long Requests for Proposals have been issued, short lists have been made, vendor proposals have been studied, consultants have been retained . . . , but for all the effort and the availability of the finest analytical minds for the exercise, the value of the new courseware will remain elusive. The monetary value of enhancing the student satisfaction and learning, saving faculty time and so on can only be a blind guess. Similarly, although the out of pocket costs of purchasing a system will play a role in picking a vendor, the magnitude of the much larger “all in” opportunity costs (e.g., the time of faculty and staff) of switching to any new courseware platform are unfathomable. Moreover the courseware contemplated for Columbia encompasses a relatively small number of functions and users. The systems used by large multinationals are vastly more complex; and as described in appendix, when it comes to Enterprise Resource Planning (ERP) software the problem of quantifying costs and benefits is even more intractable.

Venturesome Consumption, Innovation, and Globalization

187

7.5.3 Resourceful Problem Solving Innovators often face situations that require “resourceful problem solving” in the following sense: although the situation may be similar to ones the innovator has faced before, it also contains novel elements, so the innovator cannot simply repeat what has worked in the past. Experience (or human capital) that we may think of as the accumulated knowledge of similar past situations helps, but this is not enough. An innovator is more than just a dexterous and knowledgeable surgeon performing difficult but routine arthroscopic knee surgery. The innovator must also act resourcefully in the face of novel situations with a can-do attitude, imagination, willingness to experiment, and so on. Although consuming something novel requires coping with the Knightian uncertainty about its utility, it does not necessarily require resourceful problem solving. Drinking a new soft drink or showing up for an appointment for corrective surgery is not especially demanding. Other kinds of consumption such as assembling a model airplane may require patience, dexterity, and experience but, as long as the instructions are clear and complete, do not require resourcefulness or creativity. Indeed creative deviations from the prescribed instructions can lead to undesirable outcomes, but not all innovations come with clear and complete instructions. Many high-tech products, especially those with complex architectures and features, very rarely do, and deriving any utility from them requires a great deal of resourceful problem solving. Manuals for Windows-based personal computers and software, for instance, are famously bewildering. This is not simply because of the incompetence of the authors of the manuals. In considerable measure, the sometimes bewildering instructions reflect the complexity of the internal architectures of the systems, the many options and features they contain, and the difficulty of anticipating how the components will interact. But whatever its cause, my experience has been that the alluring features of new products rarely work “out of the box” simply by following the instruction manual. I have spent countless hours to get new gizmos to work or to stop inexplicable crashes. And the toil is far from mechanical: I have to guess what might be wrong, conduct experiments, and troll through postings of user groups on the Internet trying to find solutions to similar problems. To be sure, writing and testing the code that I developed for a real-time trading system some years ago required less resourceful problem solving than what I have often needed to get a new software package to work.

188

Amar Bhidé

The challenges multiply for innovations used by groups rather than individuals. For instance, as described in the appendix, the effective use of complex enterprise software involves more than just resourceful engineering; the organizational problems can in fact be far more daunting than the technical problems. 7.6

Trading Assumptions

Baumol (2002) suggests that a free market system of innovation provides a positive but small share of the gains to the innovator whereas users get the rest. The proposition makes intuitive sense but is difficult to prove. The profits of the producers can provide at least a crude handle on what they get, but we cannot directly observe the “surplus” secured by the users. And for the reasons already discussed, estimates of the value they derive from their consumption are highly problematic. Researchers have tried several ways of getting around the problem, and although the estimates vary with the method used and the industry studied, they all support the Baumol conjecture, that consumers rather than the producers secure the lion’s share. For instance, Nordhaus (2005) analyzed data for the non–farm business economy and for major industries in the United States. He finds that producers captured a “miniscule” fraction of the returns (of the order of 3 percent) from technological advances over the 1948 to 2000 period, “indicating that “most of the benefits of technological change are passed on to consumers.” Other studies reporting or implying large consumer surpluses include Mansfield (1977), Bresnahan (1986), Trajtenberg (1989), Hausman (1997), and Baumol (2002). The critical question for the purposes of this article is whether and how it matters if the producers of upstream innovation are located abroad rather than at home? Do consumers benefit from innovations that originate abroad, or could they as in the North–South models, invoked by Freeman, actually suffer harm? “Economists worry about another place owning the very next big thing—the next groundbreaking technology,” Stanford’s Dan Siciliano told Kronholz (2006). “If the heart and mind of the next great thing emerges somewhere else because the talent is there, then we will be hurt.” Are such concerns well-founded? An important determinant of whether or not innovations abroad help or hurt consumers at home depends on whether the innovations are internationally tradable. If innovators are able and willing to sell

Venturesome Consumption, Innovation, and Globalization

189

their innovations to users everywhere at the same low price compared to the value, it would not matter a great deal where the innovation originated. In fact, if international financiers provide the capital and share in the returns, the location where the innovation originates would be inconsequential. Suppose, however, that innovators export the products that embody their innovations and not the innovations. Now the country of origin secures both the profits from the innovation and the wage income associated with the production of the goods and services. Conversely, the receiving country has to generate exports not just to pay for the value of value of the innovation but also for the costs of its production. Such are the assumptions embedded in the North–South models that Freeman relies on to predict “disastrous” consequences for US workers from the loss of the US lead in cutting-edge research and technical development. In these models, upstream innovations do not cross national borders. Intermediate goods and services also don’t exist or cannot be traded. Only goods and services for final consumption cross national borders.13 Moreover imports of innovative products lead to the creative destruction of domestic businesses and reduce the purchasing power of local consumers. But how realistic are these assumptions? I have already argued that innovations don’t always destroy. Next I will suggest that actual patterns of international trade are quite the opposite of the assumption in the North–South models that final goods are freely tradable but nothing else is. Upstream innovations (and their associated goods and services) often move more easily across national borders, especially of advanced countries, than midstream or downstream innovations, goods, and services. Moreover a large proportion of downstream activity in an advanced economy is not traded at all—it is both produced and consumed in the same place. In pre-industrial times, monarchs sometimes took extreme measures to prevent the export of the distinctive know-how of domestic craftspeople, but even then there was a fair degree of cross-border learning. For instance, to realize his dream of making Russia a naval power, Peter 1 the Great personally studied ship-building in Deptford (in Britain) and in Amsterdam. While in Amsterdam, Peter 1 worked for four months in the largest private shipyard in the world, belonging to the Dutch East India Company. He then hired many skilled shipwrights and seamen and took them back to Russia.

190

Amar Bhidé

In modern times most countries impose restrictions on the export of sensitive technologies, but otherwise technology moves across borders without much let or hindrance. Advanced countries that lead in some sectors and technologies import technologies in other sectors. The United States, which has, according to Prestowitz, Freeman, and others, long been the “overall” leader in science and new technology, has also benefited from technologies developed overseas. As Carter and Williams (1964) wrote: “All advanced countries draw on the research and development results of other countries, freely or by payment of licence fees or through foreign subsidiaries.” In 1960 France “paid abroad” 273 million francs and received 63 million. In 1963 Germany paid 540 million marks and received 200 million. Even the United States, which enjoys undisputed technological leadership and receives more than it has paid abroad, has both imported and exported a “great deal of technical knowledge.” Eaton and Kortum (1995) examined the growth in productivity in West Germany, France, the United Kingdom, and the United States between 1950 and 1990. According to their analysis, the growth of the first three countries that started far behind the United States at the start of the period was “primarily the result of research performed abroad.” Moreover, notwithstanding its overall lead, “even the United States obtain[ed] over 40 percent of its growth from foreign innovations.”14 Certainly not all innovations travel easily across borders. According to David (2003) innovations are “most efficient as elements of a production system when they have been designed for a specific environment.” As mentioned, my research suggests that VC-backed businesses do indeed expend considerable effort in an iterative dialogue with customers in order to determine an optimal bundle of functions, interfaces, and so on. Variations in local conditions naturally affect optimal bundles, so products that are well suited to one country may be inappropriate for another. But such problems are usually less severe with scientific knowledge and upstream innovations, where Freeman and others are most concerned about leadership, than with midstream or downstream innovations. Scientific knowledge and upstream innovations tend to be relatively simple and universal. As proximity to endusers increases however, so does the complexity and localization of innovations; moreover the localization (and often the complexity) grows as innovations evolve over time. To illustrate: Sir Timothy Berner-Lee’s path breaking invention of the core technology of the World Wide Web had no problems from moving

Venturesome Consumption, Innovation, and Globalization

191

out of the CERN lab in Switzerland to anyone anywhere with an Internet connection. Browsers that were built around the core technology acquired some localization, mostly in the language used in the interfaces: a browser with English language menu was of limited use to someone who only spoke Swahili. Many subsequent Web-based applications then became extensively tailored to local conditions: e-commerce applications, for instance, reflect local shopping habits, and the business practices of local banks, credit card companies, merchants, and privacy regulators.15 The e-commerce example incorporates another feature of advanced economies because of which innovations abroad—be they up-, mid-, or downstream—don’t seriously threaten the incomes and purchasing power of consumers at home. As is well known, services account for much of the consumption and output in high wage countries. Moreover within the services sector, according to a McKinsey and Co. (2005) projection for 2008, no more than 11 percent of the 1.46 billion service jobs world wide could even theoretically be performed in an overseas location. The McKinsey estimates also projected that actual off-shore employment in 2008 would amount to just 3 percent of the theoretical maximum or less than one-quarter of 1 percent of total service jobs world wide.16 Since, like Willie Sutton, innovators and entrepreneurs tend to go where the money is, it is not surprising that much of their attention in advanced economies has been directed to improving the productivity of the service sector. Indeed, because most such services are domestically produced and consumed, innovations that improve the productivity in one country do not have much of an impact on other countries. An innovative e-commerce application that improves the efficiency of retailing in the United States doesn’t reduce the well-being of the Japanese, and the development of a better hospital management software system in Sweden doesn’t hurt the United States. Similarly, if the e-commerce application is eventually adapted for Japanese use and hospital software for use in the United States, this doesn’t hurt the countries where these innovations originated either. Unlike most services, material goods can and increasingly are being produced far away from where they are consumed. But there does not seem to be much of connection any longer between the locations where innovations occur and the sites where goods are physically produced. Rather, multinational companies design global supply chains, where factors such as wages, skills and distance from the market determine

192

Amar Bhidé

the placement of the individual links. For instance, the Singapore-based Creative Technology, Ltd. invented a hard drive MP3 music player, which it started selling in January 2000 as the Nomad Jukebox. About two years later, in October 2001, Apple introduced the competing iPod (which Creative alleged infringed on its patent), and the iPod soon displaced the Nomad as the market leader. But most of the production of MP3 players takes place in mainland China, not in the United States or Singapore. Similarly in recent decades the process and product innovations of Japanese car companies have allowed them to substantially increase their share of the US market at the expense of the market shares of US companies. But note: although the innovations have largely originated in Japan, the car companies have increasingly moved the production of cars for the US market to plants located in the United States. In other words, not only have consumers in the United States benefited from innovations originating in Japan that lowered prices and improved the quality of cars, the wage income derived from the manufacture of such cars has increasingly shifted to the United States. Thus, even in many manufacturing sectors, the “make where you innovate” assumption of the North–South models does not seem to be the norm. 7.7

Explaining the US Lead

Let us return to the question raised earlier: Why, contrary to “convergence” theories, and despite the alleged erosion of its lead in science and cutting-edge technologies, has the United States maintained its lead in per capita incomes vis-à-vis Europe and Japan? The analysis above suggests that the exceptional entrepreneurial capacity of firms and individuals in the United States to take advantage of upstream innovations regardless of where they might originate, has helped maintain the US lead. (From this perspective, the historical primacy of the United States in many scientific and technological fields may be more a by-product rather than a cause of US prosperity. Just as the rich make larger contributions to the arts than the not so well-off populace, prosperous countries are more likely to contribute to research on string theory or the decoding of the genome than poor countries.17 And as prosperity becomes more widespread, more countries contribute to the world’s stock of scientific knowledge. This helps rather than hurts the countries that once took the main responsibility. Venturesome consumption of innovations in IT—a sector that, according to Nick Bloom, Raffaella Sadun, and John Van Reenen (2005),

Venturesome Consumption, Innovation, and Globalization

193

has accounted for much of the acceleration of US productivity since 1995—appears to have played a particularly important role in maintaining the US lead. As mentioned, Prestowitz (2006a) sees the US trade deficit in IT and other advanced technology products as a symptom of a faltering economy. In my interpretation, this deficit is an indicator of economic strength and dynamism, not weakness. As we will see next, the United States has a voracious appetite for IT goods and services, many of which are made in China, Taiwan, and other countries where wages and manufacturing costs are relatively low. A high propensity to use IT generates deficits but also, as the evidence we will review suggests, enhances productivity in IT-using industries that account for a much larger share of economic activity than the IT industry itself. 7.7.1 Consumption Patterns Tables 7.4 and 7.5 give data on the sales of Windows, Linux, and other operating systems for personal computers and servers by selected regions and selected countries for 2001.18 To construct this table, sales of operating systems (as available, in terms of units and revenues) were divided by GDP for the region or country. The sales to GDP ratios were then scaled by the sales to GDP ratio for the United States. The tables show that Windows operating systems sales to GDP ratios for western Europe were about 25 percent lower than in the United States and in Japan more than 10 percent lower. Within western Europe only two countries, Sweden and Denmark, had higher sales to GDP ratios than the United States. The GDPs of these two countries are, however, small, so their total sales amounted to about one-twentieth of US sales. Among the relatively large west European countries, GDP to sales ratios were about a third lower than the US ratios in Germany, France, and Italy and about 10 percent lower in the United Kingdom. The United States isn’t the leading user of IT in every category. As has been pointed out by Prestowitz (2006) and many others, the United States has become a laggard in broadband deployment. As of 2005 the United States was behind 15 other countries in terms of the number of broadband subscribers per 100 inhabitants. Similarly Prestowitz cites data showing that the United States is 42nd in the world in cell phone usage: In 2003 cell phone subscriptions per 100 inhabitants in the United States were almost half the subscriptions in Italy. The United States has also lagged behind Korea, Japan, and many countries in Europe in the deployment of 3G high-speed wireless data systems. But these exceptions apart, the US leads in terms of overall IT consumption, just as it does in the purchases of operating systems.

36

23 NA

96

38

NA

NA

Central/eastern Europe

Middle East and Africa

Japan

Asia Pacific excluding Japan

57

NA

NA

36

91

73

88

24

34

NA NA

NA NA 87

36 NA

102

100

58 NA

116

Linux

80

80

NA

NA

55 NA

115

100

All systems

175

152

118a

139

123

116

121

100

Gross fixed investment/GDP

Sources: GDP data from World Development Indicators Online; operating system sales data are generously provided by Pankaj Ghemawat and Ramon Casadesus-Masanell (as described in the text); gross fixed investment to GDP ratios from EIU database. a. For MEA gross fixed investment, calculation is the weighted average of two regional aggregates provided by EIU: Middle East and North Africa and sub-Saharan Africa.

NA

54

65

60

74

Latin America

Western Europe

137

100

Canada

100

106

100

141

United States

100

Windows OS

All systems

Windows OS

Region

Linux

Revenues/GDP

Units sold/GDP

Table 7.4 Ratios of sales of operating systems (in units and revenues) to GDP and of gross fixed investment to GDP in 2001 (US ratios = 100)

194 Amar Bhidé

Venturesome Consumption, Innovation, and Globalization

195

Table 7.5 Ratios of sales of operating systems (in units) to GDP and of gross fixed investment to GDP in 2001 (US ratios = 100) Units sold/GDP Gross fixed investment/GDP

Country

Windows OS

Linux

All systems

United States

100

100

100

100

71

53

69

136

Austria Belgium

86

65

83

125

Denmark

127

92

123

122

Finland

100

73

97

125

France

68

65

68

120

Germany

65

63

64

123

Greece

72

36

68

146

Ireland

92

56

88

143

Italy

63

42

60

125

Netherlands

93

80

92

130

Norway

87

62

84

112

Portugal

94

54

89

163

Spain Sweden

47

29

45

160

117

91

114

106

Switzerland

92

73

90

137

United Kingdom

87

93

87

102

Sources: GDP data from World Development Indicators Online; operating system sales data are generously provided by Pankaj Ghemawat and Ramon Casadesus-Masanell (as described in the text); gross fixed investment to GDP ratios from EIU database.

Table 7.6 compares total IT expenditures (as estimated by the Gartner Group in its Market Data books) across selected regions and countries for 2001 to 2004. As in the previous table, total expenditures are divided by GDP and this ratio is compared to the US expenditure to GDP. Here too we find that the IT expenditure to GDP ratio in western Europe is between 15 to 20 percent lower than in the United States and in Japan it ranges from 10 to 30 percent lower than in the United States over these four years. 7.7.2 Productivity Gains Until not so long ago, there was debate about whether the IT investment had done the US economy much good. In 1987 Robert Solow wrote: “We see the computer age everywhere, except in the productivity statistics.” The following year, Steven Roach (1988) of Morgan

83

83

98

64

71

83

Latin America

Western Europe

Central/eastern Europe

Middle East and Africa

Japan

Asia/Pacific

175

85

68 74

118a 152

105

81

89

90

100

139

123

116

121

100

193

133a 155

144

129

123

130

100

GFI ratio

88

76 84

90

81

93

90

100

IT ratio

2003

200

133a 152

143

127

120

130

100

GFI ratio

88

81 87

83

84

98

88

100

IT ratio

2004

198

123a 143

134

122

121

126

100

GFI ratio

Source: IT spending estimates from Gartner Dataquest Market Databook for December 2005 and December 2003, GDP data from World Development Indicators Online, Gross Fixed Investment to GDP ratios from EIU database. a. For MEA gross fixed investment, calculation is the weighted average of two regional aggregates provided by EIU: Middle East and North Africa and sub-Saharan Africa.

90

100

United States

Canada

IT ratio

IT ratio

Region

GFI ratio

2002

2001

Table 7.6 Ratios of IT expenditures to GDP, and gross fixed investment (GFI) to GDP, where US ratios = 100

196 Amar Bhidé

Venturesome Consumption, Innovation, and Globalization

197

Stanley dubbed this the “productivity paradox.” In 1991 the Economist pointed out that the returns from IT investments were so low that firms “would have done better to have invested their money in almost any other part of their business.” On the other side, Paul David (1990) argued that it was too early to judge the value of computers—it took decades for the productivity benefits of the electric dynamo (whose technical development had been largely completed by 1880) to be realized. Griliches (1994) and others also suggested that the productivity paradox reflected a measurement problem: users realized many benefits from IT in forms such as greater variety, convenience, or quality that are missed in standard GDP accounting. By 1996 the annual IT spending of US firms had crossed half a trillion dollars and organizations like Nations Bank had annual IT budgets of $2 billion a year (Lucas 1999). Yet Gordon (1999) suggested this was all for naught so far as the economy was concerned. According to Gordon, the acceleration of productivity in the United States that occurred after the mid-1990s was almost entirely due to more efficient production of IT, particularly computers, rather than to the use of IT. Jorgenson and Stiroh (2000) came to similar conclusions: the acceleration of productivity growth after 1995 could be “traced in substantial part” to improvements in the production of IT. On the user side there had been some “capital deepening”—as computers got cheaper, firms bought more of them. The evidence was clear, they wrote, that “computer-using industries like finance, insurance, and real estate (FIRE) and services” had “continued to lag in productivity growth” despite their “massive high-tech investment.” As someone who studies entrepreneurs and businesses from the ground up, I find the Jorgenson and Stiroh claim puzzling. First, by the year 2000, computer-using industries had undergone vast changes for over nearly two decades. The rise of new big-box retailers, notably Wal-Mart, had displaced many traditional players. Similarly small regional banks had been merged into mega-sized national institutions like Citicorp and the Bank of America, or had disappeared. With the new players came new ways of doing business—Wal-Mart established global supply chains and regional distribution centers, for instance. The old players and ways of doing business didn’t fall like trees stricken with Dutch elm disease. The old order was pushed out by the new after a competitive struggle in the marketplace. How could this have happened unless in some way, shape, or form the new was more “productive” than the old?

198

Amar Bhidé

Second, could business facing relentless pressure from rivals and the capital market simply have thrown away the greater part of half a trillion dollars a year in IT spending? I once suggested (Bhidé 1986) that banks and other financial institutions overestimate the sustainability of the competitive advantages they can derive from investing in technologies because they overlook the possibility of imitation: if one bank builds ATMs that customers value, others soon will as well. But, this simply means that customers, not the banks derive most of the benefit. For the technological “race” to have no productivity benefit, competitors must all invest in innovations that even customers don’t value— much of the half a trillion dollars invested in IT would have to have been totally wasted. Third, could IT-using industries really improve their output per worker just from capital deepening, without finding new ways to use that capital? It is one thing to say that as the costs of vacuum cleaners and dishwashers decline, less labor will be required to clean floors and dishes because more capital equipment will be used. But I don’t think that my writing will go faster if I had two computers instead of one (or got a computer with twice the processor speed as my current model). The idea of simply plugging more IT gear into the existing production function of a large organization becomes particularly far-fetched in light of the extensive reengineering that the implementation of new systems actually entails. Finally it is surprising to find that the models assume perfectly competitive product and factor markets. If this were ever the case, why would anyone try to develop productivityenhancing techniques in the first place? Does not the assumption preclude the phenomenon?19 In any event, the debate seems to be largely over now. As Dedrick, Gurbaxani, and Kraemer (2003) conclude after reviewing the literature, “the productivity paradox as first formulated has been effectively refuted. At both the firm and the country level, greater investment in IT is associated with greater productivity growth.” Moreover at the firm level, “IT is not simply a tool for automating existing processes, but is more importantly an enabler of organizational changes that can lead to additional productivity gains.” The literature also suggests that the United States is at the forefront of realizing the productivity gains. Bloom, Sadun, and Van Reenen (2005) write that whereas in the United States significant gains in productivity have been realized in sectors that intensively use and produce IT, the EU has seen similar productivity acceleration just in the IT-

Venturesome Consumption, Innovation, and Globalization

199

producing sector. In sectors “that use IT intensively,” such as retail, wholesale, and financial intermediation, European countries have not enjoyed the “spectacular levels of productivity growth” as in the United States. “Britain has done better than France or Germany, but not as well as the United States.”20 Similarly Robert Gordon, who in 1999 had written that there had been virtually “no productivity growth acceleration in the 99 percent of the [US] economy located outside the sector which manufactures computer hardware,” was subsequently persuaded that “[a]fter fifty years of catching up to the US level of productivity, since 1995 Europe has been falling behind . . . studies of industrial sectors suggest that the main difference between Europe and the United States is in ICT-using industries like wholesale and retail trade”(Gordon 2004). 7.7.3 Venturesomeness of IT Users The significantly higher propensity of US businesses to buy IT does not seem to reflect a generally higher propensity to invest in fixed capital, because of some general economywide factors like low interest or discount rates or tax breaks for long term investment. Tables 7.4, 7.5, and 7.6 on operating system sales and total IT expenditures also contain columns on the ratios of total gross fixed investment (in all categories) to GDP. These columns show that the United States lags in overall fixed investment by about the same degree as it leads in IT spending. The ratio of gross fixed investment to GDP is about 50 percent higher in the Japan than in the United States and in western Europe about 25 percent higher. Table 7.5 also shows that there is not a single country in Europe where the ratio of gross fixed investment to GDP is lower than in the United States.21 My hypothesis is that the comparatively high propensity of the United States to buy IT (which is available throughout the world at roughly similar prices) despite a low overall rate of investment reflects an exceptional level of venturesome consumption: First, buyers of IT in the United States are willing to take their chances on novel technologies where no one has much evidence on the risks and the returns. Large corporations, run by the book with the help of squadrons of financial analysts, will spend tens of millions of dollars on enterprise software based on the crudest of guesses of the costs and benefits. I have no systematic knowledge about buying habits outside the United States, but at least some of the vendors of enterprise software we have interviewed said that European IT staffs tend to be risk averse

200

Amar Bhidé

and prefer more stable, older generation products. Second, US purchasers of IT may be more bold and resourceful in making the organizational changes needed to derive the full benefit of ERP and other such systems. The venturesomeness of IT users in the United States—and the large size of the market—helps attract suppliers. According to the entrepreneurs we have been interviewing recently, the United States is the market of first resort for most IT vendors. Indeed the attractiveness of the US market had caused some of our interviewees who had started their businesses in Europe to relocate to the United States to be close to their target customers. And to the extent that suppliers refine their products through a dialogue with US customers, their features are optimized to the US market. This in turn makes the product more attractive to US customers rather than to customers outside the United States; it also creates an incentive for suppliers to continue to focus their sales efforts on US customers (because selling outside the United States might require additional costs to adapt the product to local conditions.) Bloom, Sadun, and Van Reenen’s (2005) research suggests that the unusual willingness to invest in IT and the capacity to derive value from such investments is sufficiently embedded into the structures and routines of US-based companies that they carry this capacity to their operations outside the United States. They compared the establishments based in the United Kingdom that were owned by US multinationals, non–US multinationals, and domestic (i.e., UK-based) companies. They found that establishments owned by US multinationals invested about 41 percent more per employees in IT than the average for the industry; non–US multinationals about 20 percent more, while domestic companies invested about 15 percent less than the industry average. US multinationals also apparently got more for their IT buck; they enjoyed “significantly higher productivity of IT capital” —this effect accounted for “almost all the difference between the overall productivity of resources used by US-owned and all other establishments.” And, they found that that the “IT edge” of US multinationals was “confined to the same ‘IT-using intensive’ industries that largely accounted for US productivity growth acceleration since the mid-1990s.” I should emphasize that success of IT-using industries in the United States has not in any material way reduced incomes in the rest of the world, since their outputs are usually not tradable exports. When US

Venturesome Consumption, Innovation, and Globalization

201

multinationals use their productivity advantages to establish operations abroad, they may wipe out local firms but their activities (e.g., those of Japanese owned automobile plants in the United States) generate wage incomes and consumer surpluses that remain almost entirely with their host countries. And investors abroad have the same opportunity to share in the profits of US multinationals (by buying their shares) as do US nationals. I should further note that although developed countries in Europe and elsewhere may be somewhat behind the United States in using IT, they are well ahead of emerging and underdeveloped countries. My interviews suggest that despite the economic boom in China and other Asian countries, after the United States, the European market is the next port of call for most IT innovators. And even if products are initially optimized for the United States, potential users in Europe are not so very different that the products cannot be “Europeanized” or that Europeans cannot learn to live with products developed for the United States. Arguably, the superior capacity of Europe to use advanced products vis-à-vis most other countries has mitigated the effects of what many consider to be dysfunctional public policies. To slightly extend Adam Smith’s adage, there is much ruin in a nation once its capacity to use innovative goods and services is sufficiently advanced. 7.8

Elusive Underpinnings

In the Parente and Prescott (1994) model, all countries draw their technologies from a common pool that keeps getting bigger and better; however, the investments that firms have to make to take advantage of the ever-improving pool depends on the “barriers to technology adoption” in their countries. Parente and Prescott mention “regulatory and legal constraints, bribes that must be paid, violence or threats of violence, outright sabotage, and worker strikes” as examples of the forms that the barriers can take, but that is not their focus. Rather, according their theory, whatever form the barriers may take, the differences in their magnitude do not have to be “implausibly large” to account for the “huge observed income disparity” across the rich and poor countries. But what form do these barriers in fact take and why do they vary across countries?22 Nelson and Phelps (1966) offer the general (i.e., economywide) hypothesis that “education speeds the process of technological diffusion.”23 Although education surely matters, it cannot be

202

Amar Bhidé

a sufficient condition for the rapid adoption of new technologies. The erstwhile Soviet Union had a well-educated population and few qualms about infringing on foreign patents; nonetheless, it remained far behind the Western technological frontier. Nor does the supply side prescription of low taxes, free markets, and property rights seem to ensure a high level of venturesome consumption. As a native of India who grew up under avowedly socialistic government, I am acutely aware of the debilitating consequences of confiscatory tax rates, pervasive regulation, and expropriation of private property. But the economic record before Independence under a colonial regime strongly oriented toward maintaining a low tax regime, free domestic markets, free international trade, and the rule of law, was even worse. Not only did India miss the Industrial Revolution in its manufacturing industry, as Bhidé and Phelps (2005) have pointed out, under colonial rule it even failed to learn how to consume modern goods. Although, according to the prevailing imperial ideology (and marxist analyses) colonies were supposed to provide captive export markets for their European metropoles, India’s imports were persistently lower than its exports. The institutions, or what Ohzawa and Rosovsky (1973) called “social capabilities,” that help sustain the venturesome consumption and other sources of prosperity of advanced economies cannot be reduced to a few well-defined elements. The formula is both complex (see Nelson 2006) and ever changing. For instance, the common beliefs that now undergird the demand for new products and services have distinctively modern features. The widespread belief in the inevitability and desirability of technological progress is an important case in point. In earlier times a relatively small number of visionary inventors and scientists held such views. Now many popular magazines, TV shows, and management books are predicated on the assumption of scientific and technological progress. Their growing acceptance has turned such beliefs into selffulfilling prophecies. Consider, for instance, Gordon Moore’s famous observation that the number of transistors built on a chip doubles every eighteen months. Semiconductor companies that believe in this socalled law invest the resources needed to make it come true. Downstream customers, (e.g., PC manufacturers) and providers of complementary goods to their customers (e.g., applications software companies) design products in anticipation of the eighteen-month cycle. So when the new chips arrive, they find a ready market, which

Venturesome Consumption, Innovation, and Globalization

203

in turn validates beliefs in Moore’s law and encourages even more investment in building and using new chips. In principle, expectations of change can also slow it down—as David (1986) puts it “if it is expected that every one will quickly adopt the [new] technology, the inducements to bear the costs of adopting it early are reduced.” Why buy the $5,000 flat panel TV set now, when a year from now the price will surely drop and the reliability of the models will increase? Apparently a sizable number of users derive utility not just from the functions that the new products provide but from the fact of being early adopters. As Keynes (1930: 326) pointed out, people have both “absolute” needs (e.g., for health and survival) and “relative” needs that we feel “only if their satisfaction lifts us above, makes us feel superior to, our fellows.” Early purchasers of goods like flat panel TVs apparently enter into a tacit bargain with other consumers: they incur the higher risks and costs, which then drives down prices and improves the quality for the consumers who wait; in return, those who wait give the early purchasers the gratification of being first. The gratification that many modern consumers enjoy may be contrasted with the long-standing propensity to consume expensive goods for the sake of displaying status or wealth—the “conspicuous consumption” of the gilded age that Thorstein Veblen wrote about in The Theory of the Leisure Class. Only the wealthy can indulge in conspicuous consumption; moreover, as Veblen put it, to satisfy its purpose—the demonstration of wealth—the consumption “must be wasteful.” In contrast, many early purchasers of the latest gadgetry aren’t flush with cash (or even pretending to be); they seek to display, to themselves and to others, their technological sophistication, not their wealth. (The classic form of conspicuous consumption has certainly not disappeared, however.) The propensity of consumers to open their hearts and wallets to new offerings also involves the dilution of prior beliefs in the moral and economic value of thrift. Up to the end of the nineteenth century, in accord with Max Weber’s thesis, religious convictions about thrift sustained the “spirit of capitalism.” Weber argued that merchants and industrialists accumulated capital in the belief that they had a moral duty to strive for wealth as well as to lead austere lives. In fact, because venturesome production requires venturesome consumption, excessive thrift can injure rather than help modern capitalism. As it happens, modern consumers have been more inclined to keep up with the recently acquired baubles of their neighbors (if not stay ahead) than

204

Amar Bhidé

toward excessive thrift. At the dawn of the automobile era only a few very rich buffs could buy and test the vehicles. Now the not-so-well-off use credits cards—or what they “save” by buying paper napkins in bulk at a Wal-Mart—to take their chances on laser surgery and flat panel TVs without much foreknowledge of the utility their purchase will provide. The utility that individuals now derive from using cutting-edge technology can also stimulate business purchases of IT where the IT staff put their love for the latest toys ahead of their employer’s interests. Astute managers who are aware of this tendency can, of course, prevent this. But even managers who have no obsession with technology may embrace enlarging its use for another reason, namely the pressure they face to grow their organizations. As I have argued (Bhidé 2000: ch. 9), in a dynamic modern economy, competitors, customers, capital markets, and labor markets make it difficult to limit growth; as David Packard and William Hewlett (the founders of H-P) concluded, the growth of a company is “a requirement for survival.” Growth in turn facilitates and encourages firms to use IT and other innovative technologies in several ways. First, the scale effects observed by Davies (1979) in the adoption of innovations in the manufacturing sector continue to be seen in the adoption of many IT products by service companies. Unlike Wal-Mart, a small retailer cannot afford to purchase a license for a sophisticated supply chain management software package or pay for the in-house IT staff necessary to install and maintain the package. Moreover, since the vendors’ cost of marketing their packages also tends to be fixed, they tend to favor large customers who buy large (or many packages). Second, growing companies often start new facilities, where it is both easier and more economical to adopt new technologies. As David (1986) writes, while an old plant may be technologically obsolete, it may still cover its variable costs and make it rational for profit maximizing to defer replacing it with a state of the art plant. Moreover it can be operationally disruptive to pull out the old technology. Neither consideration applies with a green-field facility. It is worth noting here that the main differences in the productivity of European and US retailing are in the arena of “big box” retailers (like Wal-Mart) and within this category, the US edge derives mainly from its newly opened retail outlets. Third, the effort to grow can stimulate the search for innovative technologies that can be used to realize economies of scale and scope. For

Venturesome Consumption, Innovation, and Globalization

205

instance, in 1988, Physicians Sales and Services (a company about which I have written case studies) was an “itty-bitty” company in Florida. In 1989, the founder Pat Kelly declared that PSS would become the first national distributor of medical products to physicians’ offices in the United States. With just $20 million in revenues, it had no significant economies of scale that would justify nationwide operation. Like a lot of young companies Kelly recalls, PSS relied on “hard work,” “good people,” “seat of the pants navigation,” and “a lot of luck.” The goal of becoming a national company (which was displayed in big banners in every branch and repeated endlessly by Kelly and his top managers) provided the impetus to search for economies of scale. For instance, the company invested in an order entry system based on handheld computers that increased the speed of deliveries and enabled PSS to reduce the inventories it had to carry. Such initiatives allowed PSS to create national-level scale economies where none had previously existed. These kinds of underpinnings—the pervasive belief in change, the value placed on being the first user, and the pressure faced by managers to grow—cannot be easily measured, nor can we identify the nature of their interactions or their “ultimate” sources. But to ignore such factors on the ground of their elusiveness is to exclude from considerations the essential and distinctively modern ways in which innovation sustains the prosperity of advanced countries. 7.9

Redressing the Upstream Bias

David (1986) makes several noteworthy points about how public policy affects the use of new technologies. (1) Overt efforts to promote the diffusion of innovations are modest in terms both of money and attention devoted to them. They usually comprise efforts to disseminate information (e.g., agricultural extension or “technology transfer” programs in the United States) or the payment of subsidies to adopters of new technologies (e.g., those offered to purchasers of robots in Japan). (2) The range of policies that actually affect the adoption of new technologies is quite broad. These include the “tax treatment of investment, the funding of R&D, the education of scientists and engineers, regulation and standard setting, as well as the monetary and fiscal measures shaping the macroeconomic environment.” (3) Speeding up the rate of technology innovation isn’t always in the public interest; sometimes slowing it down can be more beneficial. (4) Policies to quicken or retard the adoption of new technologies should only be undertaken after

206

Amar Bhidé

“explicit assessments” of the varied and changing environments of different industries; an “absolutely indispensable ingredient in the formulation of rational economic policies” vis-à-vis diffusion is “detailed assessments on an industry-by-industry basis.” (5) The processes of the development and diffusion of new technologies are closely intertwined; therefore “intelligent” policy-making would take a more “integrated” approach to designing innovation and diffusion policies. The policy implications of this chapter are in many respects similar to David’s observations, save in two respects: in the utility of a caseby-case approach and the feasibility of formulating an integrated approach to promoting technology development and diffusion. On the first issue, I have little doubt that the binding constraints or pinch points vary significantly across markets and submarkets. Looking at the health care sector, for instance, we can see some diseases whose cures await an “upstream” scientific breakthrough. In other instances, improved management of hospitals and patient databases using tried and tested technologies can lead to vast improvements in productivity. And in yet other cases the social value of the increased use of therapies and techniques that can be called medical in only the broadest sense of the term seems dubious under a system where the users don’t pay. The record of case-by-case interventions, however, does not appear to be inspiring. The approach obviously invites efforts, both overt and covert, by lobbies to secure results that suit their private ends. The process of public policy making is also slow, and indeed to secure the legitimacy of openness and the accommodation of many points of view, in most cases public policy ought to be formulated with all due deliberation. But technologies and their associated bottlenecks keep changing, so interventions that might have been apropos yesterday may be irrelevant tomorrow. There is no point, for instance, in promoting “hardwired” broadband connections to the Internet if we are on the verge of a cheaper or better wireless alternative. Finally suppose that policy makers could identify the “right” bottlenecks across all industries in a timely manner; they would still be faced with the problem of formulating effective responses. As I have argued in this chapter, the development and the use of new technologies has entrepreneurial features that lie outside the domain of mainstream economics, and while we may crudely describe their manifestations, their underpinnings are elusive. But economic and policy analysts tend to focus on measurable indicators and relationships. The danger is that such an orientation may not only fail to touch the larger but

Venturesome Consumption, Innovation, and Globalization

207

more elusive barriers to progress, they may actually increase these barriers. The same concerns about our profound ignorance of the underlying factors make me skeptical about integrated approaches to the development and diffusion of innovation—a fine principle perhaps, but do we know enough to implement it? The arguments in this chapter, however, suggest the removal of the tendency of policy makers to favor upstream innovation and neglect or even impair what happens mid- and downstream. Such a bias is apparent in the promotion of research and the denigration of marketing; thus pharmaceutical companies that receive substantial tax credits or subsidies for their R&D programs get a tongue lashing for their marketing. Big-Pharma is told to spend more on research and less on peddling “frivolous” drugs. But the frivolous drugs also start in a lab. Moreover even useful drugs can only be effective if they are properly incorporated in a therapeutic regime, and as the McGettigan et al. (2001) study suggests, whereas doctors may say they get their information from reading medical journals, pharmaceutical company salespeople play a more important role in influencing their prescribing habits. Without a marketing push, breakthrough treatments may fail to catch on. Consider the history of using antibiotics to treat ulcers, which suggests an important role for marketing beyond the passive dissemination of information. Warren and Marshall demonstrated a link between helicobacter pylori to peptic ulcers in the early 1980s. In 1987 Couglan et al. published an article in Lancet showing that the eradication of H. pylori with antibiotics could effectively cure peptic ulcers. This further milestone was then followed in the first half of the 1990s by the publication of national and international guidelines on the treatment of H. pylori. But although the consensus guidelines were clear, pharmaceutical companies did not have an incentive to promote the therapies. A literature review by O’Connor (2002) showed that although there was “widespread acceptance of H. pylori as a causal agent” among physicians in principle, there was “significant undertreatment” of peptic ulcers with H. pylori therapies. And physicians who did use the therapies often used “treatment regimens of doubtful efficacy” instead of following the consensus guidelines.24 Marketing plays an even more important role in realizing the value of innovations where there are no guidelines offered by authoritative professional bodies and users face significant Knightian uncertainty about the utility of their purchases. Moreover in many innovations,

208

Amar Bhidé

ranging from corrective laser surgery to enterprise software, the downside faced by users in the aggregate (and sometimes even individually) matches or exceeds the downside of the innovator. In the absence of hard evidence that the returns are worth the risks, even consumers who are innately venturesome need some persuasion. In fact persuading users to take a chance on innovations is a crucial entrepreneurial function that often involves the use of smoke and mirrors and psychological manipulation. Policy makers and others may find the misrepresentations and manipulations distasteful, but they are an essential ingredient of technological progress. Conversely, incentives to promote R&D may help upstream innovators but do little good for the downstream and midstream players. Retailers like Wal-Mart may have very large IT budgets and staff who may even develop some in-house systems. But none of this qualifies for R&D incentives. In principle, the output of midstream innovators (e.g., the VC-backed firms I have been studying) may qualify for R&D subsidies; in practice, however, many such firms not only lack the earnings needed to take advantage of tax credits, they often cannot easily segregate R&D outlays and activities from those of their other functions such as marketing and sales. Efforts to stimulate “savings and investment” also tilt against midand downstream innovations. There appears to be a consensus among policy makers of many stripes that except possibly in recessions, saving is always virtuous and consumption undermines long-term growth—a mindset exemplified by Prestowitz’s (2006c) alarm that the United States “is building its economy into a giant consumption machine.” Mechanisms to mobilize savings such as the stock market and retirement plans are thus regarded with favor while mechanisms that facilitate consumption, like credit cards, with some suspicion. But, as I have argued, Max Weber’s thesis that capitalism is synonymous with capital accumulation ignores the role that the venturesome consumption of innovative goods plays in a modern economy. Moreover the young and the impecunious are more likely to have the recklessness of spirit necessary to perform this role. At least up to a point, their spendthrift ways and the credit cards that sustain them are a boon to economic growth, and because there is no knowing what that point might be, there is no justification for promoting or discouraging their behavior. Similarly policies to promote long-term investment, for instance, by providing tax credits for capital outlays, also seem to be outdated. The modern knowledge economy appears to have erased the old boundar-

Venturesome Consumption, Innovation, and Globalization

209

ies between long-term investment and (supposedly undesirable) shortterm spending. Much of what would traditionally have been categorized as the spending of mid- and downstream players is in fact risky, longterm investment. For instance, as discussed, the purchase price of an ERP system is a fraction of the total project cost, but businesses eligible for an investment tax credit for their purchases of computer hardware and software don’t receive a tax break for the costs of adapting the system to their needs, training users, reengineering their business processes, and so on. It may be that a tax credit for the computers also encourages the other larger outlays. But to the extent that promoting long-term investment is a worthy goal for tax policy, this seems like a roundabout and inefficient way to achieve this purpose (e.g., the tax credit may encourage a business to invest more in the computers and less on user training and reengineering). The goals of educational and immigration policies are more attuned to the labor requirements of upstream innovation. For instance, there is a long-standing claim in the United States that it should train more engineers and scientists so that the costs or labor supply constraints faced by organizations undertaking R&D could be reduced. But this invites the question: whose labor supplies would tighten—what have all the people who otherwise might have become scientists and engineers been doing? And has this been less productive than if they had been working as engineers and scientists? The data suggest that many individuals who could have been engineers or scientists have been working as managers. The progressive increase in the proportion of service sector jobs is well known; many have not noticed, however, the increasing share of managerial and professional jobs—in the United States from about one in six in 1940 to about one in three today (Bird 2004). Some of the managerial positions may be filled by individuals who have engineering degrees but such training is probably not a job requirement. The growth in managerial jobs, which in the last couple of decades has taken place in a climate of cost-cutting, restructuring, and reengineering, probably does not reflect a spontaneous increase in bureaucratization of US companies. More likely it follows from the growth of activities, particularly in the expanding service sector, that are difficult to coordinate and where economies of scale and scope are difficult to come by. Moreover these managers have been at the forefront of the challenging effort to improve the productivity of services sector. As we have seen, the use of technologies such as ERP pose organizational as

210

Amar Bhidé

well as technical challenges; arguably their implementation requires a much higher ratio of managers to technical personnel than did the productivity increasing technologies in the manufacturing sector. In other words, the labor market may not have gotten it monumentally wrong, and interventions that increase the supply of trained personnel for upstream innovators may impair productivity growth by reducing the availability of the personnel in downstream firms. We find a similar bias in the area of immigration policy that takes the form of preferring highly trained engineers and scientists (i.e., those with PhDs and masters degrees) to individuals with just a bachelors degree. Supposedly highly trained individuals required to undertake to cutting-edge R&D are scarce, whereas engineering and scientific jobs that don’t require advanced degrees can easily be filled in the local labor market. In fact as I have just pointed out, the highest valued use of talented locally born individuals may not lie in scientific and engineering jobs at all; therefore immigrants who don’t have advanced degrees probably make as valuable a contribution as those who have advanced degrees. As mentioned in my ongoing study of VC-backed businesses, for instance, we have found that a relatively small proportion of employees in these ostensibly high-tech firms have masters and PhDs. Correspondingly the number of immigrants who have PhDs or masters degrees is also much smaller than the number of immigrants (who usually work in a technical function) whose highest degree is a bachelors. I do not have data on the composition of immigrants on the technical staffs of organizations such as retailers and banks; I would suspect, however, that they have an even lower proportion of immigrants with advanced degrees. I could go on to discuss policies in the areas of antitrust enforcement, land use, and labor mobility that encourage or discourage the adoption of new technologies by influencing the incentives and ability of downstream players to grow their organizations. But rather than make this survey even more cursory, let me conclude by returning to the main topic of this conference: whether or not continental Europe is sinking, and if so, why? As must be apparent, I have not much of a notion. But my perspective, from having had the good fortune to spend my adult life in a place that has enjoyed unparalleled prosperity, is that a vibrant economy is a many-splendored thing. Although, depending on from where one looks, one or the other facet may sparkle, the magic of the system lies in the whole, in its many components and their subtle relationships.

Venturesome Consumption, Innovation, and Globalization

211

Moreover, like the innovations it generates, the system evolves constantly. Although basic human nature may remain the same, our expectations, desires, social compacts, relative positions, and technologies all change. Not so long ago we might have worried about the rust in manufacturing sector; now concerns about the productivity of services seem more pressing. This is not our grandfather’s economy or Adam Smith’s, or David Ricardo’s, or Max Weber’s economy. And while parsimonious models may provide a starting point, useful ideas—ones that a pragmatist like William James would say have “cash value”— about why the economy is surging or stalling must incorporate the many and distinctively modern features of our lives. Appendix: Enterprise Resource Planning (ERP) Systems According to the current Wikipedia entry on the topic, ERP software is used for the “control of many business activities, like sales, delivery, billing, production, inventory management, quality management, and human resources management.” The systems are supposed to integrate many functions including “manufacturing, warehousing, logistics, information technology, accounting, human resources, marketing, and strategic management.” In principle, all these activities and functions are supposed to use a single database, rather than, for instance, the human resources department and the payroll department maintaining records on the same employee in two different and incompatible databases. Costs and Benefits Most ERP systems are not built to suit; they are based on packages provided by software companies like Oracle and SAP. The premise, according to Eric Roberts (2004), professor of computer science at Stanford, is that “software systems are expensive and complex. What’s more, the expense of a software system lies almost entirely in its development; once a system is built and tested, the marginal cost of delivering that same system to other users is typically quite small. The concentration of cost in the development phase creates a strong incentive to share development expenses over a large user base. If it costs $10,000,000 to develop a system, it seems foolish for a single institution to bear that cost alone. Given that the bulk of that $10,000,000 represents development, it makes far more sense—at least in theory—for a consortium of institutions to purchase software from a vendor that can then distribute those costs over the community of users.”

212

Amar Bhidé

But there is a catch, writes Roberts: “The success of any enterprise system depends on refashioning the business practices of the institution to match the software rather than trying to change the software to accommodate the idiosyncrasies of the institution. Changing the software violates the underlying economic assumption that allows for the reduction in cost. If each institution tailors the system to suit its needs, the cost advantage vanishes.” Enterprise software vendors claim that their systems incorporate the best possible business practices. Therefore customers gain significant advantages in refashioning their business practices to fit the standard packages. Even where the packages draw their “best practices” from a variety industries and situations, there can be a considerable gap between the “best practice” configuration available in the package and the practice that works best for a particular organization. In The ABCs of ERP published online by CIO Magazine, Christopher Koch writes: While most packages are exhaustively comprehensive, each industry has quirks that make it unique. Most ERP systems were designed to be used by discrete manufacturing companies (that make physical things that can be counted), which immediately left all the process manufacturers (oil, chemical, and utility companies that measure their products by flow rather than individual units) out in the cold.

Moreover it is simply infeasible for organizations to adopt all of the specified best practices. Therefore we usually find a compromise: organizations change some of its practices to suit the system, but they also “struggle” to “modify” core ERP programs to their needs. All this makes it extremely difficult to assess the value or the costs. Koch writes that the value of the systems is hard to pin down because: . . . The software is less important than the changes companies make in the ways they do business. If you use ERP to improve the ways your people take orders and manufacture, ship and bill for goods, you will see value from the software. If you simply install the software without trying to improve the ways people do their jobs, you may not see any value at all—indeed, the new software could slow you down by simply replacing the old software that everyone knew with new software that no one does.

Similarly there “aren’t any good numbers to predict the costs” because: [T]he software installation has so many variables, such as: the number of divisions it will serve, the number of modules installed, the amount of integration that will be required with existing systems, the readiness of the company to

Venturesome Consumption, Innovation, and Globalization

213

change and the ambition of the project—if the project is truly meant to be a battering ram for reengineering how the company does its most important work, the project will cost much more and take much longer than one in which ERP is simply replacing an old transaction system. There is a sketchy rule of thumb that experts have used for years to predict ERP installation costs, which is that the installation will cost about six times as much as the software license. But this has become increasingly less relevant. . . . Research companies don’t even bother trying to predict costs anymore.

Implementation Issues The effective use of ERP requires solving both technical and organizational problems. As Koch writes in The ABCs of ERP: The inherent difficulties of implementing something as complex as ERP is like, well, teaching an elephant to do the hootchy-kootchy. The packages are built from database tables, thousands of them, that IS programmers and end users must set to match their business processes; each table has a decision “switch” that leads the software down one decision path or another. . . . [F]iguring out precisely how to set all the switches in the tables requires a deep understanding of the . . . processes being used to operate the business.

Inevitably business processes have to be reengineered. As mentioned, for a user to take advantage of the re-usability of off-the-shelf software packages, they must align their processes with the best practices incorporated into the software. And to have a system that is truly enterprisewide, organizations have to figure out processes that work best across their different units. Also inevitably individuals and organizational subunits tend to resist changing the way they do things, and even if they didn’t, business processes and their associated information systems cannot be changed overnight. Therefore, in addition to figuring out what their businesses processes should ultimately look like (and how the “switches” in the software need to be set to match the processes), organizations need to resolve how they will overcome the resistance to change and the transition from legacy processes and systems. Consultants who have implemented ERP systems in the past can help ameliorate these problems. However, the issues facing different organizations are never identical, so the consultants and their clients have to solve many novel problems. Moreover ERP packages and the other applications—for instance, supply chain, customer relationship management (CRM), and e-commerce software—that ERP is supposed to work with also change, which adds to the difficulty of deriving a tried-and-tested formula for implementation. Researchers and industry

214

Amar Bhidé

experts who have expended considerable effort to investigate what works and what doesn’t work have been unable to get beyond long and seemingly wooly lists. For instance, Somers and Nelson (2001) have formulated a list of twenty-four “critical success factors” that starts with top management support and includes items such as project team competence, interdepartmental cooperation, and having clear goals and objectives. For obvious reasons such lists are just starting points and do little to obviate the need for situation-specific problem solving. The mixed record of ERP systems also points to the difficult problems that users have to solve to realize the potential benefits. According to Holland and Light (1999), successful implementations at Pioneer New Media Technologies and Monsanto have been well publicized, but “less successful projects have led to bankruptcy proceedings and litigation.” Similarly Plant and Wilcocks (2006) note the success of ERP at companies like Cisco as well as spectacular failures at Hershey Foods and FoxMeyer and disappointments at Volkswagen, Whirlpool, and W.L. Gore. Notes 1. David repeated this observation at a 2003 conference held in memory of Zvi Grilichches when he wrote that “the political economy of growth policy has promoted excessive attention to innovation as a determinant of technological change and productivity growth, to the neglect of attention to the role of conditions affecting access to knowledge of innovation and their adoption” (David 2003). In that paper David also remarked that Griliches’s work first on the diffusion of technology and then on the sources of growth of total factor productivity had been pathbreaking, yet Griliches had not pursued the connections between the two. Instead, in his later work Griliches had focused on the “upstream” sources of productivity growth, namely R&D efforts and patenting. 2. Using a different comparative methodology, Gordon (2006) suggests that per capita GDP in Europe fell from about 75 percent of US per capita GDP in 1975 to 69 percent in 2004 3. The following is a condensed version of a prior lecture (Bhidé 2004). I am grateful to the Kauffman Foundation for funding the field study from which the ideas for this paper have emerged, to Elizabeth Gordon, the Research Associate on the study, and to Massimo Cordella who provided outstanding assistance for this paper. 4. In principle, societies could accommodate the reduction in the demand for labor by increasing everyone’s leisure. Over the last century, economic growth has helped reduce working hours and increase vacations. But somehow, beyond a certain point, societies seem unable to accommodate reductions in the demand for labor by spreading the work around. Efforts to control unemployment by mandating reductions in work weeks or increasing the number of holidays don’t seem to work.

Venturesome Consumption, Innovation, and Globalization

215

5. Although Rosenberg defers to Schumpeter’s analysis for “major innovations” involving “significant shifts to an entirely new production function,” he does not provide examples of such one shot breakthroughs. In fact I can’t think of any. Even ventures like Federal Express that started with a revolutionary concept required refinement over several years before they attained commercial viability (Bhidé 2000). 6. Around the time of the 1999 to 2000 Internet bubble, the “Silicon Valley,” venturecapital backed model of the firm was popular among management gurus and academic researchers. In a 1999, Gary Hamel’s Harvard Business Review article exhorted large companies to “bring Silicon Valley inside.” Kortum and Lerner’s “Assessing the contribution of venture capital to innovation” was published in the RAND Journal of Economics in 2000. Using a variety of methods, but then “focusing on a conservative middle ground” Kortum and Lerner estimate that “a dollar of venture capital appears to be three times more potent in stimulating patenting than a dollar of traditional corporate R&D” suggesting that “venture capital, even though it averaged less than 3 percent of corporate R&D from 1983 to 1992, is responsible for a much greater share–about 8 percent–of US industrial innovations during this decade.” After the bursting of the Internet bubble, however, the VC-backed model has apparently lost some of its luster, and there has been no major redeployment of research resources in its direction. 7. George Richardson, has extensively analyzed the general idea of complementarities across organizations, notably in his 1972 Economic Journal article on the organization of industry. 8. I argue that at one end, publicly financed firms can undertake large projects but have a relatively low tolerance for Knightian uncertainty of highly novel initiatives. At the other end, the self-financed entrepreneur tends has a comparative advantage in undertaking projects that are in a broad sense of the word, highly novel. My recent research also suggests important differences between the kind of R&D undertaken by VC-backed businesses and by the labs of large companies. VC-backed businesses undertake relatively quick and dirty projects and don’t employ many researchers with advanced degrees. And, they often use the innovations coming out of large company and university research efforts. The basic differences in category may well be responsible for the large difference in the efficiency of patenting efforts reported by Kortum and Lerner and the absence of any significant redeployment of research expenditures after the publication of their results. 9. For instance, VCs can vary in their target investment amounts and capacities and in their preferences for early or later stage investments. Some provide funding to biotechnology firms on the scientific frontier; others provide the coffee retailers Starbucks with the capital needed for geographic expansion. Similarly some self-financed entrepreneurs help shape completely new industries, as Paul Allen and Bill Gates did in 1975. Others create new supply chains and distribution systems after the product category has come of age, as did the company Michael Dell started in 1982 in the computer industry. Nonprofit organizations like university research facilities, also contribute to the process of innovation—and in quite different ways. Some undertake basic research whose commercial value, if any, is very indirect; others favor projects with an eye to licensing the intellectual property they develop; and yet others provide a venue for the interactions and tinkering that accidentally result in the start of companies like Cisco and Google. 10. Moreover the system appears to have shown a tendency toward an increasing proliferation of species and characters, although not necessarily in a smooth progression. In

216

Amar Bhidé

the nineteenth century, inventions of new products were made by a few individuals. Edison brought forth a remarkable cornucopia including incandescent bulbs, motion pictures, and gramophones, from a facility in Menlo Park (New Jersey, not California) with fewer employees than the typical Silicon Valley start-up. Alexander Graham Bell had one assistant. Automobile pioneers were one or two man shows—Karl Benz and Gottlieb Daimler in Germany, Armand Peugeot in France, and the Duryea brothers of Springfield, Massachusetts. The large professionally managed corporation became an important contributor to innovation in the first half of the twentieth century. In the second half of the century, the diversity of the entrepreneurial species further increased. Research laboratories in universities that had hitherto focused just on creating knowledge began to develop commercially useful technologies. Similarly professionally managed venture capital funds saw explosive growth. The emergence of new organizations did not, however, wipe out the old. Individual entrepreneurs weren’t (as Schumpeter suggested) made obsolete by large public firms, which in turn survived the growth of venture capital. Although they sometimes butted heads, the old and the new forms generally complemented each other’s contributions— their planned and unwitting collaborations, taking place simultaneously and in sequence, made products that initially were only kind of, sort of, commercially viable. The PC industry and the Internet do not have a solitary Alexander Graham Bell or Henry Ford. Rather, many entrepreneurs, venture capitalists, large companies, standard setting institutions, university and state-sponsored laboratories, and even investment bankers and politicians have revolutionized the way we compute and communicate. Some participants in the revolutions have acquired considerable wealth and fame; others have received neither. Dan Bricklin and Bob Frankston who created the first spreadsheet, and Sir Timothy Berners-Lee who invented the World Wide Web, did not profit at all from their contributions. And outside specialized circles, mention of their names often evokes puzzled looks. 11. This could be because of an artifact of my samples or because I have not specifically focused on the issue of user-led innovations. Moreover, although user-led innovation has been critical in many instances, it may not be as pervasive as the academic literature suggests: the academic community, like the newspapers can be prone to a “man-bitesdog” type bias—studies of producer-led innovation would not excite interest and so would not be widely undertaken or published. 12. The contribution of customers to the development process continues after the first full-blown commercial launch. As previously discussed, products can evolve so much over time that the relationship to their antecedents may be all but unrecognizable. Rosenberg (1982) suggests that “learning by using” by customers often plays a significant role in such transformations. 13. And, as long as there is no trade in intermediates, from the point of view of national incomes and expenditures, it is not unreasonable for the models to stipulate that a single player rather than a swarm develops and produces innovative products. 14. These findings, according to Eaton and Kortum are consistent with historical accounts of the importance of foreign technology to the United States such as Mueller’s (1962) description of the foreign inventions underlying DuPont’s innovations. 15. Not all consumer goods and services show great variance across markets and multinational companies that sell consumer goods do try to standardize their offerings across geographies to minimize costs: Apple, for instance, sells the same iPod in Europe as it does in the United States. That said, there are many examples where the upstream components are more localized than the downstream offerings.

Venturesome Consumption, Innovation, and Globalization

217

16. Of course, technological change could allow jobs that have to be performed locally now to be performed remotely in the future or cause such jobs to vanish altogether. Famously the jobs of most bank tellers have been replaced by a combination of mechanized ATMs and human operators in remote call centers. But history suggests that that “new want machine” keeps replenishing the demand for service jobs that have to be performed locally. At Columbia Business School, for instance, notwithstanding automation and off-shoring the number of support and administrative staff at Columbia has increased not decreased. The job of cutting paychecks could have been outsourced, and there are no punch card operators in the computer center. But twenty years ago there was no one on campus to install and service LCD projectors, personal computer, local area networks, or email systems. Unless these historical trends change dramatically, locally performed services will continue to account for a very large share of economic activity. 17. According to Teresi (2004: 38) the physicist Robert Wilson appeared before Congress to secure $250 million for building Fermilab, the largest particle accelerator in the world. A friendly Congressman tossed a “softball” question that gave Wilson the opportunity to justify the new atom smasher using national defense. Wilson insisted that it had “nothing at all” to do with national security. Rather Wilson said “It has only to do with the respect with which we regard one another, the dignity of men, our love of culture. It has to do with, are we good painters, good sculptors, great poets? I mean all the things we really venerate and honor in our country and are patriotic about. It has nothing to do directly with defending our country except to make it worth defending.” 18. The data used to construct the table were generously provided by Pankaj Ghemawat and Ramon Casadesus-Masanell who had collected and used the data (in a very different way) for their 2006 article. 19. Brynjolfsson (1993) made a strong case that the IT “productivity paradox” reflected deficiencies in the measurements and methodological tool kits. This apparently did not dissuade researchers from continuing to use these measurements and methods. 20. See also O’Mahony and van Ark (2003). 21. Moreover the gross fixed investment numbers in the tables include investments in residential real estate. The ratio of true “business” investment to GDP is likely to be even lower in the United States than in Japan and in the countries of western Europe. 22. David (1986) reviews the extensive theoretical and empirical research on the economics of technology diffusion. The review shows that a great deal of work has been done on the incentives that firms face and the costs they incur in adopting new technologies. These findings help explain differences in the rates of the adoption across industries and firms and help us analyze whether, from a social welfare point of view, the rates are too slow or fast. The research does not, however, examine the sources of the differences in rates of new technology adoption across countries (whose economies comprise many industries and firms). Similarly Cohen and Levinthal (and other researchers such as Cockburn and Henderson 1998) who have worked on “absorptive capacity” peek “under the hood” of firms to examine how their strategies affect their adoption of new technologies, but not why such strategies might vary across countries. Moreover this line of research focuses on “upstream” high-tech businesses, and it is difficult to extrapolate the findings to service sectors that now dominate advanced economies. 23. The better educated farmer, write Nelson and Phelps (1966), will adopt a profitable process more quickly since “he is better able to discriminate between promising and unpromising ideas.” In large industrial corporations, educated scientists keep abreast of

218

Amar Bhidé

technological improvements and educated top managers make the final decision. Therefore as a general principle, the “time lag between the creation of a new technique is a decreasing function of . . . average educational attainment.” 24. These data led O’Connor to suggest the use of “some of the methods used by pharmaceutical manufacturers to educate physicians about their products, which are known to be effective and often overshadow the information available in the medical literature.” But who could do this? Replicating a good marketing system is easier said than done.

References Acs, Z. J., and D. B. Audrestsch. 1991. Innovation and Small Firms. Cambridge: MIT Press. Arrow, K. J. 1982. Innovation in large and small firms. In J. Ronen, ed., Entrepreneurship. Lexington, MA: Lexington Books. Baily, M., and R. Gordon. 1988. The productivity slowdown, measurement issues, and the explosion of computer power. Brookings Papers on Economic Activity 2: 347– 431. Baumol, W. J. 2002. The Free-Market Innovation Machine: Analyzing the Growth Miracle of Capitalism. Princeton: Princeton University Press. Bhidé, A. 1983. Beyond Keynes: Demand side economics. Harvard Business Review 6 (4): 100–10. Bhidé, A. 1986. Hustle as strategy. Harvard Business Review 64 (5): 59–65. Bhidé, A. 2000. The Origin and Evolution of New Businesses. New York: Oxford University Press. Bhidé, A. 2004. Non-destructive creation: How entrepreneurship sustains development. Lecture at the Royal Society of Arts, London. November 17. Bhide, A. 2006. How novelty aversion affects financing options. Capitalism and Society 1 (1): article 1. Bhidé, A., and E. S. Phelps. 2005. A dynamic theory of China–U.S. trade: Making sense of the imbalances. Working paper 4. Center for Capitalism and Society, New York. Bird, R. 2004. Testimony on “The Department of Labor’s Overtime Regulations Effect on Small Business Committee” before Subcommittee on Workforce, Empowerment, and Government Programs, May 20. Bloom, N., Sadun, R., and J. Van Reenen. 2005. It ain’t what you do, it’s the way that you do I.T.—Testing explanations of productivity growth using U.S. affiliates. Working paper. Centre for Economic Performance, London School of Economics. Bresnahan, T. F. 1986. Measuring the spillovers from technical advance: Mainframe computers in financial services. American Economic Review 76 (4): 742–55. Brynjolfsson, E. 1993. The productivity paradox of information technology: Review and Assessment. Communications of the ACM 36 (12).

Venturesome Consumption, Innovation, and Globalization

219

Brynjolfsson, E., and L. Hitt. 1996. Paradox lost? Firm-level evidence on the returns to information systems spending. Management Science 42 (April): 541. Carter, C. F., and B. R. Williams. 1964. Government scientific policy and the growth of the British economy. The Manchester School, September, pp. 197–214. Casadesus-Masanell, R., and P. Ghemawat. 2006. Dynamic mixed duopoly: A model motivated by Linux vs. Windows. Management Science. Cohen, W. M., and D. A. Levinthal. 1989. Innovation and learning: The two faces of R&D. The Economic Journal (September): 369–96. Cohen, W. M., and D. A. Levinthal. 1990. Absorptive capacity: A new perspective of learning and innovation. Administrative Science Quarterly 35: 128–52. Cockburn, I. M., and R. M. Henderson. 1998. Absorptive capacity, coauthoring behavior, and the organization of research in drug discovery. Journal of Industrial Economics 46: 157–82. Cockburn, I. M., and S. Wagner. 2006. Patents and the survival of Internet-related IPOs. Conference paper. NBER Entrepreneurship Meeting, Cambridge, MA. March 10. David, P. A. 2003. Zvi Griliches on diffusion, lags and productivity growth . . . Connecting the dots. Paper prepared for the Conference on R&D, Education and Productivity Held in Memory of Zvi Griliches (1930–1999), August 25–27, 2003. David, P. A. 1986. Technology diffusion, public policy, and industrial competitiveness. In R. Landau and N. Rosenberg, eds., The Positive Sum Strategy: Harnessing Technology for Economic Growth, Washington, DC: National Academy Press, 373–91. David, P. A. 1990. The dynamo and the computer: An historical perspective on the modern productivity paradox. The American Economic Review Papers and Proceedings (May): 355–61. Davies, S. 1979. The Diffusion of Process Innovations. Cambridge: Cambridge University Press. Dedrick, J., V. Gurbaxani, and K. L, Kraemer. 2003. Information technology and economic performance: A critical review of the empirical evidence. Computing Surveys 35 (1): 1–28. Eaton, J., and S. Kortum. 1995. Engines of growth. Working paper 5207. National Bureau of Economic Research, Cambridge, MA. Enos, J. L. 1962. Petroleum Progress and Profits. Cambridge: MIT Press. Freeman, C. 1968. Chemical process plant: Innovation and the world market. National Institute Economic Review 45 (August): 29–57. Freeman, R. B. 2005. Does globalization of the scientific/engineering workforce threaten U.S. economic leadership. Working paper 11457. National Bureau of Economic Research, Cambridge, MA. Gordon, R. J. 1999. Has the “new economy” rendered the productivity slowdown obsolete? Unpublished manuscript. Northwestern University. Gordon, R. J. 2004. Why was Europe left at the station when America’s productivity locomotive departed? CEPR working paper. http://faculty-web.at.northwestern.edu/ economics/gordon.

220

Amar Bhidé

Gordon, R. J. 2006. Issues in the comparison of welfare between Europe and the United States. Paper presented at Center on Capitalism and Society/CESifo joint conference in Venice, July 21–22. Griliches, Z. 1994. Productivity, R&D, and the data constraint. American Economic Review 84 (1): 1–23. Hamel, G. 1999. Bringing (September–October).

Silicon

Valley

inside.

Harvard

Business

Review

Hausman, J. A. 1997. Valuation of new goods under perfect and imperfect competition. In T. F. Bresnahan and R. J. Gordon, eds., The Economics of New Goods. Chicago: University of Chicago Press, 29–66. Holland, C. P., and B. Light. 1999. A critical success factors model for ERP implementation. IEEE Software (May–June): 30–35. Jorgenson, D. W., and K. J. Stiroh. 2000. Raising the speed limit: US economic growth in the information age. Working paper 261. OECD Economics Department, Paris. Keynes, J. M. 1930. Economic Possibilities for our Grandchildren. In Collected Writings. Vol. IX: Essays in Persuasion. London: St. Martin’s Press, 1971–73. Originally published in 1930. Kronholz, J. 2006. Under a cloud: For Dr. Sengupta, long-term visa is a long way off. Wall Street Journal June 27, p. A1. Lucas, H. C., Jr. 1999. Information Technology and the Productivity Paradox: The Search for Value. Oxford: Oxford University Press. Mansfield, E., J. Rapoport, A. Romeo, S. Wagner, and G. Beardsley. 1977. Social and private rates of return from industrial innovations. Quarterly Journal of Economics 91 (2): 221–40. McGettigan, P., J. Golden, J. Fryer, R. Chan, and J. Feely. 2001. Prescribers prefer people: The sources of information used by doctors for prescribing suggest that the medium is more important than the message. British Journal of Clinical Pharmacology 51 (February): 184–89. McKinsey & Co. 2005. The Emerging Global Labor Market. New York: McKinsey Global Institute. Mueller, W. F. 1962. The origins of the basic inventions underlying DuPont’s major product and process innovations, 1920–1950. In The Rate and Direction of Inventive Activity: Social Factors. NBER Conference Series 13. Princeton: Princeton University Press. Nelson, R. R. 1993. National Innovation Systems: A Comparative Analysis, Oxford University Press. Nelson, R. R., and G. Wright. 1992. The rise and rise and fall of American technological leadership: The postwar era in historical perspective. Journal of Economic Literature 30 (December): 1931–64. Nelson, R. R. 2006. What makes an economy productive and progressive? What are the needed institutions? Mimeo.

Venturesome Consumption, Innovation, and Globalization

221

Nordhaus, W. D. 1997. Do real-output and real-wage measures capture reality? The history of lighting suggests not. In T. F. Bresnahan and R. J. Gordon, eds., The Economics of New Goods. Chicago: University of Chicago Press, 29–66. O’Connor H. J. 2002. Helicobacter pylori and dyspepsia: Physicians’ attitudes, clinical practice, and prescribing habits. Alimentary Pharmacology and Therapeutics 16 (3): 487–96. O’Mahony, M., and B. van Ark, eds. 2003. EU Productivity and Competitiveness: An Industry Perspective Can Europe Resume the Catching-up Process? Luxembourg: Office for Official Publications of the European Communities. Ohzawa, K., and H. Rosovsky. 1973. Japanese Economic Growth. Stanford: Stanford University Press. Parente, S. L., and E. C. Prescott. 1994. Barriers to technology adoption and development. Journal of Political Economy 2 (2): 298–32. Prestowitz, C. 2005. Three Billion New Capitalists: The Great Shift of Wealth and Power to the East. New York: Basic Books. Prestowitz, C. 2006a. America’s Technology Future at Risk: Broadband and Investment Strategies to Refire Innovation. Washington, DC: Economic Strategy Institute. Prestowitz, C. 2006b. The world is tilted. Newsweek, special issue. Richardson, G. 1972. The organization of industry. The Economic Journal 84: 883–96. Roach S. 1988. White collar productivity: A glimmer of hope? Special economic study, Morgan Stanley, New York. Roberts, E., 2004. Here be dragons: The economics of enterprise software systems. Letter to members of the Faculty Senate, Stanford University, May 27. Rosenberg, N. 1976. Perspectives on Technology. New York: Cambridge Univeristy Press. Rosenberg, N. 1982. Learning by using. In Inside the Black Box: Technology and Economics. New York: Cambridge University Press, ch. 6. Rucker, T. D. 1976. Drug information for prescribers and dispensers: Toward a model system. Med Care 14(2): 156–65. Schumpeter, J. A. [1911] 1934. The Theory of Economic Development. Cambridge: Harvard University Press. Schumpeter, J. A. [1942] 1961. Capitalism, Socialism and Democracy. London: George Allen and Unwin. Smith, A. [1776] 1937. An Inquiry into the Nature and Causes of the Wealth of Nations. New York: Random House. Solow, R. M. 1987. We’d better watch out. New York Times, July 12, p. 36. Teresi, D. 2004. Foreign policy, Flemish painters, and pharoah placement: The many purposes of science. InCharacter (Fall): 36–43. Trajtenberg, M. 1989. The welfare analysis of product innovations, with an application to computed tomog-raphy scanners. Journal of Political Economy 97: 444–79.

222

Amar Bhidé

van Ark, B., R. Inklaar, and R. H. McGuckin. 2002. “Changing gear.” Productivity, ICT and service industries: Europe and the United States. Research memorandum GD-60. Groningen Growth and Development Centre, University of Groningen. von Hippel, E. 1976. The dominant role of users in the scientific instrument innovation process. Research Policy 5 (3): 212–39. von Hippel, E. 1988. The Sources of Innovation. Oxford: Oxford University Press. von Hippel, E. 2005. Democratizing Innovation. Cambridge: MIT Press. Winter, S. 1984. Schumpeterian competition in alternative technological regimes. Journal of Economic Behavior and Organization 5: 287–320.

8

Cyclical Budgetary Policy and Economic Growth: What Do We Learn from OECD Panel Data? Philippe Aghion and Ioana Marinescu

8.1

Introduction

This chapter uses yearly panel data on OECD countries to analyze the relationship between growth and the cyclicality of the budget deficit. We develop new yearly estimates of the countercyclicality of the budget deficit, and show that the budget deficit has become increasingly countercyclical in most OECD countries over the past twenty years. However, EMU countries did not become more countercyclical. Using panel specifications with country and year fixed effects, we show that (1) an increase in financial development, a decrease in openness to trade, and the adoption of an inflation targeting regime move countries toward a more countercyclical budget deficit; (2) a more countercyclical budget deficit has a positive and significant effect on economic growth, and this effect is larger when financial development is lower. A common view among macroeconomists, is that there is a decoupling between macroeconomic policy (budget deficit, taxation, and money supply) which should primarily affect price and income stability,1 and long-run economic growth which, if anything, should depend only on structural characteristics of the economy (property right enforcement, market structure, market mobility, etc.). That macroeconomic policy should not be a key determinant of growth, is further hinted at by recent contributions such as Acemoglu et al. (2004) and Easterly (2005), who argue that the correlation between macroeconomic volatility and growth (Acemoglu et al.) or those between growth and macroeconomic variables (Easterly), become insignificant once one controls for institutions. The question of whether macroeconomic policy does or does not affect (productivity) growth is not purely academic. In particular, it underlies the recent debate on the European Stability and Growth Pact

224

Philippe Aghion and Ioana Marinescu

as well as the criticisms against the European Central Bank for allegedly pursuing price stability at the expense of employment and growth. In this chapter we question that view by arguing that the cyclicality of the budget deficit is significant in explaining GDP growth, with a more countercyclical budgetary policy being more growth enhancing the lower the country’s level of financial development. We also identify economic factors that tend to be associated with more countercyclical policies. These results hold in a sample of OECD countries with comparable institutional environments. The idea that cyclical macroeconomic policy might affect productivity growth, is suggested by previous work by Aghion, Angeletos, Banerjee, and Manova (2006), henceforth AABM. The argument in AABM is that credit constrained firms have a borrowing capacity that is typically conditioned by current earnings (the factor of proportionality between earning and debt capacity is called credit multiplier, with a higher multiplier reflecting a higher degree of financial development in the economy). In a recession, current earnings are reduced, and so is firms’ ability to borrow in order to maintain growth-enhancing investments (e.g., in skills, structural capital, or R&D). To the extent that higher macroeconomic volatility translates into deeper recessions, it should affect firms’ incentives to engage in such investments. This prediction finds empirical support, first in cross-country panel regressions by AABM who show on the basis of cross-country panel regressions that structural investments are more procyclical the lower the country’s level of financial development; and second, in firm-level evidence by Berman et al. (2007). Using French firm-level panel data on R&D investments and on credit constraints, Berman et al. show that (1) the share of R&D investment over total investment is countercyclical without credit constraints; (2) this share turns more procyclical when firms are credit constrained; (3) this effect is only observed during down-cycle phases—namely in presence of credit constraints, R&D investment share plummets during recessions but doesn’t increase proportionally during up-cycle periods.2 These findings in turn suggest that countercyclical macroeconomic policies, with higher government investment or lower nominal interest rates during recessions, may foster productivity growth by reducing the magnitude of the output loss induced by market failures (in particular by credit market imperfections) in a recession, which in turn should allow credit-constrained firms to preserve their growthenhancing investments over the business cycle. For example, the

Cyclical Budgetary Policy and Economic Growth

225

government may decide to stimulate the demand for private firms’ products by increasing spending. This could further increase firm’s liquidity holdings and thus make it easier for them to face idiosyncratic liquidity shocks without having to sacrifice R&D or other types of longer-term growth-enhancing investments. However, in a recession, more workers face unemployment, so that their earnings are reduced. Government spending could help them overcome credit constraints either directly (social programs, etc.) or indirectly by fostering labor demand and therefore employment; this relaxation of credit constraints in turn would allow workers to make growth-enhancing investments in human capital, relocation, and so forth. The tighter the credit constraints faced by firms and workers, the more growth enhancing such countercyclical policies should be.3 Our contribution in this chapter is threefold. It is first to compute and analyze the cyclicality of the budget deficit on a panel of OECD countries, that is, how the budget deficit responds to fluctuations in the output gap over time. Second, it is to investigate some potential determinants of the countercyclicality of the budget deficit. Third, it is to use these yearly panel data to assess the relationship between growth and the countercyclicality of budgetary policies at various levels of financial development. Our main findings can be summarized as follows: (1) the budget deficit has become increasingly countercyclical in most OECD countries over the past twenty years, but this trend has been significantly less pronounced in the EMU; (2) within countries, a more countercyclical budgetary policy is positively associated with a higher level of financial development, a lower level of openness, and the adoption of an inflation targeting regime; (3) a more countercyclical budgetary policy has a greater positive impact on growth when financial development is lower. While we argue that our results likely reflect the causality from budgetary policy to growth, at the very least they document statistical relationships between macroeconomic variables that are consistent with the theory and microevidence on volatility, credit constraints, and growth-enhancing investments. While we do not know of any previous attempt at analyzing the growth effects of countercyclical budgetary policies, analyses of the determinants of the cyclicality of budgetary policies already exist in the literature. For example, Alesina and Tabellini (2005) argue that more corrupt democracies will tend to run a more procyclical fiscal policy. The idea is that in good times, voters demand that the government cut taxes or provide more public services instead of reducing debt because

226

Philippe Aghion and Ioana Marinescu

they cannot observe the debt reduction and can suspect the government of appropriating the rents associated with good economic conditions. In equilibrium, this leads to a more procyclical policy as the moral hazard problem worsens, in the sense that governments are more likely to divert public resources in booms. They also show that this mechanism tends to be more powerful in explaining the variation observed in the data than borrowing constraints alone. While Alesina and Tabellini (2005) are using a large sample of countries and explore cross-sectional variations, in this study we use panel analysis on OECD countries. This makes the use of corruption indexes impractical for two reasons. First, there is almost no cross-sectional variation in corruption indexes within the OECD. Second, there is even less variation of these indexes across time for individual countries. In a similar vein Calderon et al. (2004) show that emerging market economies with better institutions are more able to conduct a countercyclical fiscal policy.4 Their empirical analysis is based on the International Country Risk Guide. Although the variation in this indicator is limited across OECD countries and time, it presents somewhat more variation than corruption indexes.5 Other studies such as Gali and Perotti (2003) and Lane (2003) focus, as we do, on OECD countries. Gali and Perotti investigate whether fiscal policy in the European Monetary Union (EMU) has become more procyclical after the Maastricht treaty. They find no evidence for such a development. They do find, however, that while there is a trend in the OECD towards a more countercyclical fiscal policy over time, the EMU is lagging behind that trend. Lane (2003) probably comes closer to the analysis developed in the third section of our chapter. Lane examines the cyclical behavior of fiscal policy within the OECD. He then uses trade openness, output volatility, output per capita, the size of the public sector and an index for political power dispersion to examine cross-country differences in cyclicality. The reason why power dispersion may play a role is taken from Lane and Tornell (1998): when multiple political groups compete for public spending, the latter may become more procyclical. No group wants to let any substantial fiscal surplus subsist because they are afraid that this will not lead to debt repayment, but rather to other groups appropriating that surplus. Lane finds in particular evidence that GDP growth volatility, trade openness, and political divisions lead to a more procyclical spending pattern, even though the effect of political divisions is not present for all categories of spending. We contribute to this literature

Cyclical Budgetary Policy and Economic Growth

227

by using yearly panel data to analyze the cyclicality of budgetary policy and its determinants within OECD countries, and we show that the degree of financial development is an important element to explain within country variations in such policies, while future or present EMU membership explains cross-country variations. Moreover we show that inflation targeting is associated with a more countercyclical budgetary deficit. Most closely related to our second stage analysis of the effect of countercyclical budgetary policy on growth, are Aghion-AngeletosBanerjee-Manova (2005), henceforth AABM, and Aghion-BacchettaRanciere-Rogoff (2006), henceforth ABRR. AABM develop a model to explain why macroeconomic volatility is more negatively correlated with productivity growth, the lower financial development, and they test this prediction using cross-country panel data. ABRR move from a closed real to an open monetary economy and show that a fixed nominal exchange rate regime or lower real exchange rate volatility are more positively associated with productivity growth, the lower financial development and the lower the ratio of real shocks to financial shocks. 8.2 The Countercyclicality of the Budget Deficit in the Crosscountry Panel In this section we compute time varying measures of the cyclicality of budgetary policy in our cross-country panel, and compare the extent to which budgetary policy became more countercyclical over time in some countries than in others. A main finding is that budgetary policy in the United States and the United Kingdom have become significantly more countercyclical over the past twenty years, whereas it has not in the EMU area. 8.2.1 Data Panel data on GDP, the GDP gap (ygap), the GDP deflator, and government gross debt (ggfl) are taken from the OECD Economic Outlook annual series.6 Our measure of budgetary policy is the first difference of debt divided by the GDP, which is the same as the budget deficit over GDP. Note that debt and other government data refer to general government. Financial development is measured by the ratio of private credit to GDP, and annual cross-country data for this measure of financial development can be drawn from the Levine database.7 In this latter

228

Philippe Aghion and Ioana Marinescu

measure, private credit is all credit to private agents, and therefore includes credit to households. The “average years of education in the population over 25 years old” series is directly borrowed from the Barro–Lee dataset; this measure is only available every five years and has been linearly interpolated to obtain a yearly series. The openness variable is defined as exports and imports over GDP and data on it come from the Penn World Tables 6.1. The population growth, government share of GDP, and investment share of GDP also come from the Penn World Tables 6.1. The inflation targeting dummy is defined using the dates when countries adopted inflation targeting, as summarized in Vega and Winkelried (2005). All nominal variables are deflated using the GDP deflator. Summary statistics can be found in table 8.1. The sample is an unbalanced panel including the following countries: Australia, Austria, Belgium, Canada, Denmark, Spain, Finland, France, United Kingdom, Germany,8 Iceland, Italy, Japan, Netherlands, Norway, New Zealand, Portugal, Sweden, and United States. 8.2.2 Public Deficit and Growth: The Empirical Challenge We are interested in evaluating the impact of the cyclicality of the budget deficit on the growth of GDP per capita, and how this effect may depend on the degree of financial development. Our expectation is that a more countercyclical budget deficit is more likely to enhance growth when financial development is lower. Empirically we wish to identify this effect from time variation of budgetary policy within countries. Figure 8.1 illustrates this idea for a hypothetical case: we distinguish between the situation where, in the base period t − 1, financial development is low (upper panels), and the situation where financial development is high (lower panels). We start with a baseline depicted in the left-hand side panels of figure 8.1: the budget deficit is thus initially assumed to be procyclical. The right-hand side panels of figure 8.1 illustrate the growth response in period 2 after an increase in the countercyclicality of the budget deficit in period 1, such that the budget deficit becomes strongly countercyclical. If financial development is low, then trend growth in period 2 increases substantially (upper left panel in figure 8.1). If, however, financial development is high, then trend growth increases by a smaller amount9 (lower left panel of figure 8.1). Looking at figure 8.1, the most obvious method one can think of to compute cyclicality is to regress the public deficit on the GDP growth using ordinary least squares (OLS) on the observations in period t. In

756

Inflation targeting dummy

0.112

24.106

12.440

0.006

0.061

53.633

8.236

0.801

0.025

0.608

0.578

0.316

4.983

5.709

0.005

0.066

35.641

1.989

0.392

0.026

1.065

0.752

0.563

0.046

0.295

0.019

Standard deviation

0.321

0.116

0

12.867

1

41.635

27.848

0.762 0.047

−0.018 3.008

266.883

8.705

12.250

−0.025

2.510

2.240

−0.092 0.128

−3.011

−1.112

8.972

3.337

−0.342

−0.065 2.686

1.686

0.071

Maximum

0.046

−0.070

Minimum

Sources: OECD Economic Outlook, Levine dataset, Barro Lee dataset, Penn World Tables 6.1. Note: Sample restricted to observations where the countercyclicality of budget deficit computed using Gaussian weighted rolling windows is not missing.

605 605

Government share of GDP (in %)

689

Population growth

Investment/GDP (in %)

605 756

Openness

585

Average years of schooling for the population over 25 years old

Inflation

689 585

Growth of GDP per capita

532

Countercyclicality of budget deficit (10-year rolling window)

Private credit/GDP

756

Countercyclicality of budget deficit (Gaussian weighted rolling window)

0.511

0.048

756 641

Budget deficit/GDP

0.000 0.548

756 756

GDP gap

Gross government debt/GDP

Countercyclicality of budget deficit [AR(1)]

Mean

Observations

Table 8.1 Summary statistics

Cyclical Budgetary Policy and Economic Growth 229

t

t–1

Time

Trend GDP growth Realized GDP growth Budget deficit

Increased countercyclicality of budget deficit in base period t – 1

High financial development in period t – 1

Figure 8.1 Impact of an increase in the countercyclicality of the budget deficit on growth

t

t–1

Time

Increased countercyclicality of budget deficit in base period t – 1

Low financial development in period t – 1

t–1

t–1

t

t

Time

Time

230 Philippe Aghion and Ioana Marinescu

Cyclical Budgetary Policy and Economic Growth

231

practice, it seems more reasonable to regress the public deficit on the GDP gap (defined as (GDP − GDP*)/GDP*, where GDP* is the trend GDP) rather than the GDP growth. Indeed the GDP gap is very much like a detrended measure of the GDP growth, and a forward-looking government’s budgetary policy should respond to shortfalls from trend rather than to GDP growth per se (for a theory of why fiscal policy should depend on the GDP gap; see Barro 1979). This type of regression based approach to measure the cyclicality of fiscal policies is now common in the literature and can be found for example in Lane (2003) and Alesina and Tabellini (2005). However, the methods used in these papers give rise to only one (or a few) observation of cyclicality per country. Since we want to investigate the impact of time variation in cyclicality, we need to compute for each country time-varying measures for the countercyclicality of budget deficit. Specifically, as we wish to use a yearly panel of countries, we need a measure of countercyclicality that varies yearly. This means that period t − 1 and period t in figure 8.1 are reduced to one single year each! A regression is not defined for a single observation, so we must use observations from a few years in order to compute countercyclicality. The next subsection discusses what methods can be used to compute countercyclicality. 8.2.3 Econometric Methods to Compute Countercyclicality Generally, one would like estimate the following equation for each country i: bit − bi ,t −1 = − a1it y gap ,it + a2 it + ε it , yit

(8.1)

where eit ~ N(0, s e2), and a1it measures the countercyclicality of budgetary policy. Note that there is a minus sign in front of ygap,it: when the economy is in a recession and the GDP gap is negative, the opposite of the GDP gap is positive, and so a positive a1it means that the budget deficit increases when the economy is in a recession; namely the budget deficit is countercyclical. Both a1it and the constant a2it10 are both time varying, which is why we write ajit to denote the coefficient on the variable j in country i at year t.

232

Philippe Aghion and Ioana Marinescu

The variables in equation (8.1) are defined as bit = gross government debt in country i at year t, yit = the GDP in country i and year t, in value, ygap,it = the GDP gap in country i and year t. The GDP gap is computed as (yit − y*i t)/y*i t, where y*i t is the prediction of yit using the Hodrick–Prescott filter. A lambda parameter of 25 was chosen, following Giorno et al. (1995). Note that the GDP gap computed by the OECD using a production function approach is also smoothed by a Hodrick-Prescott technique, so that in practice the difference between the OECD measure of the GDP gap and the measure used here is very limited: the correlation between the two variables is 77 percent. Our measure of the GDP gap is as expected positively correlated with the GDP per capita growth: the correlation is, however, not so strong at 36 percent. Note that bit − bi,t−1 is exactly equal to the opposite of the budget balance, so that our left-hand side variable is equal to the budget deficit as a share of GDP, which we will simply refer to as “budget deficit.” We now examine how the coefficients ajit can be estimated econometrically. One way to implement this is to compute finite (e.g., ten-year) rolling window ordinary least squares estimates. The ten-year rolling window OLS method simply amounts to estimating the countercyclicality of the budget deficit (bit − bi,t−1)/yit at year t in country i by running the following regression for each country i, and all possible years t: bit − bi ,t −1 = − a1it y gap ,iτ + a2 it + ε iτ yit

for τ ∈ (t − 5, t + 4 ) ;

(8.2)

that is, one uses a ten-year centered rolling window to estimate the countercyclicality of budget deficit at any date t. This method suffers, however, from serious shortcomings. First, by definition, we lose the first five years and the last four years of data for each country. Second, because the method involves estimating a coefficient by discarding at each time period one old observation and taking into account a new one, the coefficient can vary substantially when the new observation is very different from the one it replaces. This implies that the series may be jagged and affected by noise and transitory changes; moreover a sudden jump in the series would not be coming from changes in the

Cyclical Budgetary Policy and Economic Growth

233

immediate neighborhood of date t but from changes five years before and four years after. To deal with the shortcomings of the ten-year rolling window method, one can use smoothing such that all observations are used for each year, but those observations closest to the reference year are given greater weight. The “local Gaussian-weighted ordinary least squares” method is one way of achieving this. It consists in computing the ajit coefficients by using all the observations available for each country i and then performing one regression for each date t, where the observations are weighted by a Gaussian centered at date t,11 bit − bi ,t −1 = − a1it y gap ,iτ + a2 it + ε iτ , yit

(8.3)

where

ε iτ ∼ N ( 0, σ 2 wt (τ )) and wt (τ ) =

1 ⎛ (τ − t )2 ⎞ exp ⎜ − . ⎝ 2σ 2 ⎟⎠ σ 2Π

While the local Gaussian-weighted OLS method is less noisy than the ten-year rolling window method, it suffers from a similar shortcoming when it comes to testing the idea illustrated in figure 8.1. Indeed these two methods use observations from both the past and the future (previous years and future years) to calculate yearly countercyclicality. Ultimately we want to look at the impact of year t − 1 changes in countercyclicality on year t growth, but if countercyclicality is computed using some future observations, then in practice we are examining the impact of both past and (some) future countercyclicality on growth. Thus it is hard to be certain that year t − 1 countercyclicality causes year t growth, and reverse causality becomes a problem. One way to address this issue is to use longer lags of countercyclicality (t − 2 or t − 3 or t − 4, etc.), but this requires us to assume that the effects of countercyclicality on growth at year t are delayed for a specific number of years. An alternative method that gets around this problem by making current countercyclicality depend essentially on past observations, is to assume that coefficients follow an AR(1) process, namely using the notation from equation (8.1) for each country i and for each coefficient j: a jit = a ji ,t −1 + ε itj , ε itj ∼ N ( 0, σ a2j ) . a

a

(8.4)

234

Philippe Aghion and Ioana Marinescu

The main challenge in implementing this method is to estimate s 2aj (the variance of the coefficients) at the same time as the variance of the observation, namely the variance s 2e in the formulation of equation (8.1). Once these variances are estimated, applying the Kalman filter gives the best estimates for ajit. The optimal estimates for these variance are extremely hard to compute. While finding analytical closed form solutions turns out to be virtually impossible, Markov chain Monte Carlo (MCMC) methods provide a feasible numerical approximation. We implement the method in Matlab, assuming that the variances of the coefficients and equation are the same for all countries.12 We are thus left with three variances to estimate: two for the coefficient processes ( s 2aj , j = 1, 2) and one for the variance of the error in the equation (s 2e). Intuitively the MCMC method explores randomly (using a Markov chain, hence the name) a wide spectrum of possible values for the variances, and one then retains a set of values that is representative of probable values given the data.13 An advantage of the MCMC method over maximum likelihood type methods is that it does not get stuck in local solutions and properly represents uncertainty about the variances.14 Once we obtain the estimates of these three variances, the ajit coefficients can be calculated using the Kalman filter. AR(1) MCMC is to be preferred over the previous methods for two reasons. First, it reflects a reasonable assumption about policy, namely that policy changes slowly and depends on the immediate past. Second, and most important, it is econometrically appealing in that it makes policy reflected in the ajit coefficients mainly depend on the past (because of the AR(1) specification);15 thus, when the ajit coefficients are used as explanatory variables in panel regressions, it is less likely that there should be a reverse causation problem. 8.2.4 Results We now use the AR(1) method as described above to characterize the level and time path of the countercyclicality of budget deficits in the OECD countries in our sample. We also report some basic results with the 10-year rolling window and Gaussian weighted OLS methods. Table 8.1 summarizes the descriptive statistics of our main variables of interest. For all three measures, the budget deficit is countercyclical (positive coefficient), which is consistent with Lane’s (2003) finding that the primary surplus is procyclical. It is worth noting that the three different methods used in the first stage to estimate countercyclicality give

Cyclical Budgetary Policy and Economic Growth

235

very similar results in terms of the mean: a mean of about 0.5 means that, on average, in our sample a one percentage point increase in the opposite of the GDP gap (i.e., a worse recession) lead to about 0.5 percentage points increase in the budget deficit as a share of the GDP. In terms of the variance of these measures, we can see that the standard error is largest for the ten-year rolling window method, as expected; it is smaller for the Gaussian method, and even smaller for the AR(1) MCMC method. We now look at the evolution of the countercyclicality of budget deficit, as measured by the estimated coefficients a1it from equation (8.1). Figure 8.2 shows the evolution of the countercyclicality of the budget deficit for the United States estimated by the three methods described above. We can readily see that, as expected given the construction of these measures and their empirical standard errors, the ten years’ rolling window yields the most volatile results, and the AR(1) method is the smoothest with the Gaussian-weighted OLS method lying in between. Overall, all three methods show an increase in countercyclicality over time, with a recent trend towards decreasing countercyclicality shown by the ten-year rolling window and Gaussian-weighted OLS methods.

2

1.5

10 years rolling window AR(1) MCMC Gaussian-weighted rolling window

1

.5

0 1960

1970

1980

1990

2000

2010

Year Figure 8.2 Countercyclicality of the budget deficit in the United States. The graph plots the a1it coefficients, meaning the coefficients on the opposite of the output gap in equation (8.1), using various estimation techniques. (Source: OECD Economic Outlook)

236

Philippe Aghion and Ioana Marinescu

In figure 8.3 we then show the countercyclicality of the budget deficit estimated through the AR(1) method for a few countries in our sample. US and UK countercyclicality tends to increase over time, especially since the 1980s. On the contrary, the average countercyclicality of budgetary policy in EMU countries slightly decreases over time. Also one can observe some divergence between EMU and non-EMU countries: at the beginning of the period, the countercyclicality of the budget deficit in EMU countries was very similar to that in the United States; however, as of the 1990s, the United States and the United Kingdom became significantly more countercyclical whereas the EMU did not. In figure 8.4 we plot the same evolution, this time based on coefficients that are estimated using the Gaussian-weighted OLS. Trends in estimates are very similar to those obtained using the AR(1) method. These results are consistent with Gali and Perotti (2003), who show, splitting their sample by decades, that in general fiscal deficits in the OECD have become more countercyclical, but less so in EMU countries. Here we confirm these results using a full-fledged time-series measure of countercyclicality. 1 United States

.8

United Kingdom EMU countries

.6

.4

1960

1970

1980

1990

2000

2010

Year Figure 8.3 Countercyclicality of budget deficits using the AR(1) MCMC method. The graph plots the a1it coefficients, meaning the coefficients on the opposite of the output gap in equation (8.1), using the AR(1) MCMC method. For EMU countries (countries that are or will be part of the EMU) the line represents the average of the estimated coefficients for the EMU countries present in the sample; the average is only computed for those years where all EMU countries have non-missing observations. (Source: OECD Economic Outlook)

Cyclical Budgetary Policy and Economic Growth

2

237

United States United Kingdom EMU countries

1.5

1

.5

0 1960

1970

1980

1990

2000

2010

Year Figure 8.4 Countercyclicality of budget deficits using the Gaussian-weighted OLS method. The graph plots the a1it coefficients, meaning the coefficients on the opposite of the output gap in equation (8.1), using the Gaussian-weighted rolling window OLS method. For EMU countries (countries that are or will be part of the EMU) the line represents the average of the estimated coefficients for the EMU countries present in the sample; the average is only computed for those years where all EMU countries have non-missing observations. (Source: OECD Economic Outlook)

To summarize our descriptive results, we found that the budget deficit has become more countercyclical in the United States and the United Kingdom than in EMU countries since the 1990s. In the next section we investigate possible explanations for these observed differences in the countercyclicality of budgetary deficit across countries and over time. 8.3 First Stage: Determinants of the Countercyclicality of Budgetary Policy In this section we use the series of cyclicality coefficients derived using the AR(1) MCMC method and regress the countercyclicality of budgetary policy over a set of macroeconomic variables. Since our sample is restricted to OECD countries, little variation should be expected from the corruption or other institutional variables considered by the literature so far.16 Instead, we focus on the following candidate variables: financial development, openness, EMU membership,17 and whether the

238

Philippe Aghion and Ioana Marinescu

country has adopted inflation targeting. We also include GDP growth volatility as measured by the standard error of GDP growth, lag of log real GDP per capita and the government share of GDP as control variables. Financial development is a plausible suspect as it influences both the ability and the willingness of governments to borrow in recessions in order to finance the budget deficit. Lower financial development should thus translate into lower countercyclicality of budget deficit. While OECD countries are arguably less subject to borrowing constraints than other countries in the world, there is still a fair amount of cross-country variation in financial development among OECD countries. Openness is also a plausible candidate as one can expect foreign capital to flow in during booms and flow out during recessions, implying that the cost of capital is higher during recessions than during booms. This in turn tends to increase the long-run cost of financing countercyclical budget deficit policies while maintaining the overall debt constant on average over the long run. The EMU dummy is also a plausible candidate, given (1) our observation in figures 8.2 and 8.3 that the budget deficit is less countercyclical in the eurozone than in the United States or the United Kingdom; (2) the deficit and debt restrictions imposed by the Stability and Growth Pact and also the restrictions that individual countries imposed on themselves in order to qualify for EMU membership. Inflation targeting should also improve a country’s willingness or ability to conduct countercyclical budgetary policy. In particular, one potential factor that might discourage governments to borrow in recessions, is people’s expectation that such borrowing might result in higher inflation in the future, for example, as a way for the government to partially default on its debt obligations. This in turn would reduce the impact of current government borrowing on private (long-term) investment. Inflation targeting increases the effectiveness of government borrowing in recession by making such expectations less reasonable. Table 8.2, where the countercyclicality measures are derived using the AR(1) MCMC method, shows results that are consistent with these conjectures, namely (1) while countries that are more financially developed tend to have a less countercyclical budgetary policy (column 1), as a country gets more financially developed, it exhibits a significantly more countercyclical budget deficit (column 2); using the results from column 2, our estimates imply that a ten percentage points increase in

Cyclical Budgetary Policy and Economic Growth

239

Table 8.2 Determinants of the countercyclicality of budget deficits (1) Year f.e.

(2) Country year f.e.

Private credit/GDP

−0.453 (0.115)***

EMU country

−0.127 (0.038)***

Standard error of GDP growth

−3.364 (0.818)***

Lag[log(real GDP per capita)]

0.011 (0.017)

0.072 (0.071)

Government share of GDP (in %)

0.000 (0.005)

0.004 (0.001)***

Inflation targeting

0.292 (0.081)***

0.112 (0.015)***

Openness

−0.007 (0.001)***

−0.002 (0.001)***

Observations R-squared

515 0.21

0.196 (0.018)***

515 0.99

Sources: OECD Economic Outlook, Levine dataset, Barro Lee dataset, Penn World Tables 6.1. Notes: Robust standard errors in parentheses; * significant at 10 percent, ** significant at 5 percent, *** significant at 1 percent. The explained variable is the coefficient on the opposite of the GDP gap in equation (8.1), estimated using the AR(1) MCMC method. EMU country is a dummy variable equal to 1 for all countries that are part of the EMU as of 2006.

private credit over GDP is associated with an increase of about 0.0196 in the countercyclicality of the budget deficit; in other words, it is precisely when the countercyclicality of the budget deficit is more positively correlated with growth, namely when financial development is low, that budgetary deficit countercyclicality seems hardest to achieve; (2) when using country and year fixed effects (column 2) more trade openness is positively and significantly associated with budgetary deficit countercyclicality (the table shows a positive coefficient on openness); (3) EMU countries and countries with a larger standard error of GDP growth appear to have a harder time achieving budgetary deficit countercyclicality (column 1); the EMU dummy implies that on average EMU countries’ budgetary policy countercyclicality is lower by 0.127, which is about a fourth of a standard deviation; the effect of

240

Philippe Aghion and Ioana Marinescu

the EMU dummy is more likely to be explained by rigidities already imposed by the precursor EMS regime and then reinforced by the Maastricht Treaty, rather than the 1999 implementation of the EMU itself;18 further investigation of this question is, however, beyond the scope of this paper; (4) a higher share of government in the GDP is associated with a more countercyclical budgetary policy; (5) pursuing inflation targeting is associated with a more countercyclical budget deficit. Note that the coefficient on the inflation targeting dummy in column 2 is of the same magnitude as the coefficient on the EMU dummy in column 1, but of opposite sign. Hence a lower level of financial development, a higher degree of openness, belonging to the EMU group, and the absence of inflation targeting, are all associated with a lower degree of countercyclicality in the budget deficit. In the next section we move to second stage analysis of the effect of budget deficit cyclicality on growth. 8.4

Second Stage: Cyclical Budget Deficit and Growth

In this section we regress growth on the cyclicality coefficients for budgetary policy derived in section 8.2, financial development, the interaction between the two variables, and a set of controls. Our discussion of the theory and microeconomic evidence on volatility, credit constraints, and R&D/growth at the start of this chapter suggests that the lower financial development, the more positive the correlation should be between growth and the countercyclicality of budgetary policy: the idea is that a more countercyclical budgetary policy can help reduce the negative effect that negative liquidity shocks impose on credit-constrained firms that invest in R&D and innovation. 8.4.1 Empirical Specification and Results Our empirical specification is Δyit = β1 a1i ,t −1 + β 2 pi ,t −1 + β 3 ai ,t −1 pi ,t −1 + Xit β 4 + γ i + δ t + ε it ,

(8.5)

where Δyit is the first difference of the log of real GDP per capita in country i and year t; a1i,t−1 is the countercyclicality of the budget deficit as estimated by the AR(1) MCMC method. Since a1i,t−1 is an estimated coefficient, we weigh each observation by the inverse of the variance of this coefficient (aweights in Stata), thus giving higher weight to coefficients that are more precisely estimated in the first stage. pi,t−1 is the ratio of private credit to GDP borrowed from Levine (2001); Xit is a

Cyclical Budgetary Policy and Economic Growth

241

vector of control variables that vary with the specification considered, g i is a country fixed effect, di is a year fixed effect, and eit is the error term. In table 8.3 we first report results with a limited set of controls representing the most widely accepted determinants of growth: lag of log real GDP per capita, population growth and investment over GDP (column 1). We then add more controls, namely schooling, trade openness, inflation, government share of GDP, and inflation targeting (column 2). Note that since we control for inflation, we indirectly control for the impact of monetary policy on growth. The prediction is that of a positive b1 coefficient for the effect on growth of the countercyclicality of the budget deficit when private credit over GDP is 0, and of a negative b3 coefficient on countercyclicality interacted with financial development. In the first column of table 8.3, using a limited set of controls, we see that the corresponding coefficients have the anticipated signs and are statistically significant: a more countercyclical budget deficit is positively correlated with growth, but the interaction term between countercyclicality and financial development is negative. Including a richer set of controls in column 2 does not change the results. If anything, the point estimates are larger: a coefficient of 0.11 of the lagged countercyclicality of budget deficit means that if private credit over GDP is 0, then increasing the countercyclicality of the budget deficit by one percentage point increases growth by 0.11 percentage points. For each percentage point increase in private credit over GDP, this positive effect of countercyclicality diminishes by 0.0004. The effect of the interaction is thus small: private credit over GDP would need to be larger than 2.75 for a countercyclical budgetary policy to become growth reducing. Such a high value of private credit over GDP is not observed in our sample: the United States in 2000, at 2.24, has the highest value of this variable in our sample. Then, in columns 3 and 4, we repeat the same specifications as in columns 1 and 2 but allow the impact of the interaction between the countercyclicality of the budget deficit and private credit over GDP to differ by quartiles of the private credit over GDP (the first quartile is then the excluded category). For example, the dummy “2ndq (Private credit/GDP)” is equal to one if the Private credit/GDP ratio lies in the second quartile, and is equal to zero otherwise. As the results in these columns show, the interaction between cyclicality and financial development is nonlinear, with a significant jump occurring when the

242

Philippe Aghion and Ioana Marinescu

Table 8.3 Effect of the countercyclicality of budget deficits on growth, AR(1) MCMC method (1)

(2)

(3)

(4)

lag(Countercyclicality of budget deficit)

0.075 (0.021)***

0.110 (0.024)***

0.058 (0.016)***

0.081 (0.018)***

lag(Private credit/GDP)

−0.010 (0.008)

−0.005 (0.008)

−0.014 (0.007)**

−0.008 (0.007)

lag(Countercyclicality of budget deficit*Private credit/ GDP)

−0.030 (0.012)***

−0.040 (0.014)***

Lag[Countercyclicality of budget deficit*2ndq(Private credit/GDP)]

−0.006 (0.003)**

−0.009 (0.003)***

Lag[Countercyclicality of budget deficit*3rdq(Private credit/GDP)]

−0.022 (0.007)***

−0.024 (0.008)***

Lag[Countercyclicality of budget deficit*4thq(Private credit/GDP)]

−0.023 (0.008)***

−0.030 (0.010)***

Lag[log(real GDP per capita)]

−0.140 (0.022)***

−0.132 (0.022)***

−0.142 (0.022)***

−0.131 (0.022)***

Investment/GDP (in %)

0.002 (0.000)***

0.002 (0.000)***

0.002 (0.000)***

0.002 (0.000)***

Population growth

−1.490 (0.268)***

−1.702 (0.284)***

−1.484 (0.272)***

−1.635 (0.290)***

Average years of schooling for the population over 25 years old

0.002 (0.003)

0.003 (0.003)

Government share of GDP (in %)

−0.001 (0.000)***

−0.001 (0.000)**

Inflation

−0.049 (0.022)**

−0.053 (0.021)**

Inflation targeting

−0.004 (0.005)

−0.001 (0.005)

Openness

0.001 (0.000)**

0.001 (0.000)***

Observations R-squared

477 0.60

467 0.64

477 0.62

467 0.65

Sources: OECD Economic Outlook, Levine dataset, Barro Lee dataset, Penn World Tables 6.1. Notes: Robust standard errors in parentheses; * significant at 10 percent, ** significant at 5 percent, *** significant at 1 percent. The explained variable is the first difference of the log of real GDP per capita. All specifications include country and year fixed effects. Columns 3 and 4 allow for the effects of countercyclicality of the budget deficit to differ with quartiles of private credit/GDP.

Cyclical Budgetary Policy and Economic Growth

243

private credit ratio moves from the second to the third quartile. In other terms, it is only at fairly high levels of financial development that countercyclical budgetary policy becomes noticeably less growth enhancing. Table 8.3 is thus consistent with the prediction of a positive effect of a countercyclical budget deficit on growth, whereas we see a negative and significant interaction effect between private credit and the countercyclicality variable. Thus the less financially developed a country is, the more growth enhancing it is for the government to be countercyclical in its budgetary policy. In particular, we observe that EMU countries have budgetary policies that are on average far less countercyclical than in the United States (0.37 vs. 0.61), even though the United States are more financially developed than the EMU: thus the ratio of private credit to GDP in 2000 in the EMU countries is equal 1.02 against 2.24 in the United States. Then, to the extent that it reflects the causality from cyclical budgetary policy to growth, the regression in table 8.3 suggests that increasing the countercyclicality of the budgetary policy would be more growth enhancing for the EMU than for the United States. 8.4.2 Robustness Tests This section discusses various potential issues with our table 8.3 estimates. We take as the reference specification for this discussion the specification shown in table 8.3, column 2. Therefore, when we report on alternative specifications, they are all based on this reference specification. A potential first source of concern for our estimation strategy is autocorrelation of residuals, which is typical in panel growth regressions. This implies that the standard errors may be biased. To correct for this potential bias, we used Newey errors to adjust the standard errors in the reference specification. Allowing for autocorrelation of errors up to lag 1 increased the standard errors very slightly and left the coefficients significant at the 1 percent level. Allowing for autocorrelation up to 5 lags leaves the effect of the countercyclicality of the budget deficit at the same level of statistical significance, but makes the interaction between the countercyclicality of the budget deficit and private credit be only significant at the 2 percent instead of the 1 percent level. Globally it does not seem that autocorrelation of residuals substantially affects the standard errors of our estimates. Second, the reader may wonder about what components of the budget deficit increase growth when they are more countercyclical. For

244

Philippe Aghion and Ioana Marinescu

example, is it the countercyclicality of total government spending that ultimately matters for growth? What about transfers and social security spending? We have run the same analysis for these variables as for the budget deficit19 and found that their countercyclicality was not significant in explaining economic growth. This indicates that the cyclical behavior of automatic stabilizers is unlikely to fully account for our results: namely it is not the case that just increasing transfers and social security spending in recessions increases economic growth. What matters for growth is not the countercyclicality of spending per se (be it discretionary or not) but rather the degree to which this spending is financed by debt, namely the degree to which the government injects extra liquidity in the economy. Third, the reader may be interested in knowing what happens if we replace the AR(1) MCMC estimate of countercyclicality by the Gaussian-weighted or the ten-year rolling windows OLS. In the case of the Gaussian, the coefficients on the countercyclicality of the budget deficit and on its interaction with private credit have the same sign as in the reference specification and are significant at the 1 percent level. The only difference is that the value of the coefficient on the countercyclicality of the budget deficit is lower. In the case of the ten-year rolling windows method, the coefficients of interest are of the same sign, but are not statistically significant, which is not surprising since these estimates are much noisier. Fourth, one may still be skeptical about the causal interpretation of our estimates. As mentioned in section 8.2, our AR(1) MCMC estimate of countercyclicality should be, in principle, mostly uncorrelated with the future, reducing the endogeneity problem. First, to check whether indeed future countercyclicality is independent of growth, we include both the lag and the lead of the countercyclicality measure in the reference specification. Doing so does not significantly change the coefficient on the lagged countercyclicality but yields an insignificant and positive coefficient on the lead of procyclciality. These results are consistent with countercyclicality causing growth and not the reverse. Second, we noticed that inflation targeting is associated with a less countercyclical budget deficit (table 8.2) but is insignificant in explaining growth (table 8.3). This raises the possibility of using inflation targeting as an instrument for countercyclicality in a GMM framework. In the GMM estimation, we instrument both the countercyclicality variable and the lagged GDP per capita. For the latter, we use the classic instruments second and third lag of GDP per

Cyclical Budgetary Policy and Economic Growth

245

capita. Excluded instruments in our GMM regression are thus second and third lag of GDP per capita and the inflation targeting dummy. Moreover our GMM estimates allow for Newey errors of lag 1. We have thus re-estimated the reference specification using GMM. First-stage estimates are significant, but the explanatory power of inflation targeting for countercyclicality is limited. Overidentifying restrictions are not rejected by the J test. However, we do not reject that countercyclicality and its interaction with private credit are exogenous, which means that GMM is not more appropriate than OLS. The coefficients on countercyclicality and its interaction with private credit are of similar magnitudes as in the reference specification but they are not significant (P-values around 30 percent). This exercise thus confirms that our countercyclicality measure is unlikely to be endogenous. Finally, one may be interested in the time horizon of our effects: when the countercyclicality of the budget deficit changes in a given year, how far in the future does the effect on growth persist? One way to answer this question is to modify the reference specification by replacing the lag of the countercyclicality of the budget deficit, private credit over GDP and the interaction of the two by further lags. When using the second lag of these variables, the coefficients of interest (b1 and b3) are still significant and of the same sign, but the R2 diminishes slightly. When using the third lag of these variables, the coefficient on the countercyclicality of the deficit is still significant, but the interaction with private credit is no longer significant. Using even further lags makes the coefficients of interest become insignificant. Thus it seems that an increase in the countercyclicality of budgetary policy affects GDP growth up to two or three years later. 8.5

Conclusion

In this chapter we have analyzed the dynamics and determinants of the cyclicality of budgetary policy on a yearly panel of OECD countries, and the relationship between this cyclicality, financial development, and economic growth. Our findings can be summarized as follows: first, countercyclicality of budget deficits has generally increased over time. However, in EMU countries, the budget deficit became slightly less countercyclical. Second, countercyclicality of budgetary policy appears to be facilitated by a higher level of financial development, a lower degree of openness to trade, and a monetary policy committed

246

Philippe Aghion and Ioana Marinescu

to inflation targeting. Third, we found that countercyclical budget deficits are more positively associated with growth the lower the country’s level of financial development. The line of research pursued in this chapter bears potentially interesting growth policy implications. In particular, our second-stage regressions suggest that growth in EMU countries could be fostered if the budget deficit in the eurozone became more countercyclical. Our first stage regression suggests that this in turn could be partly achieved by having the EMU area move toward inflation targeting, for example, following the UK lead in this respect, and also by improving the coordination among finance ministers in the eurozone on fiscal policy over the cycle so as to make it become more countercyclical.20 The analysis in this chapter should be seen as one step in a broader research program. First, one could try to perform the same kind of analysis for other groups of countries, for example, middle income countries in Latin America or in central and eastern Europe. Second, one could take a similar AABM-type of approach to volatility, financial development and growth to further explore the relationship between growth and the conduct of monetary policy. For example, to which extent allowing for higher procyclicality of short term nominal interest rates, can help firms maintain R&D investments in recessions and/or improve governments’ ability to implement growth-enhancing countercyclical budgetary policies? Finally, one could investigate the possible interactions in growth regressions between countercyclical budgetary policy and structural reforms in the product and labor markets. Appendix: The AR(1) MCMC Method for Calculating Cyclicality in the First Stage The aim of this appendix is to give a brief description of how we used the Kalman filter together with Markov chain Monte Carlo methods (MCMC) in order to estimate the coefficients ajit from equation (8.1) under the assumption that they follow an AR(1) process as desribed by equation (8.4). The implementation was carried out in Matlab. Estimating the means and variances of the coefficients of interest— that is, ajit in equation (8.4)—involves two procedures: Kalman filtering21 and MCMC.

Cyclical Budgetary Policy and Economic Growth

247

To compute the coefficients with the Kalman filter for each country, we need to know the values of three variances: s 2aj in equation (8.4), for j = 1, 2, meaning the process variances in the terminology of the Kalman filter. •

s e2 of the error term et in equation (8.1), meaning the measurement error variance in the terminology of the Kalman filter. •

Moreover, to use the Kalman filter, we need a prior for the first period of observation for each country that is a specification of our expectation over the values ajit at the first time step. As we do not have any meaningful prior information about cyclicality at the first observed period, we use a very high variance around the prior mean so that this prior has a negligible effect on the estimates. Specifically, the set of initial values for the coefficients were chosen to be the OLS estimates of the coefficients using the first ten years of data for each country, and the value of the initial variance is set to be 100,000 times the estimated variance of these coefficients. However, the process variances s 2aj and the measurement error variance s 2e are unknown, and we do not have any meaningful prior over them. We therefore need a method to find reasonable values for these three unknown variances. This is where MCMC methods are useful. One can think of MCMC as the opposite of simulating. In the case of simulation we know the parameters of our process, for example, the variances, and every time we run a simulation program, it gives us a set of possible observed data. More specifically, the probability of getting any set of observed data is the probability defined by the model that we have and the parameters. MCMC is the opposite: we assume that we have a given dataset, and we are producing a set of possible parameters. This is done in such a fashion that the probability of accepting a parameter value is identical to the probability that this parameter value has actually produced the data. Specifically, in our implementation, we use the classic Metropolis– Hastings (MH) sampler to do MCMC (see Chib and Greenberg 1995 for an introduction to MCMC and Metropolis–Hastings,). In MH one starts with arbitrary parameters values. At every iteration one proposes a random change (in our case a small Gaussian change) of the parameters. This is what is called the proposal distribution. Subsequently this change is either accepted or rejected. The probability of acceptance is

248

Philippe Aghion and Ioana Marinescu

p (data new _ parameters ) ⎞ ⎛ paccept = min ⎜ 1, . ⎝ p (data previous _ parameters) ⎟⎠

(8A.1)

It is easy to prove that this procedure is actually sampling from the correct posterior distribution over the parameter values. MCMC algorithms go through two different stages. In the first stage the sampler converges to a probable interpretation of the data in terms of the parameters. This stage is called burn-in, and it took about 500 iterations in our case. Within these 500 iterations, probabilities increased dramatically and then converged to a stable high level. Afterward the MCMC algorithm is exploring the space of relevant parameters. Over three runs we took 10,000 samples per run after the end of burn-in. To avoid the autocorrelation that typically characterizes a Markov chain, we only retain samples at every 100 iterations in order to compute the final estimates. From these three runs we get a total of 300 essentially uncorrelated samples for each of the three parameters we wish to estimate. Convergence of the Markov chain was assessed comparing the within chain correlation with the across chain correlation. From these 300 samples we can then directly estimate means and variances of the three parameters of interest. In order to correctly infer the effect of cyclicality on growth in our second-stage regressions, we need to determine not only the value of the cyclicality (a1it) but also the uncertainty we have about it. To estimate this uncertainty, namely the standard deviation of the cyclicality estimates, it is necessary to consider the relevant sources of uncertainty. Two sources are relevant in our case. One is the uncertainty that is represented by the Kalman filter that stems from the finite number of noisy observations. The other source of uncertainty is uncertainty about the three parameters that are modeled by the MCMC process. To combine them, we use the approximation var iance total = var iance MCMC + var ianceKalman , where varianceKalman denotes the average variance over the 300 Kalman filter runs using the 300 samples that we retained from the MCMC estimates of the three variances. This approximation becomes correct if the variance as estimated by the Kalman filter is similar over different runs of the Markov chain, which was a good approximation for our data. Finally, a full general statistical description of the methods used here can be found in Kording-Marinescu (2006).

Cyclical Budgetary Policy and Economic Growth

249

Notes This work owes a lot to Robert Barro who contributed abundant advice, and to the very helpful comments and editorial suggestions of Daron Acemoglu, Olivier Blanchard, Ken Rogoff, and Michael Woodford. We also thank Ricardo Caballero and Anil Kashyap for their useful discussions. At earlier stages this project benefited from fruitful conversations with Philippe Bacchetta, Tim Besley, Laurence Bloch, Elie Cohen, Philippe Moutot, Jean Pisani-Ferry, Romain Ranciere, and of our colleagues in the Institutions, Organizations and Growth group at the Canadian Institute for Advanced Research. We are very grateful to Ann Helfman, Julian Kolev and Anne-Laure Piganeau for outstanding research assistance. Finally, we thank Konrad Kording for his collaboration on the first stage analysis section and more specifically for helping us implement the MCMC methodology. 1. For example, Lucas (1987) analyzes the welfare costs of income volatility in an economy with complete markets for individual insurance, taking the growth rate as given. Atkeson and Phelan (1994) analyze the welfare gains from countercyclical policy in an economy with incomplete insurance markets but no growth. Both find very small effects of volatility (or of countercyclical policies aimed at reducing it) on welfare. 2. As pointed out by several authors, some of these results may be biased because of an endogeneity problem that may come from the potential simultaneous determination of sales and investment. BEAAC check the robustness of their results by instrumenting the variation in sales by an exchange rate exposure variable that depends on exchange rate variations and firms’ export status. This variable is strongly correlated with sales variation without being affected by investment decisions. Their results are robust to this instrumentation. 3. That government intervention might increase aggregate efficiency in an economy subject to credit constraints and aggregate shocks has already been pointed out by Holmstrom and Tirole (1998). Our analysis in this section can be seen as a first attempt to explore potential empirical implications of this idea for the relationship between growth and public spending over the cycle. 4. Also Talvi and Vegh (2000) argue that high output volatility is most likely to generate a procyclical government spending. The idea is that running a budget surplus generates political pressures to spend more: the government therefore minimizes that surplus and becomes procyclical. This movement is then accentuated by a volatile output, and therefore a volatile tax base. 5. We have also used these indicators in our analysis. However, they typically have no significant effect on GDP growth over time in our sample. Moreover, as they are less widely available than our main variables of interest, their use considerably restricts the available sample, leading to less precise estimates. We have therefore decided not to use these indicators in the results reported here. 6. Codes in parenthesis indicate the names of variables in the dataset. Full documentation available at www.oecd.org. Data can be downloaded from sourceoecd.org for subscribers to that service. 7. Data downloadable from Ross Levine’s homepage. 8. All level variables are adjusted for the German reunification. The adjustment involves regressing each variable of interest on time and a constant in the ten years before 1991

250

Philippe Aghion and Ioana Marinescu

(data based on West Germany only). We then use the estimated coefficients to predict the values for 1991 to 2000. We take the average ratio between actual and predicted values in the years 1991 to 2000. We use this ratio to proportionally adjust values before 1991. 9. The effect of a decrease in the countercyclicality of public deficit could become negative at high enough levels of financial development, if the government’s deficit crowds out more efficient private borrowing and spending. 10. The constant a2it can be interpreted as a measure of structural budgetary deficit: indeed, by construction, it corresponds to the part of budget deficit that does not depend upon the business cycle. 11. In practice, we chose a value of 5 for s. While this choice is somewhat arbitrary, changing this smoothing value slightly does not qualitatively affect the results. 12. This assumption is reasonable since the OECD countries in our sample share similar institutions and degrees of economic development. Moreover this assumption is similar to assuming no heteroskedasiticty across panels when estimating a panel regression, which is the standard assumption. Finally, assuming country-specific variances would make estimates much more imprecise due to the fact that our relatively small number of observations would have to be used to identify many more parameters. 13. See the appendix for more details on the implementation of this method. 14. It is indeed also possible to use maximum likelihood type methods to estimate the variances, but these are precisely liable to get stuck in local solutions. In a previous version of this chapter, we used such a method, amended so that it does not systematically get stuck in a local solution. In practice, the estimates of the coefficients ajit that we had obtained using that method are highly correlated with the ones obtained here using MCMC. 15. The coefficients also depend on the future in as much as their variance is calculated using the full sample of available observations. 16. As mentioned previously, using ICRG indicators turns out not to be of interest for our analysis. 17. This dummy variable takes a value of 1 for all countries that currently belong to the EMU, and 0 for all the other countries. This is because the EMU has been prepared for many years so that the countries that would eventually join might be different even before the EMU is fully effective. 18. We have experimented with an interaction between the EMU dummy and a post1999 dummy, but this interaction was typically insignificant, indicating that there is no substantial change occurring with the full implementation of the EMU in 1999. 19. Specifically, in equation (8.1), we replaced the first difference of debt by the first difference of each of these variables. 20. The Sapir report (Sapir et al. 2003) recommended the setting-up of “rainy day” funds supervised by the European Commission. 21. For an excellent overview of the Kalman filter and smoother, see the notes by Max Welling “Kalman Filters” available on the web at http://www.ics.uci.edu/~welling/ classnotes/classnotes.html.

Cyclical Budgetary Policy and Economic Growth

251

References Acemoglu, D., S. Johnson, J. Robinson, and Y. Thaicharoen. 2003. Institutional causes, macroeconomic symptoms: Volatility, crises, and growth. Journal of Monetary Economics 50: 49–123. Aghion, P., M. Angeletos, A. Banerjee, and K. Manova. 2005. Volatility and growth: Credit constraints and productivity-enhancing investment. Working paper 11349. NBER, Cambridge, MA. Aghion, P., P. Bacchetta, R. Ranciere, and K. Rogoff. 2006. Exchange rate volatility and productivity growth: The role of financial development. Working paper 12117. NBER, Cambridge, MA. Aghion, P., D. Hemous, and E. Kharroubi. 2010. Cyclical budgetary policy, credit constraints, and industry growth. Mimeo. Harvard University. Alesina, A., and R. Perotti. 1996. Fiscal adjustments in OECD countries: Composition and macroeconomic effects. Working paper 5730. NBER, Cambridge, MA. Alesina, A., and G. Tabellini. 2005. Why is fiscal policy often procyclical? Working paper 11600. NBER, Cambridge, MA. Atkeson, A., and C. Phelan. 1994. Reconsidering the costs of business cycles with incomplete markets. NBER Macroeconomics Annual 9: 187–207. Berman, N., L. Eymard, P. Aghion, P. Askenazy, and G. Cette. 2007. Credit constraints and cyclical R&D investment: Evidence from French firm-level panel data. Mimeo. Banque de France, Paris. Barro, R. 1979. On the determination of public debt. Journal of Political Economy 87: 940–71. Batalgi, B. 2001. Econometric Analysis of Panel Data, 2nd ed. New York: Wiley. Berman, N., L. Eymard, P. Aghion, P. Askenazy, and G. Cette. 2007. Credit constraints and the cyclicality of R&D investment: Evidence from French firm-level panel data. Working paper 2007–4. Weatherhead Center for International Affairs, Harvard University. Calderon, C., R. Duncan, and K. Schmidt-Hebbel. 2004. Institutions and cyclical properties of macroeconomic policies. Working paper 285. Central Bank of Chile. Santiago. Chib, S., and E. Greenberg. 1995. Understanding the Metropolis–Hastings algorithm. The American Statistician 49: 327–35. Easterly, W. 2005. National policies and economic growth: A reappraisal. In P. Aghion and S. Durlauf, eds., Handbook of Economic Growth. Amsterdam: Elsevier/North-Holland, 1015–59. Gali, J., and R. Perotti. 2003. Fiscal policy and monetary integration in Europe. Economic Policy 18: 533–72. Giorno, G., P. Richardson, D. Roseveare, and P. van den Noord. 1995. Estimating potential output, output gaps and structural budget balances. Working paper 152. OECD Economics Department, OECD, Paris.

252

Philippe Aghion and Ioana Marinescu

Hallerberg, M., R. Strauch, and J. von Hagen. 2004. The design of fiscal rules and forms of governance in European Union countries. Working paper 419. European Central Bank, Berlin. Kording, K., and I. Marinescu. 2006. Sampling approaches to Kalman learning. Work in progress. Lane, P. 2003. The cyclical behavior of fiscal policy: Evidence from the OECD. Journal of Public Economics 87: 2661–75. Lane, P., and A. Tornell. 1998. Why aren’t Latin American savings rates procyclical? Journal of Development Economics 57: 185–200. Lucas, R. 1987. Models of Business Cycles. New York: Basil Blackwell. Talvi, E., and C. Vegh. 2000. Tax base variability and procyclical fiscal policy. Working paper 7499. NBER, Cambridge, MA. Vega, M., and D. Winkelried. 2005. Inflation targeting and inflation behavior: A successful story? International Journal of Central Banking 1 (3): 153–75.

9

Policies to Create and Destroy Human Capital in Europe James J. Heckman and Bas Jacobs

9.1

Introduction

The labor market prospects of unskilled workers are jeopardized by skillbiased technical changes and the globalization of the world’s production activities (e.g., see Katz and Autor 1999). Moreover some scholars argue that the position of unskilled workers has become more vulnerable in recent, more turbulent labor markets (Ljungqvist and Sargent 1998, 2002). As the relative demand for unskilled labor diminishes, and governments or unions attempt to protect workers with low skills through labor market regulations or minimum wages, nonemployment among the unskilled will increase (Bertola 2003). Raising payroll taxes to support such efforts also reduces demand. Reinvention of human capital policy is required to combat the emergence of an underclass. All available evidence shows that welfare state dependency in Europe is heavily concentrated among unskilled persons. For example, unskilled persons have higher unemployment rates, higher take-up rates of welfare benefits, and larger participation rates in active labor market programs (OECD 2005a, 2006a, b). In addition many social problems are associated with lack of skill, such as deviant social behavior (drug use), working in the unofficial economy, criminal behavior, teenage pregnancies, and so on (European Commission 2005). Social cohesion may be undermined further as migrant populations are predominantly low skilled and their welfare dependency rates are high. Another feature of European labor markets is that European human capital stocks remain idle during large parts of the life cycle due to nonemployment and early retirement. Human capital needs be utilized and maintained over the life cycle for human capital investments to earn a sufficiently high return. High levels of taxation, generous social benefits, and strong labor market regulations reduce labor force

254

James J. Heckman and Bas Jacobs

participation rates, hours worked, and employment and thereby lower the utilization rates of human capital. Generous early retirement and pension schemes encourage older people to retire many years before statutory retirement ages (Gruber and Wise 1999). Low labor force participation rates of older workers imply that the time horizons over which investments in human capital are harvested are short. In addition there are often weak economic incentives to maintain skills through training on the job. Insurance schemes for disability, unemployment, and sickness create important moral hazard problems. Once out of work, older workers will often never be able to find a new job. Weak incentives to utilize and maintain skills over the life cycle become manifest with the ageing of the population. The utilization rate of European human capital will fall and substantial parts of human capital stocks will be written off as workers retire. Reinvention of human capital policies is required for both efficiency and equity reasons. In order to address the challenges imposed by skillbiased labor demand shifts resulting in larger wage premiums for skilled workers, investments in human capital should expand. Such a policy also helps to contain the growing divide between the skilled and the unskilled. Policy makers should acknowledge strong dynamic complementarities in skill formation. Governments should put a strong emphasis on interventions early on in the life cycle. Once individuals drop out of secondary school, labor market institutions or government policies often prevent them from finding employment at established wage minimums. If one wishes to maintain high levels of minimum income support and redistribution toward the poor, human capital policy is more urgent than ever to avoid increasing dependency on welfare states. Only when individuals acquire sufficient human capital at the beginning of their life cycles, they can avoid getting stuck in poverty and productivity traps later on in life. There is no trade-off between equity and efficiency at early ages of human development, but there is a substantial trade-off at later ages. Later remediation of skill deficits that are formed in early years is often ineffective. Active labor market and training policies should therefore be reformulated. In addition policies to foster human capital cannot be seen in isolation from labor market policies, tax and benefit systems, and pension schemes. Current welfare state arrangements often create substantial implicit tax burdens on human capital investments because the incentives for investments in human capital are undermined by low utilization rates of human capital and short time horizons over which

Policies to Create and Destroy Human Capital in Europe

255

investments in skill materialize. Labor force participation, hours worked, training on the job, and later retirement are all complementary to human capital investments. High marginal tax rates and generous benefit systems reduce labor force participation rates and hours worked and thereby lower the utilization rate of human capital. Tax-benefit systems should be reconsidered as they increasingly redistribute resources from outsiders to insiders in labor markets, which is both distortionary and inequitable. Early retirement and pension schemes should be made actuarially fairer as they entail strong incentives to retire early, and human capital is thus written off too quickly. Reforms in labor markets, pension systems and tax-benefit systems might thus not only have beneficial static effects on labor market performance but also important dynamic efficiency gains by lowering implicit tax wedges on skill formation over the life cycle. We ground our policy analysis in insights from previous research on the technology of skill formation (Cunha, Heckman, Lochner, and Masterov 2006; Cunha and Heckman 2007; Heckman 2007). Human capital accumulation is a dynamic process. The skills acquired in one stage of the life cycle affect both the initial conditions and the technology of learning at the next stage. Human capital is produced over the life cycle by families, schools, and firms. Different stages of the life cycle are critical to the formation of different types of abilities. When the opportunities for formation of these abilities are missed, remediation is costly, and full remediation is often prohibitively costly. These findings highlight the need to take a comprehensive view of skill formation over the life cycle so that effective policies for increasing the low level of skills in the workforce can be devised. This chapter extends this line of reasoning to the entire life cycle. We argue that, due to the same dynamic complementarities in skill formation over the life cycle, skill formation is impaired when the returns to skill formation are low due to low skill use and insufficient skill maintenance later on in life. We develop a theory of earnings, schooling, training, and retirement that is capable of describing some stylized features of Europe’s labor markets and illustrates the impact of various policies. The consequences of low skill formation both in the AngloSaxon world and mainland Europe are equally present. However, when it comes to skill use and skill maintenance, we show that mainland Europe differs markedly from the Anglo-Saxon world due to low skill use and poor skill maintenance. Europe’s future problems with low skills are therefore exacerbated by labor market institutions and

256

James J. Heckman and Bas Jacobs

government policies that lower utilization rates of human capital and promote steep depreciation of human capital over the life cycle. 9.2 Stylized Facts on Skill Formation, Skill Use, and Skill Maintenance in Europe This section describes in detail some salient stylized facts on inequality, skill formation, skill use, and skill maintenance for various countries and in developments over time. Wherever possible we distinguish among the Anglo-Saxon countries (United Kingdom, Australia, New Zealand, Canada, and the United States), the Nordic countries (Denmark, Norway, Sweden, and Finland), continental European countries (Netherlands, Belgium, France, and Germany), and Mediterranean countries (Portugal, Spain, Italy, and Greece). 9.2.1

Economic Environment

Growing Earnings and Income Inequality Davis (1992), Gottschalk and Smeeding (1996), Katz and Autor (1999), and Brandolini and Smeeding (2006) analyze trends in earnings inequality and conclude that inequality had been steadily increasing in Western countries during the last decades of the twentieth century, although the rapid growth in the 1980s appears to level off in the 1990s.1 The increase in inequality was most notable in the Anglo-Saxon countries. The Nordic countries appeared to have contained the increase in inequality. As noted by Bertola (2003) and Atkinson (2008), the rise in inequality countries is mainly concentrated in the upper part of the earnings distribution and not so much in the lower part of the earnings distribution. Gottschalk and Smeeding (1996) and Brandolini and Smeeding (2006) show that inequality in net disposable household income did increase as well but to a much lesser extent than labor earnings. Piketty and Saez (2003) and Atkinson and Salverda (2005) document large increases in earnings inequality at the very top of the income distribution for the United States and the United Kingdom. Piketty (2003) and Atkinson and Salverda (2005) show that the income distribution at the top did not change much in France and the Netherlands. Minimum Wages, Inequality, and Trade-off between Equality and Incentives in the Labor Market Falling real minimum wages appear to have caused growing inequality at the bottom of the earnings dis-

Policies to Create and Destroy Human Capital in Europe

257

tribution in the United States (see also Autor, Katz, and Kearny 2008). Minimum wages reduce inequality for workers at the cost of lower employment. Indeed inequality at the bottom of the earnings distribution did not increase much for European countries, but unemployment rates among low-income earners have been increasing instead (Davis,1992). Bertola (2003) shows that employment declines (unemployment rates increase) especially in those countries where inequality in the lower part of the earnings distribution remained rather constant. Increases in unemployment rates are disguised to an important extent by enrolling unemployed workers in active labor market and training programs. Figure 9.1 from Heckman, Ljunge, and Ragan (2006) shows that many European countries and especially corporatist countries (Denmark, Finland, Norway, and the Netherlands) almost halve their open unemployment rates by placing more unemployed workers in Active Labor Market (ALM) programs, where they are no longer counted as unemployed workers. Adding these trainees back to the unemployed substantially boosts the unemployment rate. Trade-off between Equality and Incentives for Human Capital Investment Low inequality may not only be a virtue, but may also be a vice 9.0 8.0

Percentage points

7.0 6.0 5.0 4.0 3.0 2.0 1.0

iu

m

s

lg

nd rla

Be

en he et N

m

ed

Sw

ar k

nd

nd

la

en D

la

Ire

Fi

er itz

nl a

nd

y an m Sw

er

st ra lia

G

m

Ki te d

U

ni

Au

e

ng do

ec re G

ze ch C

N

or

w

ay

0.0

Figure 9.1 Differences between open and full employment, 1998 to 2004, averages. The difference between the true (“full”) unemployment rate and the reported (“open”) unemployment rate is due to disguising unemployed persons as trainees (source: Heckman, Ljunge, and Ragan 2006).

258

James J. Heckman and Bas Jacobs

Tertiary attainment 25 to 34 cohorts

50

40

AU FR AU AUFR FR AU FR NL NL NL AU FR NL AU NL FR AU AU NL FR DE DE DEDE FR DE DE

30 NO

BE BE

20

10

2.5

CA CA US US

GB GB GB GBGB GBGB GB GB

ES

PT

IT IT IT ITIT

2

USUS USUS US US US

CA

FI FI FIFI FI FI FI FI

3

3.5

4

4.5

P90/P10 earnings ratio Figure 9.2 Eamings inequality and higher educational attainment (source: OECD 2006a, c)

when low inequality reflects weak economic incentives. Figure 9.2 plots higher educational attainment rates of 25- to 34-year-old cohorts against earnings inequality as measured by the 90/10 percentile ratio. Both variables are taken from the OECD Labor Force Database. A clear positive correlation emerges between earnings inequality and higher educational attainment. This positive correlation remains robust using tertiary educational attainment rates of 25- to 65-year-old cohorts, employing 90/50 or 50/10 percentile ratios for inequality or doing panel regressions that allow for country-specific fixed effects. There is not only a trade-off between the quantity of employment and equality but also between the quality of employment and inequality. More compressed wage distributions imply weaker incentives for skill formation. Frederiksson (1997) is one of the few studies that directly estimates the effect of a larger skill-premium on enrollment and finds very substantial effects for Sweden. The empirical general equilibrium model for the United States of Heckman, Lochner, and Taber (1998) also predicts a quite elastic response of investments in human capital to larger skill premia.2 Rising Returns to Education Income inequality is increasing in part because the returns to education display an upward trend. Studies for the United States have documented a strong and steady increase in the

Policies to Create and Destroy Human Capital in Europe

259

college-premium during the 1980s and 1990s (Katz and Autor 1999; Autor, Katz, and Kearny 2008). Peracchi (2006) reviews a large number of country studies and shows that, in general, skill premia have been constant or increasing in recent years for most Western countries. Gottschalk and Smeeding (1996) in their cross-country comparison find that an important driving force behind growing earnings inequality is the growing skill premium. Using a panel of selected OECD countries, Nahuis and De Groot (2003) show that rising skill premia during the 1980s and 1990s are not only present in the United States but in the whole of the Western world.3 By now there is a firmly established consensus that the mean rate of return to a year of schooling, as of the 1990s, exceeds 10 percent and may be as high as 17 to 20 percent (see Heckman, Lochner, and Todd 2006). This return is higher for more able people (Taber 2001; Carneiro and Heckman 2003) and for children from better socioeconomic backgrounds (Altonji and Dunn 1996 present some evidence in support of this claim but their own interpretation is more equivocal). Those from better backgrounds and with higher ability are also more likely to attend college and earn a higher rate of return from doing so. This evidence is robust to alternative choices of instrumental variables and to the use of alternative methods for controlling for self-selection. Both cognitive and noncognitive skills raise earnings through promoting schooling and through their direct effects on earnings (see the evidence in Taber 2001; Heckman, Hsee, and Rubinstein 2001; Carneiro, Hansen, and Heckman 2001; 2003; Cunha, Heckman, Lochner, and Masterov 2006; Heckman, Stixrud, and Urzua 2006; Borghans et al. 2008).4 9.2.2

Skill Creation

Slowing Down of the Growth in Supply of Skills Educational attainment has grown enormously in most of the Western world. Figure 9.3 plots higher educational attainment rates (as a fraction of each birth cohort) over the 1960 to 1995 period for various countries from the De la Fuente and Doménech (2006) data set.5 Enrollment rates doubled virtually everywhere. The development in the average number of years of education is similar (not shown). A striking feature is the large heterogeneity between countries in higher educational attainment. The Mediterranean countries lag miles behind the Nordic and Anglo-Saxon countries. The continental European countries are somewhere in the middle. We have to note here that institutional differences between

Percent of population

260

James J. Heckman and Bas Jacobs

50 45 40 35 30 25 20 15 10 5 0 1960

1965

1970

Percent of population

Netherlands

1975

1980

Belgium

1985 France

1990

1995

Germany

50 45 40 35 30 25 20 15 10 5 0 1960

1965

1970

Denmark

1975 Norway

1980

1985 Sweden

1990

1995

Finland

Figure 9.3 Higher education attainment rates as percentage of each birth cohort across countries over time (source: De la Fuente and Doménech 2006)

countries make good comparisons difficult due to, for example, differences in the duration of higher educational programs. Literacy scores indicate that high levels of educational attainment in some countries do not necessarily match with high levels of literacy (Heckman and Jacobs 2006). Education systems differ across different countries, and these comparable tests may provide a better measure of the stock of skills of a country, at least for the purpose of international comparisons. Hanushek and Kimko (2000) use these tests as a measure of the quality of the labor force and argue that these are an important determinant of economic growth. The massive increase in the level of education of Europe’s workforces probably cannot be maintained indefinitely. Figure 9.4 shows that there are strongly decreasing returns to raising education levels as the growth rate of in education levels during 1960 to 1995 is negatively correlated with the initial level of education in 1960. Therefore one can

Percent of population

Policies to Create and Destroy Human Capital in Europe

50 45 40 35 30 25 20 15 10 5 0

1960

1965

1970

Percent of population

Portugal

1975

1980

Spain

1985

1990

Italy

261

1995

Greece

50 45 40 35 30 25 20 15 10 5 0 1960

1965

1970

United Kingdom

1975

1980

Australia

1985

1990

Canada

1995

United States

Figure 9.3 Continued

Average growth rate per year

0.012 0.01

Growth(1960–1995) = -0.001*stock(1960) + 0.015 R 2 = 0.817

IT GR

PT ES

FI

0.008 IE

0.006

NL BE SW FR

JP

ADE

AT

GB

0.004

NO

CA US GH NZ DK

0.002 0 4

5

6

7

8

9

10

11

12

Average years of initial education 1960 Figure 9.4 Growth rate of human capital stocks, 1960 to 1990/95, and initial levels of education (source: De la Fuente and Doménech 2006)

262

James J. Heckman and Bas Jacobs

expect that the returns to education will be rising in the years to come because the demand for college-educated workers outstrips supply. Nonincreasing Resources Invested in Human Capital Resources invested in human capital in Europe also remain rather stagnant despite the rising returns. Overall investment levels as a fraction of GDP do not change much over time in continental and Mediterranean countries; see figure 9.5. Notable are the decreases in some countries (Finland and Norway). The financial resources per student in higher education invested are again roughly constant in continental European and Mediterranean countries. In Anglo-Saxon and Nordic countries, resources invested in higher education have increased in recent years (see OECD 2005a). Figure 9.6 plots the share of private contributions to the direct costs of education for different countries. Virtually all European countries heavily rely on state funding for education and that is probably also the reason why public budgets haven’t kept pace with increasing enrollment rates in some countries. Primary and secondary education are generally free everywhere. As regards higher education, tuition is subsidized and students receive (means tested) grants and study loans with interest subsidies. Again, only the Anglo-Saxon countries have a nontrivial share of private investments, whereas especially the Nordic and Mediterranean countries almost exclusively rely on state funding for education. Although larger skill premia would naturally give stronger incentives to invest in more human capital, it is by no means certain that this will also happen in the stiffly regulated higher education sectors in Europe. Due to the aging of the population and the EMU criteria for deficits and debts, most government budgets are under pressure. There is currently not much hope for extra public funding whereas there are good reasons for more private investments in higher education. Nevertheless, most governments obstruct private funding by repeatedly raising accessibility issues and failing to understand the basic incentive issues facing schools and students (see Jacobs and Van der Ploeg 2006). Private funding of education should expand if governments do not free up enough resources for investment in human capital, especially in higher education. There is a strong efficiency case for public support in primary education in well-developed welfare states. Poverty traps create not only large tax burdens on work effort but also on skill formation; if it does not pay work it does not pay to invest in skills either.

0.09

0.08

0.07

0.06

0.05

0.04 1992

1993

1994

1995

1996

1997

1998

1999

2000

Netherlands

France

Germany

Portugal

Spain

Italy

2001

2002

2001

2002

0.09

0.08

0.07

0.06

0.05

0.04

1992

1993

1994

1995

1996

1997

1998

1999

2000

Denmark

Norway

Sweden

Australia

Canada

United States

Finland

Figure 9.5 Total expenditure on education (percent of GDP) with change in total resources invested in primary, secondary, and tertiary education across countries over time (source: OECD 2005b, c)

264

James J. Heckman and Bas Jacobs

100 90

Percent

80 70 60 50

Pre-primary education

ar k N or w ay Sw ed en U F in ni l te an d d Ki ng do Au m st ra l C ia an U ad ni te a d St at es

e

m

ec

ly Ita

G re

D en

N et

he

rla n Be ds lg iu m Fr an ce G er m an Po y rtu ga l Sp ai n

40

Primary, secondary, and post-secondary nontertiary education

Tertiary education

Figure 9.6 Share of public expenditures in total direct expenditures on education in 2002; Canada values 1995 (source: OECD 2005a)

If governments do not want to dismantle welfare states and income support programs for the poor, they should ensure that sufficient investments in human capital lift vulnerable groups above the minimum income floors. This public support may come in the form of public funding for primary and secondary education, but also in the form of minimum school leaving ages and outlawing child labor (Bovenberg and Jacobs 2003). Skill-Biased Demand for Labor The sharp increase in educational attainment in some countries in Europe has put downward pressures on skill premia. That is probably why returns to education have not been rising so much in some European countries as opposed to the United States where growth in the supply of skilled workers choked off already in the 1990s (Carneiro and Heckman 2003; Autor, Katz, and Kearney 2008). Nevertheless, overall wage inequality did increase and returns to education certainly did not decrease to a large extent. Gottschalk and Smeeding (1996) and Peracchi (1996)

Policies to Create and Destroy Human Capital in Europe

265

show that skill premia remained rather constant in most countries and generally increased in recent years. Everywhere in Europe labor markets have absorbed the enormous influx of skilled workers without large reductions in skill premia. In other words, the demand for skilled workers has been increasing at the same or even higher speed than the supply of skilled workers. Many explanations have been put forward for these labor demand shifts but skill-biased technical changes appears to be the most important one. Increasing international trade and capital-skill complementarities could be supplementary explanations for this phenomenon (see also Katz and Autor 1999). The outward shift of relative demand for skilled workers is radically transforming labor markets and economies. To get an impression of its quantitative importance one may ask the counterfactual question how much wage differentials would have increased had the supply for skilled workers remained constant. In the US wage differentials between skilled and unskilled workers would have increased at a rate of 3 percent per year (Katz and Murphy 1992) and about the same is found for Canada (Murphy, Riddell, and Romer 1998). Jacobs (2004) documents a skill bias of about 2 percent per year in the Netherlands. A skill bias of a 1 percent increase in college premium per year is found in Sweden by Edin and Holmlund (1995). Low-Skilled Workers Have Weak Incentives to Train Unskilled individuals receive little training on the job, either because they opt out of it when it is offered to them or because employers choose to offer training to workers with better skills. This is illustrated in figure 9.7, from OECD (2003), which shows the proportion of people at each literacy level who receive job training. A low score signifies a low level of literacy. As emphasized by Carneiro and Heckman (2003), there are strong complementarities between early human capital investments and adult human capital investments. Low-skilled workers have difficulty in benefiting from adult training because they have a low stock of human capital on which adult investments can build and be productive. Remediation investments in adulthood are very costly and ineffective for low-skilled individuals (Knudsen, Heckman, Cameron, and Shonkoff 2006; Cunha, Heckman, Lochner, and Masterov 2006). Preventive investments that take place earlier in the life cycle of individuals generate much larger returns.

266

James J. Heckman and Bas Jacobs

100 Total participation rate

Level 1

Level 2

Level 3

Level 4/5

Percent

80

60

40

20

Fi nl D and en m a Sw rk ed en N No ew rw a U ni Ze y te d alan Ki n d Sw gd om i U tze rla ni te d nd St a Au tes st ra C lia N ana et he da rla nd C ze Slo s ch ve R nia ep Be ub lg lic iu m Ire (F lan la nd d e H rs) un ga ry C hi Po le rtu g Po al la nd

0

Figure 9.7 Literacy and adult education participation, showing percentage of population aged 16 to 65 participating in adult education and training during the year preceding the interview at each literacy level and in total (source: International Adult Literacy Survey 1994–1998, OECD 2003). Document scales 1994 to 1998; countries are ranked by total population rate. Document literacy measures the knowledge and skills required to locate and use information contained in various formats, including job applications, payroll forms, transportation schedules, maps, tables, and graphics. Training is measured as enrollment in some form of organized adult education or training during the year preceding the interview.

Large Spending Labor Market Programs Many European governments spend large amounts of resources on active and passive labor market programs, as demonstrated in figure 9.8. continental European countries lead in total spending, followed by the Nordics. Mediterranean countries have some labor market programs, especially Spain. Anglo-Saxon countries have virtually no labor market programs compared to the rest. Below we will argue that these programs are largely ineffective in lifting individuals out of poverty and raising their standards of living. Taxes, Subsidies and the Incentives to Acquire Skills Flat labor income taxes do not harm skill formation as long as all costs are subsidized or deductible at the flat income tax rate. Direct costs and the opportunity costs of education—forgone labor earnings while in edu-

Policies to Create and Destroy Human Capital in Europe

267

5 2001 2002 2003 2004

4.5

Percent of GDP

4 3.5 3 2.5 2 1.5 1 0.5 e

ar m

D en

ly

ec

Ita

G re

k N or w a Sw y ed en U Fi ni n te la d Ki nd ng do Au m st ra lia C an U ni te ada d St at es

N

et

he

rla n Be ds lg iu m Fr an ce G er m an Po y rtu ga l Sp ai n

0

Figure 9.8 Total expenditure on training and passive/active labor market programs as percentage of GDP (source: OECD 2006a)

cation—are then taxed at the same rate as the future labor earnings (Heckman 1976). Only if marginal costs are taxed at lower rates than the marginal benefits, tax distortions on skill formation emerge. If marginal tax rates on labor incomes are increasing with income, future earnings are taxed at higher rates than forgone labor earnings and taxation discourages investment in human capital. This is the case in most European countries (see also figure 9.16, which shows that Musgrave and Musgrave’s coefficient of residual income progression is generally below one).6 Further, if education requires nondeductible expenses or effort costs, labor taxation reduces investment in human capital. Education expenses for formal schooling or training are generally not deductible for the income tax. Some exceptions occur in Italy, the Netherlands and Portugal (see also Gordon and Tchilinguirian 1998). Large subsidies on education and training do however correct for tax disincentives on skill formation (Bovenberg and Jacobs 2005). Indeed many governments seem to oversubsidize higher education from a fiscal perspective; namely there is a net subsidy rather than a net tax on education and training (De la Fuente and Jimeno-Serrano 2005; Bovenberg and Jacobs 2005). Costs of training on the job are generally deductible by firms.

268

James J. Heckman and Bas Jacobs

Nonpecuniary costs and benefits escape the tax system and cannot be subsidized either. Given the high returns on human capital investments, one is tempted to conclude that nonpecuniary costs of education appear to be empirically more important than the nonpecuniary benefits. Findings by Carneiro et al. (2001, 2003) and Cunha, Heckman, and Navarro (2005) surveyed in Cunha, Heckman, Lochner, and Masterov (2006) suggest that nonpecuniary costs can be very important indeed. Therefore it can still be the case that taxation distorts skill formation even though direct costs are heavily subsidized. Additionally, large subsidies on observable inputs in human capital formation (e.g., years enrolled in education) will crowd out nonsubsidized complementary inputs in human capital formation like study effort (see Bovenberg and Jacobs 2005; Jacobs 2007). High subsidies on education may then go hand in hand with long study durations, high dropout rates and low student performance. Finally, labor income taxation depresses labor supply and thereby the utilization rate of human capital. Consequently labor income taxation indirectly depresses human capital investments, even if all costs are deductible and labor taxes are flat (see also Jacobs 2005, 2007). Slowing Growth in Skills Lowers Productivity Growth Researchers have established a robust, causal relation between education and earnings at the microeconomic level (e.g., see Card 1999; Harmon, Oosterbeek, and Walker 2003; Heckman, Lochner, and Todd 2006), although there is a debate about the magnitude of the relationship. A growing body of evidence suggests that the macroeconomic returns are of the same magnitude as the conventional microeconomic estimates (Heckman and Klenow 1998; Sianesi and Van Reenen 2002; Krueger and Lindahl 2001; De la Fuente and Doménech 2006; Ciccone and Peri 2006). The fact that education appears to be roughly equally productive at the macroeconomic level as at the micro level largely disqualifies the “signalling hypothesis” as put forward by Arrow (1973).7 Skills are crucial determinants of labor productivity. These findings can also be taken as evidence that—at current levels of public spending—external effects of education are absent. Figure 9.9 gives the average annualized growth rates of labor productivity during the last twenty years. This graph shows that the continental European and Mediterranean countries have witnessed the lowest rates of productivity growth. Unsurprisingly, the countries with large levels of investment in human capital (Nordics and Anglo-Saxon

Policies to Create and Destroy Human Capital in Europe

269

3 2.5

Percent

2 1.5 1 0.5

m

an Fr

iu lg

Be

ce G er m an Po y rtu ga l Sp ai n Ita ly G re ec e D en m ar k N or w ay Sw ed en U F ni te inla d nd Ki ng do Au m st ra lia C an U ad ni te a d St at es

N

et

he

rla

nd

s

0

Figure 9.9 Annualized percentage increase in labor productivity growth, 1993 to 2006 (source: OECD 2005b)

countries) appear to generate the highest levels of productivity growth. A slowdown in the rate of skill acquisition therefore appears to threaten the standards of living of future generations. 9.2.3

Skill Utilization

Small Labor Force Attachment Reduces the Utilization Rate of Human Capital A possible reason for low average returns to education is that labor force participation rates are low. This causes acquired human capital to remain idle. Hence a lower utilization rate of human capital reduces the returns to investments in schooling and training.8 Figure 9.10 shows that labor force participation rates are lowest in Mediterranean and continental European countries. Nordic and AngloSaxon countries have higher participation rates. However, labor force attainment has been increasing in recent years in many countries as women started to participate in especially the Nordic and continental European countries. Growth in participation rates was more modest in Anglo-Saxon countries where participation rates were already high. Mediterranean countries seem to be stuck in a trap of low participation rates.

270

James J. Heckman and Bas Jacobs

Percent of population

70 65 60 55 50 45 1956 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2003 Netherlands

Belgium

France

Germany

Percent of population

70 65 60 55 50 45 1956 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2003 Denmark

Norway

Sweden

Finland

Figure 9.10 Labor force participation rates, 1951 to 2003 (source: OECD 2006a)

Labor force attachment strongly increases with the level of education as can be seen from figure 9.11. This pattern holds across all countries. Although often overlooked in the human capital literature, an important benefit of higher skills is therefore the increased labor force attachment of more educated workers, especially in Europe where employment rates are low. The mirror image is that there is a strong dependence of unemployment on the level of education (graph not shown). As already noted before, official unemployment statistics are misleading because many countries enroll unemployed workers in ALM programs (see Heckman, Ljunge, and Ragan 2006: fig. 1). The latter authors show that real unemployment rates can be twice as high as official statistics suggest. Some corporatist countries may just be good in hiding unemployment. European countries have also hidden a lot of unemployment in sickness and disability benefits. Figure 9.12 gives the nonemployment rates of disabled workers as a fraction of the population aged 20 to 64. This depressing picture shows that a fraction of around 8 percent of the

Policies to Create and Destroy Human Capital in Europe

271

Percent of population

70 65 60 55 50 45 1956 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 Portugal

Spain

Italy

Greece

Percent of population

70 65 60 55 50 45 1956 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 United Kingdom

Australia

Canada

United States

Figure 9.10 Continued

population aged 20 to 64 receives a sickness or disability benefit and does not work (full time). Large fractions of workforces are disabled or sick and not participating in the labor market especially in the Netherlands, Germany, Portugal, Denmark, Sweden, and the United Kingdom, where the rate of sickness/disability is around 10 percent or higher. Disability related expenditures are especially high in the Netherlands and some Nordic countries (between 4 and 5 percent GDP). The other continental European countries and the Anglo-Saxon countries do not have a lot of disability spending (between 1 and 2 percent of GDP; OECD 2005c). Falling Working Hours Lower Utilization Rates of Human Capital Not only is the employment rate of European workforces low in comparison to the Anglo-Saxon world, but also hours worked. This fact is documented many times and one of the suggested reasons is the high level of taxation. However, also unionization of labor markets and collective labor agreements on reductions in working hours and holidays

100 90

Employment rate (%)

80 70 60 50 40 30 20 10

ar k N or w ay Sw ed en U Fi ni nl te a d Ki nd ng do m Au st ra l C ia an U ad ni te a d St at es

e

m

ec

ly Ita

G re

D en

N et

he

rla n Be ds lg iu m Fr an c G er e m an y Po rtu ga l Sp ai n

0

Primary education or lower Lower secondary education Higher secondary vocational (short) Higher secondary vocational (long) Higher secondary general Higher education (short) Higher education (long)

Figure 9.11 Employment rates by level of education, 2003 (source: OECD 2005a)

12

Percent of labor force

10 8 6 4 2

k N or w ay Sw U ed ni te en d Ki ng do m Au st ra lia C an U ad ni a te d St at es

ar

ly

m D en

Ita

n ai

l ga

Sp

m

rtu Po

an

y

ce

er

iu m

an Fr

lg Be

G

N et

he

rla

nd

s

0

Figure 9.12 Nonemployed sick and disabled workers (source: OECD 2005c)

Policies to Create and Destroy Human Capital in Europe

273

matter (e.g., see Prescott 2004; Alesina, Glaeser, and Sacerdote 2005). The average number of hours worked is falling quite steadily over time in many countries in the last two decades (OECD 2006a). The only exception is Sweden where average hours worked actually increased. Anglo-Saxon countries featured rather stable patterns of hours worked over time with the exception of the United Kingdom. Part of this development is the mirror image of increasing labor force participation rates by female workers who tend to work more in part-time jobs (OECD 2006a). This pattern cannot readily be explained by increasing tax rates everywhere because tax rates have been falling in recent years in many countries (see also figure 9.16). Generous Benefit Entitlements Reduce Employment Replacement incomes when unemployed can be high as indicated in figure 9.13. Replacement incomes for the pool of the unemployed in Nordic and continental European countries are around 50 to 60 percent of earned income. Anglo-Saxon countries have much lower replacement rates in the order of 20 percent of earned income, or even less on average. The Mediterranean countries are in the middle with replacement rates of about 30 percent with a notable exception of Italy. Figure 9.14 shows how eligibility for unemployment benefits changes over time. Individuals quickly loose their benefits in the Anglo-Saxon, Nordic, and Mediterranean countries. Only in the continental European countries unemployment benefits often extend to five years or more without large reductions in benefit levels. Theoretical work by Layard et al. (1991), Bovenberg and van der Ploeg (1994), Pissarides (1998), Sørensen (1999), and others, shows that larger replacement benefits reduce employment in both competitive and noncompetitive labor markets characterized by unions, efficiency wages, or matching frictions. In a neoclassical world, unemployment benefits lower labor supply through income effects and act as subsidies on leisure. In labor markets with unions, higher replacement rates raise unions’ wage demands and this lowers employment. In labor markets with frictions or efficiency wages, higher replacement incomes increase reservation wages for workers and thereby lower employment as workers receive higher wages. So both in competitive and noncompetitive labor markets, higher (unemployment) benefits reduce employment (or increase unemployment). Generous benefit entitlements are probably one of main the reasons why unemployment rates are high, but also contributing are extensive

274

James J. Heckman and Bas Jacobs

Percent of earned income

70 60 50 40 30 20 10 0 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 Netherlands

Belgium

France

Germany

Percent of earned income

70 60 50 40 30 20 10 0 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 Denmark

Norway

Sweden

Finland

Figure 9.13 Replacement incomes of unemployed workers, 1961 to 2001 (source: OECD 2004b)

duration of benefits, strict labor market regulations with respect to hiring and firing of workers, and large union coverage (see Layard et al. 1991; Nickell 1997). The main problem, however, with these macroeconomic studies is that the time-series variation within countries is rather limited and identification of effects heavily relies on the crosscountry dimension. Adding country fixed effects often destroys the cross-country correlations found (see also Van Ours and Belot 2001; Blanchard 2006). In contrast to the macroeconomic literature, a pile of microeconomic studies suggests that employment sharply decreases with the generosity and duration of benefits because workers search less actively for work both in the United States and in European countries (see the overview in Lalive et al. 2006) and the estimates they present. Abbring

Policies to Create and Destroy Human Capital in Europe

275

Percent of earned income

70 60 50 40 30 20 10 0 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 Portugal

Spain

Italy

Greece

Percent of earned income

70 60 50 40 30 20 10 0 1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 United Kingdom

Australia

Canada

United States

Figure 9.13 Continued

et al. (2005) and Lalive et al. (2005), and the numerous papers they cite, find that sanctions on benefit levels and durations may be highly effective in getting unemployed workers back to work. Some recent studies document strong cohort effects in the take-up rates of benefits (Lindbeck and Nyberg 2006; Ljunge 2006). Younger generations are more likely to collect some benefit than older generations which can be due to the erosion of work ethic.9 These arguments provide an alternative explanation for the high level on which European unemployment rates have been stuck since the end of the 1980s. Labor Market Protection Harms Labor Market Performance Figure 9.15 gives the OECD summary statistic on labor market regulations. This statistic summarizes the severity of legal restrictions on hiring and

276

James J. Heckman and Bas Jacobs

100 80 60 40 20 0

0

5

10

15

AUS

20 AUT

25 BEL

30

35

CAN

40

45

CZE

50 DNK

55

60

FIN

100 80 60 40 20 0

0

5

10

15

FRA

20 DEU

25 GRC

30

35

HUN

40

45

ISL

50 IRL

55

60

ITA

100 80 60 40 20 0

0

5

10

15

JPN

20 KOR

25 LUX

30

35

NLD

40

45

NZL

50 NOR

55

60

POL

100 80 60 40 20 0

0

5

10

15

20

25

30

35

40

45

50

55

60

Time (months) PRT

SVK

ESP

SWE

CHE

GBR

USA

Figure 9.14 Net replacement rates over a five-year period in 2002, in percent (source: OECD 2004b). No entitlement to social assistance, one-earner married couple with two children. Month one refers to the first month of benefit receipt, that is, following any waiting period. Previous in-work earnings are equal to the average production wage. Children are aged 4 and 6, and neither child care benefits nor child care costs are considered.

Policies to Create and Destroy Human Capital in Europe

277

4.5 1990 1998 2003

Labor market regulations indicator

4 3.5 3 2.5 2 1.5 1 0.5

Ita l G y re ec e D en m ar k N or w a Sw y ed en U Fi ni n la te nd d Ki ng do m Au st ra lia C an U ni te ada d St at es

m

an Fr

iu lg Be

ce G er m an Po y rtu ga l Sp ai n

N et

he

rla

nd

s

0

Figure 9.15 OECD Indicator-I Labor Market Regulations (source: OECD 2006a). The indicator measures the stringency of national legislation on employment protection (EPL), which is a cardinal summary of a set of indicators converted to cardinal scores and normalized to range from 0 to 6, with higher scores representing stricter regulation. Included are indicators for individual dismissals for regular employment, temporary employment, collective dismissals, regular procedural inconveniences, notice and severance pay for dismissals by tenure categories, difficulty of dismissal for regular contracts, legislation on fixed-term contracts, temporary work agency contracts, the definition of collective dismissal, additional notification requirements, additional delays involved, and other special costs to employers.

firing, flexibility in labor contracts, working time restrictions, minimum wages, and employees’ representation rights (work councils, company boards). Mediterranean countries have the least flexible labor markets, followed by the continental European countries. Nordic countries, and especially Denmark, appear to have more flexible labor markets. It goes without saying that the Anglo-Saxon countries have most flexible labor markets in the world. Measures of labor market protection appear to be associated with bad labor market performance in macroeconomic studies (see Layard et al. 1991; Nickell 1997). Again, time-series variation in these crosscountry panel analyses is often too limited and solid conclusions cannot be drawn in general. Heckman and Pagés (2003, 2004) present evidence from microeconomic studies for Latin American countries to assess the impact of labor market regulations. They find that job security regulations indeed have large efficiency costs. In addition the distributional

278

James J. Heckman and Bas Jacobs

consequences appear to be perverse. Insiders gain from labor market regulations at the expense of outsiders: young and unskilled workers. However, using a panel of both Latin American and OECD countries, and exploiting exogenous variation induced by various policy reforms, macroeconomic evidence on the impact of labor market regulations is found to remain fragile. Payroll taxes are the only really robust variable in explaining lower employment and higher unemployment rates. Large Tax Burdens Weaken Labor Market Performance Figure 9.16 gives the average and marginal tax burdens on earned income including the value-added or sales taxes. Large average and marginal tax burdens suggest that labor supply is distorted substantially (Prescott 2004; Alesina et al. 2005). Marginal tax rates are generally in the order of 50 to 60 percent in continental European and Nordic countries. Marginal tax rates are substantially lower in the Mediterranean countries and the Anglo-Saxon world. Here the continental European countries have the steepest graduation in tax rates. The other countries are relatively close in terms of tax rate progression. A huge micro literature shows that high levels of taxation depress labor supply in terms of hours worked (for extensive overviews, see Pencavel 1986; Killingsworth and Heckman 1986; Blundell and MaCurdy 1999). Substitution effects in labor supply are dominant (especially for women), and income effects appear to be modest. Much less empirical evidence can be found on the effects of taxation on other choice margins than hours worked. As stressed by Heckman (1993) and Saez (2002), the participation (“extensive”) margin is more elastic than the hours worked (“intensive”) margin. Taxation potentially also affects human capital formation, search efforts of the unemployed, and wage setting institutions. The effects of higher taxes on labor earnings are not clear-cut in noncompetitive, European style labor markets. Typically in noncompetitive labor markets the wage level and the unemployment rate increase if marginal tax rates are lower, average tax rates are higher, and benefit levels/replacement incomes are higher—ceteris paribus. Higher marginal taxes generally increase employment for given average taxes and labor supply (see Bovenberg and van der Ploeg 1994; Sørensen 1999; Pissarides 1998; Bovenberg 2006; Van der Ploeg 2006). Unions are punished when seeking higher wages in response to higher marginal tax rates as the government taxes away wage increases at higher rates. This force moderates wage demands, labor demand

0.8 2000 2003

0.6

0.4

0.2

es

a

ia

at

ad

St te

d

C

Au

st

an

m

ra l

d

do ng

ni

U

ni

U

te

N

d

Ki

Fi

nl

an

en

ay

ed Sw

ar k

N

or w

ce

m en D

G

n

l

ly

re e

Ita

ai Sp

ga

y

G

Po

rtu

an

ce

er m

m

Fr an

iu lg

et

he

Be

rla

nd

s

0

0.6 2000 2003

0.4

0.2

es

a

at

te

d

C

St

an

st Au

ad

lia

m

ra

do ng

ni

U

ni

U

te

N

d

Ki

Fi

nl

an

d

en

ay Sw

or N

ed

w

ar

k

e

m

ec

en D

n

l

ly

re G

Ita

ai Sp

rtu

ga

y an m

er G

Po

ce

m

an Fr

iu lg

et

he

Be

rla

nd

s

0

1.1 2000 2003

1 0.9 0.8 0.7 0.6

at e

s

da U

ni

te d

St

lia

an a C

ra

Au st

om

an d nl U

ni

ng d

Ki

te d

Fi

ed en

ay

Sw

ar k

or w

N

D

en m

ly re ec e G

Ita

an y Po rtu ga l Sp ai n

er m

G

m

an ce Fr

Be lg iu

N

et h

er

la nd

s

0.5

Figure 9.16 Marginal tax, average tax wedges, and income progression (source: OECD 2005e). (Top) Total marginal tax wedge on labor income, including consumption taxes (percent of income). (Middle) Total marginal tax wedge on labor income, including consumption taxes (percent of income). (Bottom) Musgrave and Musgrave’s coefficient of residual income progression. All tax rates apply to a single worker earning the average production wage without children. For the United States no consumption tax data were available, so we assumed a US consumption tax of 5 percent. The effective tax rates on income are computed as 1 − (1 − income tax)/(1 + consumption tax), since the consumption tax only applies to the spending of after-tax labor income.

280

James J. Heckman and Bas Jacobs

expands and unemployment falls. In labor markets characterized by search frictions, marginal tax rates can boost employment. Since workers and firms Nash-bargain over the surplus of firm-worker matches, workers capture less of the surplus when marginal tax rates increase. Firms need to pay a higher gross wage to provide a given net wage to the worker. And, conversely, workers need to accept a lower net wage to maintain a given level of profits for the firm. Hence the negotiated wage falls, labor demand expands, labor market tightness increases, and equilibrium unemployment falls. In market environments characterized by efficiency wages, firms find it harder to recruit, retain, or motivate workers by increasing wages when governments tax away these wage increases at higher rates. Therefore firms pay lower wages, labor demand expands, and equilibrium unemployment falls. Higher average income taxes—for given marginal tax rates—may also have opposite effects in comparison with neoclassical models, depending on the response of the net replacement rate to higher taxes (see Bovenberg and van der Ploeg 1994; Sørensen 1999; Pissarides 1998; Bovenberg 2006; Van der Ploeg 2006). A higher average income tax increases net replacement rates (net benefit divided by the net wage rate) if benefits are untaxed. In response to higher average taxes, unions will demand higher wages as the position of their working members worsens in comparison with the nonworking members so that unemployment rates go up. Firms paying efficiency wages see that it becomes more difficult to recruit, retain, or motivate workers because net replacement rates increase. As a result, labor costs rise and equilibrium unemployment increases. With search frictions, higher average tax rates on wage income increase wage demands of workers, which push up wage costs for firms, so labor market tightness falls and unemployment increases. However, when benefits are indexed to net wages, replacement rates remain fixed; unions, firms, or workers do not change wage setting behavior (a lot) and there are much smaller (or even zero) effects of higher average tax rates on unemployment. Most of the theoretical papers cited above analyze noncompetitive labor markets where labor supply is treated as exogenous, thereby overstating the case for tax progression. Sørensen (1999: 449) demonstrates that the welfare gains of tax progression in union and efficiency wage models are virtually negligible when the compensated wage elasticity of labor supply approaches lower bounds found in the empirical literature (around 0.20/0.25); see Browning, Hansen, and Heckman (1999) and Blundell and MaCurdy (1999) for evidence on this question.

Policies to Create and Destroy Human Capital in Europe

281

Indeed Sørensen’s findings suggest that deviations from a flat rate income tax to cushion labor market distortions are quite costly in terms of reduced labor supply. Sørensen (1997) develops a CGE model for Denmark. He analyzes larger tax credits for low-wage earners, financed by raising marginal tax rates on all labor incomes. This policy, which makes the tax system more progressive, has ambiguous welfare effects, depending critically on elasticities of the wage equation with respect to marginal and average tax rates on wages. Bovenberg et al. (2000) develop a detailed CGE model for the Netherlands, which incorporates labor supply on the intensive and extensive margins, on-the-job training, search frictions, and wage setting by unions. They demonstrate that the negative (positive) effects of high marginal (average) tax burdens on labor supply and training are outweighed by the small positive effects arising from wage moderation and reductions in labor market frictions. The possibility of setting wages above the market-clearing level depends on the ability of consumers or workers to shift taxes to firms. Most theories of wage determination in noncompetitive labor markets show that the wage markup over the market-clearing wage declines if the labor demand elasticity increases, so that it is more difficult to shift taxes toward firms (e.g., see Bovenberg 2006; Van der Ploeg 2006). In small open economies with perfect capital mobility and perfectly competitive goods markets, it is not possible to shift the tax burden to firms; labor demand would be perfectly elastic at the world wage rate. Thus the distortions created by unions, search frictions, and efficiency wages could be less relevant for many small open European countries. Finally, to fully understand the effects of taxes in distorted labor markets, one should not only focus on changes in tax rate progression for given levels of benefits and average tax rates, as most of the aforementioned papers do. Empirically, larger marginal tax rates are strongly associated with higher nonemployment benefits and higher average tax burdens; see figures 9.13 and 9.16. A policy of increasing marginal taxes, while using tax revenues to finance benefits or larger levels of public spending (so that average tax rates increase) generally reduces labor supply and employment in both competitive and noncompetitive labor markets. To conclude, it is not clear that taxation is always less harmful in noncompetitive than in competitive labor markets. This depends on how responsive labor supply is, how much of the incidence of taxes can be shifted to firms, and how tax revenues are spent.

282

9.2.4

James J. Heckman and Bas Jacobs

Skill Maintenance

Decreasing Retirement Ages Causes Quicker Depreciation of Skills Apart from labor force participation and hours worked, the age of retirement also constitutes an important element of the utilization of human capital over the life cycle. At retirement, human capital is written off completely. If workers retire later, they will have larger returns on their investments in education and training on the job as the time horizon over which the investments mature expands. Figure 9.17 shows that that labor force attachment of the average worker is rapidly declining with age. This development is also carefully documented by Gruber and Wise (1999). Labor force participation rates of 55- to 64-year-old workers are only in the order of half or even less. Especially the continental European and Mediterranean countries have low participation rates of older workers. Nordic countries outperform the Anglo-Saxon countries as regards the labor force participation rates of 55- to 59-year-old workers, but the Anglo-Saxon countries do better in the 60- to 64-year cohorts. Figure 9.18 shows the development of labor force participation rates of cohorts of workers aged 55 to 59 years. Generally, the labor force 100 90

Participation rate (%)

80 70 60 50 40 30 20 10 0 15–19

20–24

25–29

30–34

35–39

40–44

45–49

50–54

55–59

60–64

Age group Netherlands Spain Sweden

Belgium Italy United Kingdom

France Greece Australia

Germany Denmark Canada

Figure 9.17 Labor force participation rates by age, 2002 (source: OECD 2006a)

Portugal Norway United States

1960 1970 1980 1990 2000

United States

Norway

France

Australia

Participation rate men

1960 1970 1980 1990 2000

West Germany

Portugal

Germany

Belgium

Participation rate all

1960 1970 1980 1990 2000

Spain

Greece

Canada

Figure 9.18 Labor force participation of 55- to 59-year-old cohorts (source: OECD 2006a)

1 0.8 0.6 0.4 0.2

1 0.8 0.6 0.4 0.2

1 0.8 0.6 0.4 0.2

1 0.8 0.6 0.4 0.2

Participation rate women

1960 1970 1980 1990 2000

Sweden

Italy

Denmark

1960 1970 1980 1990 2000

United Kingdom

Netherlands

Finland

Policies to Create and Destroy Human Capital in Europe 283

Age

284

James J. Heckman and Bas Jacobs

72 70 68 66 64 62 60 58 56 1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002

Age

Netherlands

Belgium

France

Germany

72 70 68 66 64 62 60 58 56 1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 Denmark

Norway

Sweden

Finland

Figure 9.19 Average effective retirement age of men, 1960 to 2001 (source: OECD 2005c)

participation rates have been falling and show a turn around in recent decades. However, if there has been an increase in overall labor force participation rates of 55- to 59-year-old cohorts, it is mainly driven by the general increases in female force participation rates. Belgium (slightly), Denmark, Finland, and the Netherlands are the only four countries who have witnessed both an increase in male and female labor force participation rates and in recent years which is probably due to policy changes in early retirement schemes.10 Figure 9.19 plots the effective age of retirement, conditional upon being in the labor force. There has been a landslide in effective retirement ages. Over the last forty years effective retirement ages went down massively everywhere in the Western world, including the Anglo-Saxon countries. Again, the continental European countries have witnessed the largest decreases in the retirement ages (see also Gruber and Wise 1999). In recent years we see that the decrease in effective retirement ages has come to a halt at a low plateau.

Age

Policies to Create and Destroy Human Capital in Europe

285

72 70 68 66 64 62 60 58 56 1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002

Age

Portugal

Spain

Italy

Greece

72 70 68 66 64 62 60 58 56 1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 United Kingdom

Australia

Canada

United States

Figure 9.19 Continued

Figure 9.20 shows that labor force participation rates of 55- to 59- and 60- to 64-year-old cohorts are much higher when individuals have more initial education. Better skilled workers retire much later. This graph reinforces our notion that labor supply and skill formation over the life cycle are strongly complementary activities. Generous Pensions and Early Retirement Schemes Reduce Labor Force Participation of Older Workers Pension benefits can be generous as is clear from figure 9.21. Pension replacement incomes in continental European are quite high and about 60 to 80 percent of preretirement earnings for an average worker. Mediterranean countries have exceptionally generous pension schemes which entail pension benefits of 80 to 100 percent of last earnings (up to 100 percent in Greece). The Nordics, on the contrary, have much more modest pension benefits in the order of 40 to 60 percent of pre-retirement earnings. The AngloSaxon countries have on average the lowest pension benefits which are around 40 to 50 percent of final earnings. Another interesting feature is

286

James J. Heckman and Bas Jacobs

100 90 80 70

Percent

60 50 40 30 20 10 at St d

te

C an

ad

a

es

lia

m

ra

Au

st

d

do

U ni

te

U ni

d

Ki

ng

nl

an

en

Fi

Sw ed

k ar

D en

N or

m

w ay

e

ly

ec

Ita

G re

ai Sp

Po

rtu

ga

l

n

ce

m Fr

an

iu lg

Be

N et

he

rla

nd

s

0

90 80 70

Percent

60 50 40 30 20 10 es

a

St

U

ni

te

d

C

an

at

ad

lia

m

ra st

Au

ng

do

d U

ni

te

N

d

Ki

Fi

nl

an

en

ay w

ed

k ar

or

Sw

D

N

m en

re

ec

e

ly G

Ita

n ai Sp

ga

l

ce

rtu Po

m

an Fr

iu lg

Be

et

he

rla

nd

s

0

Less than upper secondary level

Upper and post-secondary level

Tertiary nonuniversity level

Levels that correspond to both ISCED 5A and 6

Figure 9.20 Labor force participation of older workers by level of education in 2002 (source: OECD 2006a). (Top) Cohorts aged 55 to 59. (Bottom) Cohorts aged 60 to 64.

that pension systems are PAYG state pensions almost everywhere. Exceptions are the Anglo-Saxon countries, the Netherlands, Sweden, and Denmark that also rely on substantial private funding, either through DB/DC occupational pensions or individual saving schemes; see also OECD (2005d). Note finally that net pension incomes are always larger than gross pension incomes. The reason is that all governments give tax deductions or subsidies on pension savings. Many workers retire long before statutory retirement ages via all kinds of early retirement schemes. It is not easy to make international

Policies to Create and Destroy Human Capital in Europe

287

120 Gross Net

Percent of final earnings

100 80 60 40 20

s

m iu

nd

lg

rla

Be

he N et

Fr an c G er e m an Po y rtu ga l Sp ai n Ita ly G re ec D en e m ar k N or w ay Sw ed en U F ni te inla d Ki nd ng do m Au st ra lia C an U ad ni te a d St at es

0

Figure 9.21 Gross and net pension replacement incomes for workers with average earnings (percent of earnings) (source: OECD 2005d).

comparisons because the institutional details vary from country to country. However, we can summarize the impact of early retirement schemes on the labor market by the implicit marginal tax rates imposed on an additional year of work (see also Gruber and Wise 1999). Figure 9.22 shows that early retirement schemes do indeed cause very high marginal tax rates on pre-retirement incomes.11 Moreover retirement ages and benefit generosity are very negatively related. Gruber and Wise (1999) present strong evidence that this is a causal relation. It should therefore not come as a surprise that the continental European and Mediterranean countries have low labor force participation rates of elderly workers as they have the most generous early retirement schemes. In recent years some countries have reformed their pension schemes. The Netherlands, Germany, France, and Italy are examples. Labor force participation of older workers appears to be picking up in recent years in some countries due to reforms, general trends such as rising female labor force participation, or because workers anticipate future reforms. 9.3

The Technology of Skill Formation12

Figure 9.23 summarizes the major theme of Heckman (2000) and Carneiro and Heckman (2003). It plots the rate of return to human capital

288

James J. Heckman and Bas Jacobs

0

ISL

–5

JAP

NZL

–10

NOR

SWE

USA AUS CAN

–15

GBR CHE

IRL

DEU PRT

–20 –25

FRA

ESP FIN

NLD

AUT

–30 ITA

–35

BEL LUX

–40 0

10

20

30

40

50

60

70

80

90

100

Implicit tax on continued work, 55–59 (%) 0 –10 –20 –30 –40 –50 –60 –70 –80

ISL CHE

KOR USA AUS

NOR NZL CAN SWE

JAP

PRT

IRL

GBR ITA

ESP FIN

DEU BEL AUT

0

10

20

30

40

50

60

70

LUX FRA

80

90

NLD

100

110

120

Implicit tax on continued work, 60–64 (%) Figure 9.22 Fall in male labor participation between two consecutive age groups and implicit tax on continued work, 1999 (source: OECD 2004c). The implicit tax rates are calculated for a single worker with average earnings in 1999. In some cases the results differ from those presented in figure 9.4, which refer to currently legislated systems. These differences reflect either policy changes that took place after 1999 (e.g., Finland, France) or the future implementation of measures that were already legislated but had not yet come into effect in 1999 (e.g., the future maturation of the Superannuation Guarantee Scheme in Australia, the transition from the “old” to the “new” pension system in Italy). (Top) Percentage change in labor force participation between men aged 50 to 54 and aged 55 to 59, calculated as (Pr55–59 − Pr50–54)/Pr50–54 per cent. (Bottom) Percentage change in labor force participation between men aged 54 to 59 and aged 60 to 64, calculated as (PR60–64 − Pr55–59)/Pr55–59 per cent.

at different stages of the life cycle for a person of given abilities. The horizontal axis represents age, which is a surrogate for the agent’s position in the life cycle. The vertical axis represents the rate of return to investment assuming the same amount of investment is made at each age. Ceteris paribus, the rate of return to a dollar of investment made while a person is young is higher than the rate of return to the same dollar made at a later age. Early investments are harvested over a longer horizon than those made later in the life cycle (Becker 1964). In addition,

Policies to Create and Destroy Human Capital in Europe

Rate of return to investment in human capital

289

Preschool programs

Schooling r Job training

Preschool

0

School

Post school

Age Rates of return to human capital investment initially setting investment to be equal across all ages

Figure 9.23 Rates of return to human capital investment, when initially setting investment to be equal across all ages (source: Carneiro and Heckman 2003)

because early investments raise the productivity (lower the costs) of later investments, human capital is synergistic. Learning begets learning; skills (both cognitive and noncognitive) acquired early on facilitate later learning. Early deficits make later remediation difficult. Finally, young children’s cognition and behavior are more malleable than cognition and behavior in adults. For an externally specified opportunity costs of funds r, an optimal investment strategy is to invest relatively less in the old and relatively more in the young. A central empirical conclusion of their analysis is that at current investment levels, efficiency in public spending would be enhanced if human capital investment were directed more toward the disadvantaged young who do not receive enriched early environments, and less toward older, less skilled, and illiterate persons for whom human capital is a poor investment. Abilities are multiple in nature. They are both cognitive and noncognitive. Both cognitive and noncognitive abilities matter in determining participation in crime, teenage pregnancy, drug use, and other deviant activities. These abilities are themselves produced by the family and by personal actions. Both genes and environments are involved in producing these abilities. Environments affect genetic expression

290

James J. Heckman and Bas Jacobs

mechanisms (e.g., see Turkheimer et al. 2003; Björklund, Lindahl, and Plug 2006). This interaction has important theoretical and empirical implications for skill policies. It suggests an important role for environment-enriching policies in fostering human skills. Differences in cognitive ability across family types appear early and persist over time (Carneiro and Heckman 2003; Cunha and Heckman 2007). Measured cognitive ability is susceptible to environmental influences, including in utero experiences. Education barely affects test score gaps by family income or socioeconomic status after the early years of schooling (Carneiro and Heckman 2003; Cunha, Heckman, Lochner, and Masterov 2006; Raudenbush 2006).13 Noncognitive abilities such as motivation, self-discipline, and time preference—associated with the development of the prefrontal cortex— are also subject to environmental influences. Noncognitive abilities and cognitive abilities affect schooling attainment and performance, and a wide array of behaviors. Using a novel empirical approach, Heckman, Stixrud, and Urzua (2006) identify a low-dimensional vector of latent cognitive and noncognitive skills that explains a diverse array of social and labor market outcomes. For many dimensions of social performance, cognitive and noncognitive skills are equally important. Heckman and Rubinstein (2001), Heckman, Hsee, and Rubinstein (2001), and Heckman and LaFontaine (2006) study the GED program14 and show that the cognitive ability of GED participants is on average equal to that of high school graduates who do not enroll in college and even higher than the ability of high school dropouts. However, GED recipients earn less than high school dropouts once the analyst controls for cognitive abilities. Consequently noncognitive ability appears to be an important determinant of earnings that GED recipients lack. As is true for cognitive skills, gaps in noncognitive skills (motivation, trustworthiness, and behavioral skills) emerge early and are substantially reduced once long-run family factors influencing the child’s early years are controlled for (Carneiro and Heckman 2003). IQ is fairly well set by age ten. Noncognitive abilities are more malleable over the life cycle than cognitive abilities. Much of the effectiveness of early childhood interventions comes in boosting noncognitive skills and fostering motivation (e.g., see Heckman, Malofeeva, Pinto, and Savelyev 2010). Short-term increases in cognitive skills (test scores) fade out over time. Successful programs increase noncognitive skills and result in more social behavior and less

Policies to Create and Destroy Human Capital in Europe

291

crime. Programs are more successful if parents are part of the treatment, which bolsters the notion that improvements in the home environment have long-lasting effects. For overviews of the literature and evidence on a diverse array of early intervention programs, see Heckman (2000), Cunha, Heckman, Lochner, and Masterov (2006), Carneiro and Heckman (2003), and Cunha, Heckman, Lochner, and Masterov (2006). Given the quantitative importance of noncognitive traits, social policy should be more active in attempting to alter them especially for children from disadvantaged environments who receive little encouragement and discipline at home. Interventions in adolescent years partially remediate but do not remedy insufficient early childhood investments at current levels of investment. Just as early intervention programs have a high payoff primarily because of the social skills and motivation they impart to the child and the improved home environment they produce, so do interventions that operate during the adolescent years, and for many of the same reasons. The impacts they achieve are modest, but positive. One cannot expect substantial benefits from public job-training programs that are primarily targeted to disadvantaged workers. Surveying mainly microeconomic studies, Heckman, Lalonde, and Smith (1999); Martin and Grubb (2001), and Calmfors, Forslund, and Hemström (2001) conclude that these programs are largely ineffective.15 The comparison of job-training programs suggests a few important lessons. First, you get what you pay for. The recently terminated JTPA program in the United States cost very little but produced very few results. An exception to the rule is classroom training, for which the returns are substantial (Heckman, Hohmann, Khoo, and Smith 2000). Second, the effects of treatment vary substantially among subgroups (Heckman, LaLonde, and Smith 1999). Third, job-training programs also have effects on behavior beyond schooling and work that should be considered in evaluating their full effects. Reductions in crime may be an important effect of programs targeted at male youth. The evidence summarized in Heckman, LaLonde, and Smith (1999) indicates that the rate of return to most US and European training programs is far below 10 percent, although the benefits to certain groups may be substantial. Some programs survive a cost–benefit test, but many do not. And even the most successful programs have only small impacts on poverty rates and few are lifted out of it. The study by Calmfors, Forslund, and

292

James J. Heckman and Bas Jacobs

Hemstöm (2001) presents an extensive overview of the Swedish experience with active labor market policies, and they conclude that ALMP have been inefficient. Europe’s skill policy should not look to public job training to remedy or alleviate substantially skill deficits that arise at early ages. The wisest long-term investment policy is to invest in the young. Returns are highest for investments in children from disadvantaged families where children receive inadequate parental resources (Heckman 2006; Heckman and Masterov 2006). Universal programs generate dead weight because they misallocate resources to children from advantaged families that have lower returns due to higher parental investment. Politicians face a practical problem of the transition. Older persons and disadvantaged younger persons are unemployable at current wage minimums. Investing in them has a low economic return. A better policy is to subsidize their employment to give them dignity and social inclusion, and to benefit from what they can offer society at large. The essays in Phelps et al. (2003) argue strongly for carefully constructed wage subsidies. Such subsidies should be cohort specific and phased out over time. Otherwise, newer generations will have weak incentives to develop skills and the problem of poverty will perpetuate across generations. 9.4

A Theory of Skill Formation, Skill Use, and Skill Maintenance

How can we reconcile the empirical findings of the European experience with theory? In this section we develop a partial equilibrium lifecycle model of schooling, on-the-job training, labor supply, saving, and retirement. By simultaneously analyzing schooling, training labor supply, and retirement decisions, the model allows us to spell out various complementarities over the life cycle. First, we show that human capital investments feature dynamic complementarities over the life cycle even after initial education. Both initial schooling and later on-the-job training are complementary activities. The returns to initial schooling are larger when individuals engage more in on-the-job training later on during their working careers. Also individuals will invest more in on-the-job training when they have more initial schooling. Second, complementarities exist between skill formation and labor market participation in its broadest sense. That is, the more individuals work and the later they retire, the larger will be the returns to investments in initial schooling and on-the-job training. The reason is that

Policies to Create and Destroy Human Capital in Europe

293

the costs of leisure and retirement increase when individuals become better skilled. The reverse also holds: later retirement and more hours of work boost skill formation by increasing its financial rewards. Our model builds on Mincer (1974) and Ben-Porath (1967) and adds an endogenous retirement decision. We focus on labor supply on the intensive (hours) margin and retirement.16 We maintain the assumption of full employment as labor markets are perfectly competitive and frictionless. We acknowledge at the outset that this is probably not the best description of the labor markets in Europe, but economic theory does not provide us yet with useful models that allow for the joint determination of labor supply, human capital formation, and wages in noncompetitive labor markets.17 9.4.1 Model We assume that a representative individual is born at time t = 0 and has a life span T that is exogenously given. This individual undergoes S years of initial education at the beginning of his life. Then the individual works. After the working career the individual retires at date R. The time constraint states that total time in school S, in the labor market (R − S), and in retirement (T − R) should equal the life span T of the individual: T = S + (R − S) + (T − R).

(9.1)

At each date the individual derive instantaneous utility U(Ct ) from consumption Ct . To simplify the analysis, we assume that only when the individual is in the labor market (S < t ≤ R) does he derive utility V(Lt) from leisure Lt. Similarly the individual does not engage in training on the job before entering the labor market and stops with the job training when he leaves the labor market. The time constraint while working states that the fraction of time working Lt, plus the fraction of time invested in training It plus the fraction of time consumed as leisure Lt should be equal to the total time endowment—which is normalized to one: 1 = Lt + It + Lt,

S < t ≤ R.

(9.2)

Alternatively, one could interpret Lt as the labor force participation rate, It as aggregate training efforts, and Lt as the nonemployment rate in this representative agent setting. Individuals derive utility X(T − R) from the years they are retired T − R where R denotes the retirement age. Retirement is a discrete deci-

294

James J. Heckman and Bas Jacobs

sion to exit the labor market completely. The individual does not derive direct (dis)utility from being in school. Life-time utility of the individual is a time-separable function of instantaneous consumption and leisure felicities and retirement utility

∫

T

0

R

U (Ct ) exp ( − ρt ) dt + ∫ V (Lt ) exp ( − ρt ) dt + X (T − R), S

(9.3)

with U′(Ct) > 0, U″(Ct) < 0, V′(Lt) > 0, V″(Lt) < 0, X′(T − R) > 0, and X″(T − R) < 0, where r is the subjective rate of time preference. These preferences simplify our analysis. The costs of forgone labor time are measured by forgone labor earnings. The value of retirement leisure is governed by X(T − R).18 The representative individual optimally decides the number of years S in education. W(S) is the rental rate of human capital of type S. This rental rate is assumed to be constant over time, and it differs between individuals with different skill levels. W(S) features positive but diminishing marginal returns of additional initial schooling: W′(S) > 0, W″(S) < 0. Alternatively, one may interpret W(S) as the production function of human capital. The costs of education are the forgone earnings W(S) while not working and the direct costs P per year of education (we ignore the utility or disutility of education). Without loss of generality, we keep the direct costs of education fixed. We assume here that the government can affect decision on the optimal years of schooling only through the tax system and education subsidies. In the real world, governments affect the education choices of individuals through a host of other mechanisms, for example, by outlawing child labor, by setting minimum school leaving ages, and so on. These alternative instruments can be regarded as implicit rather than explicit subsidies on education. As such, our model is still suited to capture the main incentive issues, and we do not think that this undermines our main story of the various complementarities over the life cycle. Nevertheless, some of these policies may require fewer public resources than education subsidies and could be preferred for that reason. The individual starts his life with A0 in financial assets which are normalized to zero for convenience (A0 = 0). He borrows in a perfect capital market at constant real interest rate r to finance the costs of living and the costs of education in the periods when he is enrolled in initial education. The flow budget constraint of the individual while still in school (t ≤ S) is therefore given by

Policies to Create and Destroy Human Capital in Europe

. At = (1 − tA)rAt − Ct − (1 − s)P,

0 ≤ t ≤ S,

295

(9.4)

where a dot denotes a time derivative. Since A0 = 0, and Ct and P are both positive, the individual accumulates debt in the first periods of his life. tA is the tax on interest income. Interest payments of education loans are deductible for the interest tax; s is the subsidy rate on direct educational costs. After graduation the individual starts earning gross labor income W(S)HtLt. Ht is the stock of human capital, which is gathered through training on the job in a manner that is made precise below. The flow budget constraints after graduation until retirement (S < t ≤ R) state that the increase in financial assets should equal total interest income (which is negative when individuals repay debts), net labor income (1 − tL)W(S)HtLt (where tL is the labor income tax rate), minus consumption Ct: . At = (1 − tA)rAt + (1 − tL)W(S)HtLt − Ct, S < t ≤ R. (9.5) After retirement, until death (R < t ≤ T), the representative individual does not work anymore and runs down his accumulated assets for consumption purposes: . At = (1 − tA)rAt + B − Ct, R < t ≤ T, (9.6) where B is the net retirement benefit in each year spent in retirement. One should interpret the pension benefit B as that part of pension benefits that is actuarially completely nonneutral, since individuals only receive retirement benefits conditional upon full retirement. Any actuarially fair pension savings are covered by the voluntary saving decision. The individual has no bequest motive and ends his life with zero wealth: AT = 0. The representative individual can increase his human capital by allocating time It to learning activities, while forgoing labor earnings or leisure time. It’s assumed that on-the-job training does not require direct costs. The individual has one unit of on-the-job human capital when he enters the labor market (HS ≡ 1). On-the-job human capital accumulates according to a Ben-Porath (1967) type of production function . S < t ≤ R, (9.7) Ht = G(S)F(It, Ht) − dHt, where FI(It, Ht) > 0, FH(It, Ht) > 0, FII(It, Ht) < 0, FHH(It, Ht) < 0, and FIH(It, Ht) > 0. G(S) denotes the productivity of on-the-job training, which increases

296

James J. Heckman and Bas Jacobs

with the initial level of education at a diminishing rate: G′(S) > 0 and G″(S) < 0. This captures the main idea of dynamic complementarity in skill formation. Larger levels of initial education increase the productivity of investments in on-the-job training. Furthermore there is dynamic complementarity in human capital formation on-thejob because the marginal product of training investments G(S)FI(It, Ht) increases with the level of human capital Ht as indicated by the positive cross-derivative FIH(It, Ht) > 0. Larger levels of human capital increase the productivity of later human capital investments. d denotes the rate of depreciation of human capital. Browning, Hansen, and Heckman (1999) survey empirical estimates of Ben-Porath earnings functions. Specification (9.7) is consistent with estimates reported by Heckman, Lochner, and Taber (1998).19 Integrating the asset accumulation constraints and imposing the initial and terminal conditions on financial wealth gives the life-time budget constraint of the individual

∫

T

0

S

Ct exp ( − r *t ) dt + ∫ (1 − s) P exp ( − r *t ) dt 0

R

T

S

R

= ∫ (1 − τ L ) W (S) Ht Lt exp ( − r *t ) dt + ∫ B exp ( − r *t ) dt ,

(9.8)

where r* ≡ (1 − tA)r is the net discount rate. The individual maximizes life-time utility by choosing consumption, labor supply, leisure, on-the-job training, education, and retirement subject to the household budget constraint, the time constraints, and the accumulation equation for on-the-job human capital. The appendix contains the derivation.20 Using standard results, we obtain the Euler equation for consumption C t = θt (r* − ρ) , 0 ≤ t ≤ T , Ct

(9.9)

where qt ≡ [−U″(Ct)Ct/U′(Ct)]−1 is the intertemporal elasticity of substitution in consumption. If the rate of time preference is lower than the real after-tax return on financial saving, consumption features an upward-sloping profile over the life cycle. A larger intertemporal elasticity of substitution results in a stronger upward-sloping consumption profile and a stronger sensitivity of savings with respect to after-tax returns.21

Policies to Create and Destroy Human Capital in Europe

297

The labor supply equation is governed by the first order condition: V ′ (Lt ) = (1 − τ L ) W (S) Ht , S < t ≤ R. U ′ (Ct )

(9.10)

The marginal willingness to demand leisure time decreases with the net wage rate and increases with the level of taxation. The gross wage rate increases with education S and on-the-job human capital Ht. Hence better skilled workers supply more labor if the substitution effect dominates the income effect in labor supply (which is the empirically plausible case; see Browning, Hansen, and Heckman 1999). Therefore this equation implies that labor supply and skill formation are complementary activities. Indeed the data previously discussed show that more educated workers have higher participation rates and lower unemployment rates. The optimal number of years in initial education follows from the first-order conditions for education, leisure demand, labor supply and training22

∫

R

S

(1 − τ L ) W ′ (S) Ht Lt exp ( − r * (t − S)) dt +

G ′ (S ) F ( IS , 1) (1 − τ L ) W (S) G (S) FI ( IS , 1)

LS ⎞ ⎛ = (1 − s) P + (1 − τ L ) W (S) ⎜ LS + ⎟ , ⎝ εS ⎠ (9.11) where et ≡ V′(Lt)Lt[V(Lt)]−1 > 0 is the elasticity of the leisure subutility function at time t. This is a modified Mincer equation stating that the net present value of marginal returns to initial education (evaluated at the time of graduation S) should be equal to net marginal costs on an additional year of schooling. The latter comprise direct, subsidized expenditures, and net forgone labor earnings. Years spent in initial education increase when the returns to human capital investments are larger. This is the case when the working life is longer and individuals retire later (R larger). We note here that the returns at later ages are more heavily discounted. Consequently, a higher retirement age has only small effects when the discount rate is sufficiently large. The feedback between retirement and education may gain in strength due to training on the job. On-the-job training becomes more profitable when individuals retire later (see below), and this in turn enhances initial investment in education. Returns increase when individuals invest more in on-the-job training

298

James J. Heckman and Bas Jacobs

during their working lives (Ht larger) and supply more labor (Lt larger). The standard Mincer equation ignores the interaction with labor supply and training on the job. In addition the time horizon is finite, and direct costs of education are not negligible as also noted by Heckman, Lochner, and Todd (2006).23 Moreover individuals with a higher level of education have a larger return on investments in on-the-job human capital as indicated by [G′(S)/G(S)][F(Is, 1)/FI(Is, 1)](1 − tL)W(S), which denotes the discounted value of larger human capital investments in training due to more initial education. All these results underscore the second important complementarity. Educational investments increase when the utilization of human capital is larger and when skills are better maintained through on-the-job training. Initial schooling is therefore complementary to later retirement, hours worked and on-the-job training. Again, this is in conformity with the data presented earlier. Labor taxation directly reduces investments in initial education as long as the subsidy rate is smaller than the tax rate (tL > s). If the subsidy rate s equals the tax rate on labor tL taxation is neutral with respect to human capital investments because then all costs and benefits of human capital formation are symmetrically affected by the tax and subsidy rates. Capital income taxation (as reflected in a lower r*) boosts initial education. The reason is that higher capital income taxation lowers the net discount rate at which marginal benefits of education are discounted. Alternatively, one can say that higher taxes on capital income induce substitution in the household lifetime asset portfolio from financial toward human assets (see Heckman 1976). Labor taxation nevertheless reduces labor supply and lower the retirement age (shown below); hence labor taxation still discourages investments in initial education by lowering the utilization rate of human capital. Optimal retirement is given by X ′ (T − R ) 1 − LR ⎞ ⎛ = (1 − τ L ) W (S) H R ⎜ (1 − ζ ) LR + ⎟, ⎝ λR εR ⎠

(9.12)

where lt ≡ U′(C0)exp(−r*t) is the marginal value of income at time t, and z ≡ B/[(1 − tL)W(S)HRLR] is the implicit marginal tax rate on additional years of work due to the presence of (early) retirement incomes.24 The marginal willingness to pay for an additional year in retirement should be equal to the marginal costs of an extra year in retirement. The marginal benefit is the marginal rate of substitution between retirement utility and consumption at the date of retirement. The marginal

Policies to Create and Destroy Human Capital in Europe

299

costs are given by the value of net the forgone labor earnings in the last year on the labor market.25 The individual has stronger incentives to retire later if the individual has more initial education S, has accumulated a higher stock of on-thejob human capital HR, and supplies more labor effort LR in the retirement year R. The labor tax directly distorts retirement decisions, because retirement utility is not taxed, whereas continued work is. This direct impact of labor taxation on retirement is not often discussed in the literature on retirement (e.g., Gruber and Wise 1999). Indeed this literature mainly focuses on the implicit marginal tax rate on additional years of work, z, due to the presence of actuarially unfair (early) retirement incomes. The implicit tax on continued work z exacerbates the impact of the labor tax tL on the decision to exit the labor market. The wealth effects in the retirement decision are captured by lR ≡ U′(C0)exp(−r*R). Richer individuals, with a lower marginal utility of income, retire earlier—ceteris paribus. The third complementarity is therefore that retirement is delayed when individuals utilize and maintain their skills better through working life. Hence more skilled workers retire later when the income effect of higher skills are outweighed by the substitution effects of higher skills. Again, this is in conformity with the data. Investment in on-the-job training is governed by the following equation:

μt G (S) FI ( It , Ht ) = (1 − τ L ) W (S ) Ht , λt

(9.13)

where m t/lt is the shadow value—expressed in monetary units—of one unit of human capital at time t. This equation states that the marginal costs of on-the-job human capital investment (right-hand side) should be equal to the discounted value of marginal benefits in terms of higher wages (left-hand side). The benefits and costs of OJT investments increase when schooling levels are higher and when the individual has a higher stock of human capital. The benefits also increase when the shadow value mt/lt of human capital is large. Now we see that higher levels of initial education both increase the opportunity costs of training on the job and the marginal benefits of training on the job. The same holds for a higher stock of on-the-job human capital. We assume that F(It, Ht) ≡ [Φ(It, Ht)]f, where 0 < f < 1 and s ≡ [ΦI(It, Ht)ΦH(It, Ht)]/[ΦIH(It, Ht)ΦH(It, Ht)]. f is a returns-to-scale parameter and s is the elasticity of substitution between I and H in the homothetic constant returns-to-scale subproduction function Φ(It, Ht). We

300

James J. Heckman and Bas Jacobs

then find an arbitrage equation between on-the-job human capital investments and financial saving: t FI ( It , Ht ) Lt ⎛ ω H H + 1− + (1 − φ ) ω H ⎞ ⎝ ⎠ Ht Ht σ  ωH It + (1 − φ ) (1 − ω H )⎞ = r * + δ , +⎛ ⎝ σ ⎠ It

G (S) FH ( It , Ht ) + G (S)

(9.14)

where wH ≡ [ΦH(It, Ht)Ht]/Φ(It, Ht).26 The left-hand side gives the total returns of one extra unit of human capital. The right-hand side gives the required rate of return on investments in OJT; the net returns on financial savings plus the rate of depreciation. The fourth complementarity in skill formation follows from the last equation. First, individuals with more initial education S will engage in more on-the-job training because the productivity of OJT investments is enhanced by higher initial education. This is, again, in conformity with the data; more educated workers engage more in training. Second, if labor supply increases and human capital is more heavily utilized, the marginal returns to investments in on-the-job training increase. Therefore individuals who work more hours or participate more in the labor market have higher returns on training. Third, if individuals maintain their skills though their working career, later investments in human capital become more profitable. s measures the dynamic complementarity of on-the-job human capital investments. If s = 0, it is not possible to remedy neglect of on-the-job training in the early years of the working career. Early and late investments are perfect complements. If s > 0, it is to some extent possible to remedy poor skill maintenance early in working careers. If s = ∞, initial and later investments are perfect substitutes. wH then measures the plasticity of investments in human capital. If wH > 1/2, plasticity is smaller at later ages than at early ages. If wH is close to 1, it may be very costly to remediate deficient early career investment in on-the-job training. When wH = 1, it is impossible.27 The tax rate on labor incomes is absent in the training arbitrage equation. Linear income taxes affect marginal costs and benefits of training equally and therefore do not directly reduce training investment.28 Note, however, that labor supply and retirement are distorted by higher taxes. So taxes do indirectly affect the returns to training by lowering the marginal benefits as the utilization rate of the stock of on-the-job human capital falls and the payback time of investments decreases. Capital income taxes boost training by lowering the required

Policies to Create and Destroy Human Capital in Europe

301

rate of return on training investments. Intuitively, a higher capital income tax increases the net present value of additional labor earnings resulting from larger training efforts. 9.4.2 Policy Impacts In general, explicit analytical solutions to the model can be found only if one imposes (strong) functional form restrictions on preferences or technologies. To illustrate some of the important interactions described in the previous section, we present some numerical simulations of the model where it is assumed either that OJT investments are fixed (Ht ≡ 1, It ≡ 0)—and labor supply, initial investment, and retirement are endogenous—or that training and retirement are endogenous but labor supply and initial investment in human capital are fixed (i.e., Lt ≡ 1 and S = constant). In the simulations the uncompensated labor supply elasticity is set at 0.18 (if endogenous), the uncompensated retirement elasticity is set at 0.2, and the elasticities of the human capital production function are based on findings from the literature. The baseline policy variables are tL = 0.5, s = 0.75, and z = 0.3. These values match (unweighted) averages for the sample of sixteen continental European and Anglo-Saxon countries we analyzed in section 9.2. Total marginal tax wedges on labor income (including employer contributions and local taxes) are 51 percent for a single household without dependents that earns the average production wage (OECD 2007a). Seventy-nine percent is the average of subsidies on higher education, which probably corresponds better to a marginal subsidy than the average subsidy on all levels of education (OECD 2007b). Gruber and Wise (1999), OECD (2004c), and Duval (2004) show that the implicit tax on work after age 55 amounts to around 30 percent for an older worker aged between 55 and 65, although there are substantial cross-country differences.29 In order to abstract from various complications in determining effective marginal tax rates on capital income, we set capital income taxes to zero in the simulations (tA = 0).30 None of our conclusions depend on this particular assumption. Jacobs (2009a, b) provides more details on the calibration of the models and also extensive sensitivity analyses. Figure 9.24 plots the simulated life-cycle labor supply patterns for various policy experiments for the model where investments in OJT are fixed throughout the life cycle. Each panel shows that individuals are enrolled in initial education for the first years of their life cycle, labor is supplied during working life, and the final years of life are spent in retirement. The time path of labor supply during working lives

302

James J. Heckman and Bas Jacobs

(a) Labor taxation 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81

Age Baseline (τ = 0.5) τ = 0.6

τ = 0.2 τ = 0.7

τ = 0.3 τ = 0.8

τ = 0.4

(c) Implicit tax retirement 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81

Age Baseline (ζ = 0.3) ζ = 0.4

ζ = 0.1 ζ = 0.5

ζ = 0.2 ζ = 0.6

Figure 9.24 Optimal paths of labor supply, education, and retirement for various policies. Baseline parameters: pure rate of time preference r = 0.02, real interest rate r = 0.04, intertemporal elasticity of substitution in consumption q = 2, intertemporal elasticity of labor supply e = 0.5, human capital production function: W(S) ≡ ASa, a = 0.55, A > 0, uncompensated elasticity labor supply 0.18, uncompensated elasticity retirement 0.2, direct costs of education P = 10, time horizon T = 75 years. Baseline policy: tax rate labor tL = 0.5, tax rate savings tA = 0, subsidy on education s = 0.75, and retirement wedge z = 0.3. (See Jacobs 2009a for more details.)

is downward sloping over the life cycle due to income effects as the consumption profile rises with age (not shown). In the baseline simulation, labor supply at the end of the life cycle is around 60 percent. If labor supply is interpreted as the employment rate, the downwardsloping path matches falling labor force participation rates over the life cycle quite well (OECD 2006a). All figures show that various decisions over the life cycle are affected by policy.

Policies to Create and Destroy Human Capital in Europe

303

(b) Education subsidies 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81

Age Baseline (σ = 0.75) σ = 1.0

σ=0 σ = 1.25

σ = 0.25 σ = 1.5

σ = 0.5

(d) International comparison 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81

Age Continental Europe Nordic countries

Mediterranean countries Anglo-Saxon countries

Figure 9.24 Continued

As expected, figure 9.24a shows that higher labor taxes result in less education, lower labor supply, and earlier retirement. The model yields an uncompensated wage elasticity of labor supply of approximately 0.18, which is not extreme. Although taxes directly distort labor supply, human capital investment and retirement, life-cycle interactions between labor supply and human capital decisions reinforce the impact of taxes. As a result very substantial declines in labor supply, retirement ages and years of education are found as labor taxes increase. Indeed the uncompensated elasticity of the tax base is 0.46, which more than twice as large as the simple labor supply elasticity (Jacobs 2009a). Figure 9.24b shows the impact of education policy. This graph most clearly illustrates the importance of the dynamic interactions over the

304

James J. Heckman and Bas Jacobs

life cycle. The subsidy only directly increases human capital investments, not labor supply and retirement. However, higher education subsidies indirectly boost labor supply and the retirement age as individuals become better educated. The costs of leisure and retirement increase with the level of human capital and quite strong effects of higher education subsidies on labor supply and retirement decisions are found in the simulations. Similarly a larger retirement wedge also shows that human capital investments and retirement decisions are interacting; see figure 9.24c. A lower retirement wedge directly results in later retirement and indirectly also in larger investment in human capital. The effects are not very large because returns to education at later points in the life cycle are heavily discounted. Labor supply responds marginally to a lower retirement wedge. Substitution effects in wages—due to a higher education level—are relatively small and income effects—due to a longer working life—are relatively large. Both effects roughly cancel out. Life-cycle interactions only gain in importance when on-the-job training is allowed to be endogenous. Jacobs (2009b) analyzes a model with exogenous labor supply and initial investment in human capital (i.e., Lt ≡ 1 and S = constant), but with endogenous retirement and OJT. Figure 9.25 plots the effects of labor taxes and retirement policies on investments in OJT, retirement ages, and lifetime earnings profiles. Labor earnings drop to zero at the retirement age. Investment in OJT is high at the beginning of the working career, and declines until the retirement age is reached. The reason is that the payback time of investments continuously decreases. The life-cycle profile of labor earnings steadily increases until it peaks and then decreases slightly afterward. This reflects both the investment in OJT before the peak and the depreciation of human capital after the peak. There would be no decline in labor earnings at the end of the life cycle in the absence of depreciation of human capital. Labor productivity (not shown) peaks before earnings, since individuals are still investing some of their time endowment in OJT at the peak in labor productivity (see Ben-Porath 1967; Heckman 1976). The individual also has a valuable time endowment after retirement, although it is steadily depreciating over time. Investment in human capital drops to zero at retirement, since the marginal value of investment in human capital has become zero then. Life-cycle investments in OJT are affected by the labor tax. However, the labor tax impacts human capital investments only indirectly, since all costs of OJT are deductible and labor supply is exogenous. A higher

Policies to Create and Destroy Human Capital in Europe

305

(explicit) tax on retirement reduces OJT investments to a considerable extent, since the payback period of investment in human capital falls. Since retirement is distorted by the presence of the positive implicit tax on retirement, the effect is substantial. Consequently life-cycle earnings profiles shift toward the origin, and earnings peak earlier. Since less time will be invested in OJT, earnings when young increase slightly. However, lower growth of human capital stocks implies that earnings at later ages will be (much) lower. This in turn promotes earlier retirement as the opportunity costs of retirement decrease if wages are lower. A higher implicit tax on retirement gives similar results as a higher labor tax, since it also gives stronger incentives to retire early. As a result investment in human capital declines at all ages. Since less human capital will be accumulated, workers end up with lower wages at the end of their careers and they retire earlier. Thus actuarially very unfair retirement schemes seriously impair investments in OJT over the life cycle. This finding confirms the notion that individuals do not invest in skills because they retire early, and they retire early because they do not invest in skills. The model simulations clearly demonstrate that the policy environment is very important to understand life-cycle interactions between education, labor supply, OJT, and retirement. Human capital formation, labor supply, investment in OJT, and later retirement are complementary. In figure 9.24d we have simulated the model with exogenous OJT with a set of policy parameters that correspond with the unweighted averages of the policies of Anglo-Saxon, continental European, Mediterranean, and Nordic countries; see also section 9.2. We can infer from these simulations that America’s labor supply is larger than in Europe due to the high marginal taxes on work effort in Europe. Moreover the main driver behind the earlier retirement of Europeans is the larger total tax wedge on retirement. Early retirement in turn depresses skill maintenance in OJT and lowers wage growth over the life cycle. The simulations illustrate the vital importance of human capital policy for Europe. Indeed many European countries subsidize initial education much more than the United States, especially in the Nordic countries. Human capital policy is needed to offset the strong disincentives on skill formation created by various welfare state arrangements. Without sufficient skill formation, explicit and implicit taxes on human capital formation would generate too much dependence on welfare states as skill formation, skill utilization, and skill maintenance would be retarded. In other words, there is not only a trade-off between equity

306

James J. Heckman and Bas Jacobs

(a) OJT investment as a fraction of total time over the life cycle for varying labor tax rates 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 20

25

30

35

40

45

50

55

60

65

Age tl =0

tl =0.2

tl =0.4

tl =0.6

tl =0.8

(c) OJT investment as a fraction of total time over the life cycle for varying implicit tax rates on retirement 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 20

25

30

35

40

45

50

55

60

65

70

75

80

Age tr =–0.4 tr =0.4

tr =–0.2 tr =0.6

tr =0

tr =0.2

Figure 9.25 Optimal paths of OJT-investment and retirement for tax and retirement policies. Baseline parameters: pure rate of time preference r = 0.02, real interest rate r = 0.04, intertemporal elasticity of substitution q = 1.25, uncompensated elasticity retirement = 0.2, Cobb– Douglas production function for OJT: F(It, Ht) ≡ (ItHt)a, a = 0.6, time horizon T = 60 years. Baseline policy: tax rate labor tL = 0.5, tax rate savings tA = 0, and retirement wedge z = 0.3. (See Jacobs 2009b for more details.)

and efficiency in the quantity of labor supply but also a trade-off between equity and the quality of labor supply. 9.5

Policy Conclusions

European welfare states that attempt to protect incomes and labor market prospects for persons with low skills face important policy

Policies to Create and Destroy Human Capital in Europe

307

(b) Labor earnings over the life cycle for varying labor tax rates 50 40 30 20 10 0 20

25

30

35

40

45

50

55

60

65

70

75

80

Age tl =0

tl =0.2

tl =0.4

tl =0.6

tl =0.8

(d) Labor earnings over the life cycle for varying implicit tax rates on retirement 60 50 40 30 20 10 0 20

25

30

35

40

45

50

55

60

65

70

75

80

Age tr =–0.4 tr =0.4

tr =–0.2 tr =0.6

tr =0

tr =0.2

Figure 9.25 Continued

challenges. Labor demand has shifted toward the skilled workers as can be witnessed from increasing earnings inequality and the rising returns to education. The growth in the supply of human capital is likely to choke off in years to come. Despite increasing enrollment rates at higher levels of education, resources invested in Europe remain rather stagnant at all levels (except for the Nordic countries) and often more targeted on higher education than preschool and primary education. A substantial fraction of immigrant youths have literacy problems, drop out from secondary education, do not assimilate, and end up disproportionally in crime or welfare state arrangements. Poverty traps result in not only insufficient incentives to work but also insufficient incentives to invest in human capital. As relative demand for unskilled

308

James J. Heckman and Bas Jacobs

labor decreases, low-skill workers become increasingly dependent on welfare state arrangements such as unemployment benefits, public training, and labor market policies. In the end, social cohesion is undermined with a growing divide on labor markets between the skilled and the unskilled and a larger dependency of low-skill workers on welfare state arrangements. European welfare states do not only affect skill creation, but also Europe’s skill utilization is low for a variety of reasons. Hours worked are low and decreasing. Labor force participation rates are relatively low—especially in continental Europe and Mediterranean countries— but increasing that is in part due to larger female participation rates. Take-up rates of benefits for unemployment, sickness, and disability are substantial. Many unemployed workers appear to be hidden in generally ineffective active labor market and training policies. Generous social benefits and high levels of taxation lower labor force participation and hours worked and thereby lower returns on human capital investments. Generous welfare states create substantial implicit taxes on the returns to human capital investments through the interaction with the labor market. Not only is the utilization rate of European human capital low, also the maintenance of human capital is worrisome. Effective retirement ages have fallen dramatically and have landed on a low plateau in recent years. Declining labor force participation rates of older workers are showing signs of a trend reversal in some countries (after controlling for the increase in female labor force participation rates). Individuals spend about a third of their lifetime in retirement. Incentives to retire long before statutory retirement ages are strong due to generous pension and early retirement schemes. Short payback times of investments in human capital and steep depreciation rates of skills undermine the incentives to create and maintain skills through education and on-the-job training. In order to maintain welfare states, human capital policies need to be reinvented. The economic returns to initial investments in human capital are very high. The economic return to investment at older ages is, however, lower. Human capital investment is self-productive and investments at different ages are complementary. Self-productivity and complementarity are the reasons why skill begets skill and learning begets learning. Complementarity implies that early investments need to be followed by later investments if the early investments are to pay off. Investments in the human capital of children should expand for

Policies to Create and Destroy Human Capital in Europe

309

both equity and efficiency reasons. The returns to human capital are high and rising. There is no trade-off between equity and efficiency at early ages of human development but a substantial trade-off at later ages. Once skills are crystallized, complementarity implies that the returns are highest for investment in the most able. At the youngest ages it is possible to form ability and create the stock of skills that enrich late adolescent and early adult human capital investment. Thus early interventions targeted toward the disadvantaged can be highly effective. Later investments are not. Policy should therefore focus heavily on early childhood interventions for children from disadvantaged families. At later ages policies are generally too costly and ineffective. Given public spending constraints, resources should be shifted away from higher education to preschool and primary education. Private funding for higher education should expand, possibly through income contingent loans to warrant access. Labor market and training programs for older workers should be reformed or abolished in their current form as their benefits are doubtful and the costs are high. Successful policies focus on both noncognitive and cognitive skills. The benefits of lower crime rates and socially more acceptable behavior are substantial. A greater emphasis needs to be placed on family policy. Early cognitive and noncognitive deficits can be partially remedied. Dynamic complementarities are not only important for initial investments in life but also for the utilization and the maintenance of skills during working life. Returns to investments early in life will not materialize if early investments are not followed up by later investments. A precondition for sufficient returns to investments during working life is a sufficient level of investment early in life. European human capital policy should take into account the impact of tax-benefit and pension systems and the functioning of labor markets. European labor markets are distorted due to severe labor market regulations, high taxes, generous benefit schemes and insider–outsider problems in wage setting institutions. Eligibility for various types of benefits should become stricter and perverse income redistribution from the outsiders toward the insiders through all kinds of benefit schemes should be reduced. This is both efficient and equitable, since the outsiders on labor markets are hurt by the privileged insiders. Reducing distortions in labor markets increases the utilization rates of human capital and enlarges the benefits of initial education and skill maintenance over the life cycle. Retirement is heavily subsidized via early retirement schemes

310

James J. Heckman and Bas Jacobs

and pension subsidies. These policies create perverse incentives to utilize and maintain human capital over the life cycle. Therefore earlyretirement schemes and pension plans should be made actuarially fairer. Appendix The Hamiltonian for maximizing lifetime utility is given by T

R

0

S

H ≡ ∫ U (Ct ) exp ( − ρt ) dt + ∫ V (Lt ) exp ( − ρt ) dt + X (T − R) R T + λ ⎡ ∫ (1 − τ L ) W (S ) Ht Lt exp ( − r *t ) dt + ∫ B exp ( − r *t ) dt ⎤ R ⎣ S ⎦ T S − λ ⎡ ∫ Ct exp ( − r *t ) dt + ∫ (1 − s) P exp ( − r *t ) dt ⎤ 0 ⎣ 0 ⎦ + μt [G (S) F ( It , Ht ) − δ Ht ] + vt [1 − Lt − It − Lt ] ,

(9A.1)

where l is the marginal utility of lifetime income, mt is the co-state variable at time t associated with the on-the-job human capital accumulation equation, and nt is the shadow value of the time constraint at time t. First-order conditions for a maximum are given by ∂H = U ′ (Ct ) exp ( − ρt ) − λ exp ( − r *t) = 0, ∂Ct ∂H = V ′ (Lt) exp ( − ρt ) − vt = 0, ∂Lt

0 ≤ t ≤ T,

S < t < R,

∂H = λ (1 − τ L ) W (S) Ht exp ( − r *t) − vt = 0, ∂Lt

(9A.2) (9A.3)

S < t < R,

(9A.4)

∂H = −V (LS ) exp ( − ρS ) − λ (1 − s) P exp ( − r *S) ∂S − λ (1 − τ L ) W (S) HS LS exp ( − r *S S) + λ ∫ (1 − τ L ) W ′ (S) Ht Lt exp ( − r *t ) dt + μSG ′ (S) F (I S , HS ) = 0 , R

S

(9A.5)

∂H = V ′ (LR ) exp (− ρR ) − X ′ (T − R) ∂R + λ ((1 − τ L ) W (S) H R LR − B) exp ( − r *R) = 0,

(9A.6)

∂H = μtG (S) FI ( It , Ht ) − vt = 0 , ∂I t

(9A.7)

S < t < R,

Policies to Create and Destroy Human Capital in Europe

∂H = λ (1 − τ L ) W (S) Lt exp ( − r *t ) + μt [G (S) FH (It , Ht ) − δ ] ∂H t = − μ t , S < t < R.

311

(9A.8)

In addition we have to impose a transversality condition on the co-state variable that mR = 0. The equations in the main text are derived as follows. Note that l = U′(C0) from the first-order condition for C0. Totally differentiating the first-order condition for Ct with respect to time gives the Euler equation for consumption in equation (9.9). Combining the first-order conditions for Ct, Lt, and Lt yields equation (9.10). Equation (9.13) can be found by substituting the first-order condition for Lt into the first-order condition for It. To derive equation (9.11), we obtain from the first-order conditions for Lt and Lt: V (Lt ) 1 exp ( − ρt ) = (1 − τ L ) W (S ) Ht Lt, λt εt

(9A.9)

where lt ≡ U′(C0) exp(−r*t), and et ≡ V′(Lt)Lt[V(Lt)]−1. Substituting (9.13) and (9A.9) in the first-order condition for S and using HS ≡ 1 yields equation (9.11). Similarly we substitute (9A.9) in the retirement equation to find (9.12). Finally, we totally differentiate equation (9.13) to find that −

μ t FII It It ⎛ FIH Ht ⎞ H t = +⎜ − 1⎟ + r *. ⎠ Ht μt FI It ⎝ FI

(9A.10)

Substitute this result and equation (9.13) in the first-order condition for Ht : G (S) FH ( It , Ht ) +

t G (S) FI ( It , Ht ) Lt FII It It ⎛ FIH Ht ⎞ H = r* + δ . − + ⎜1− ⎟ Ht FI It ⎝ FI ⎠ Ht (9A.11)

Homogeneous functions of degree f have the property that its partial derivatives are homogeneous of degree f − 1. Consequently we can substitute FIIIt = (f − 1) FI − FIHHt into the last equation: G(S)FI ( It , Ht ) Lt Ht t  ⎛ F F F H FIH F FH H t ⎞ H ⎛ IH H t ⎞ It + ⎜1− φ + + ⎜1− = r * +δ . ⎟ ⎟ ⎝ FI FH F ⎠ I t ⎝ FI FH F ⎠ H t

G (S) FH ( It , Ht ) +

(9A.12)

We assume that F(It, Ht) ≡ [Φ(It, Ht)]f, which has the following derivatives: FI = fΦf−1ΦI, FH = fΦf−1ΦH, FIH = fΦf−1ΦIH(1 + (f − 1)s). Therefore

312

James J. Heckman and Bas Jacobs

we find FHH/F = fwH and FIHF/(FIFH) = [1 + (f − 1)s]/(fs). Substitution of these results yields equation (9.14). Notes The writing of this chapter was generously supported by the National Science Foundation (SES-0241858, SES-0099195, SES-0452089); the National Institute of Child Health and Human Development (R01HD43411); a grant to the Committee on Economic Development PAES Group from the PEW Foundation; support from the J.B. Pritzker Consortium on Early Childhood Development at the Harris School of Public Policy, University of Chicago; and the Netherlands Organisation for Scientific Research (NWO Vidi Grant 452–07–013, “Skill Formation in Distorted Labour Markets”). An extended version of this chapter was presented at the joint conference of CESifo and Center on Capitalism and Society, “Perspectives on the Performance of the Continent’s Economies,” Venice International University, San Servolo, July 21–22, 2006. We especially thank Dennis Snower, Edmund Phelps, Hans-Werner Sinn, and the participants for their useful comments and suggestions. In addition we thank Lans Bovenberg, George Gelauff, Coen Teulings, and participants of the Netspar-SER-CPB conference, “Reinventing the Welfare State,” April 27–28, 2006, The Hague, The Netherlands, for their comments and suggestions. 1. This finding is also reported for the United States by Autor, Katz, and Kearney (2008). 2. Numerous studies find only small impacts of larger tuition rates on enrollment (Kane 1994, 1995; Hilmer 1998; Heckman et al. 1998; Dynarski 2003; Card and Lemieux 2001; Cameron and Heckman 2001). Part of the explanation is that tuition costs are a relatively minor fraction of the total costs of education since forgone labor earnings are by far the most important part (Becker 1964). Another explanation is that psychic costs play a substantial role in explaining college choices (Cunha, Heckman, and Navarro 2005). 3. Carneiro and Lee (2010) show that standard measures for skill premia between higher and lower educated workers are even biased downward due to selection into higher education on nonobserved characteristics. 4. Cunha and Heckman (2008) and Heckman, Lochner, and Todd (2006) survey a large number of studies that show that nonpecuniary factors (associated with psychic costs, motivations, etc.) play a major role in explaining why minorities and persons from lowincome families do not attend college even though it is financially profitable to do so. Returns to schooling for marginal entrants attracted into college by changes in tuition are below those of the average participant. Returns to schooling are lower for people less likely to attend college. 5. De la Fuente and Doménech (2006) have constructed a panel data set for OECD countries of educational attainment for various education levels and average years of education based on data from national statistical offices which are supplemented by data from the OECD. 6. This coefficient gives the ratio (1 − marginal tax rate)/(1 − average tax rate) and is smaller than one if the marginal taxes are higher than average taxes. 7. This was first noted by Heckman and Klenow (1998). 8. Standard measures for the returns to education or the skill premium rarely allow for the utilization rate of human capital, however.

Policies to Create and Destroy Human Capital in Europe

313

9. These findings are consistent with a “social multiplier” for leisure demand as hypothesized by Alesina et al. (2005). 10. All countries have witnessed declines of labor force participation rates of 60- to 64-year-old cohorts since 1960 (not shown). Dramatic declines are found in Finland, France, Germany, Netherlands, and Spain. In Australia, Canada, Denmark, Norway, Sweden, and the United States the decline is to an important extent offset by increases in female force participation. In the other countries this offsetting effect has been largely absent as female participation rates were falling too. However, more recent increases in Canada, Netherlands, and Sweden (as from 1990) are the result of increasing male and female labor force participation rates. This is, again, likely to be the result of changes in early retirement schemes. 11. We have to note that some countries have started to reform their early retirement schemes in recent years, for example, in the Netherlands. The graph may therefore give a too pessimistic view of the adverse incentives to retire early. 12. This section draws upon research by Heckman (2000); Carneiro and Heckman (2003); Cunha, Heckman, Lochner, and Masterov (2006); Carneiro, Cunha, and Heckman (2005); and Cunha and Heckman (2007, 2008) that develops the economic foundation for skill acquisition in modern economies. 13. In the popular literature, achievement tests and IQ tests are often confused. Achievement test scores are affected by IQ, schooling inputs, and noncognitive skills, and are malleable over a much greater range of ages than is IQ (see Hansen, Heckman, and Mullen 2004; Cunha and Heckman 2008; Cunha, Heckman, and Schennach 2010). Abilities have an acquired character although they differ in their malleability at different ages. 14. The General Educational Development (GED) program allows secondary-school individuals to obtain certification through an equivalency exam administered to dropouts that is comparable to a high school degree. 15. Some macroeconomic cross-country studies by Layard, Nickell, and Jackman (1991) and Nickell (1997) suggest that active labor market programs can be effective in reducing unemployment rates, but these studies do not control for country-specific effects. 16. For simplicity of exposition we do not work with the more general model of skill formation developed and estimated by Cunha, Heckman, and Schennach (2010). 17. One should bear this qualifier in mind at our discussion of the model simulations. These simulations are only meant to stress the importance of various complementarities over the life cycle and these complementarities are in our view relevant in both competitive and noncompetitive labor markets. 18. The separability between leisure and retirement from consumption in the utility function avoids discontinuities in the marginal utility of consumption. 19. Initial education is acquired in both families and schools. Our current formulation lumps human capital formation in families and schools together during the initial phase. However, human capital formation in schools and families are by no means perfect substitutes; for example, see Cunha, Heckman, Lochner, and Masterov (2006) for more on this. Investment in on-the-job human capital takes place mainly in firms. The degree of complementarity between human capital formation through schools/families and firms is captured by the productivity of learning in the on-the-job human capital production function.

314

James J. Heckman and Bas Jacobs

20. We assume that first-order conditions are necessary and sufficient. The latter condition is not necessarily fulfilled due to the feedbacks between labor supply and human capital accumulation. In order to guarantee an interior solution, elasticities of human capital decisions (schooling and training) and labor supply decisions (work effort, retirement) should not be too high. Otherwise, higher investments in human capital (schooling and training) will boost labor supply (work effort and retirement), which in turn increases the return to human capital investments. This increases human capital investment and labor supply expands in a second round, which again increases human capital investments, and so on. Only sufficiently strong decreasing returns in schooling and training and a sufficiently concave leisure and retirement subutility functions ensure an interior solution. We assume that these conditions are met. See Arrow and Kurz (1970), Blinder and Weiss (1976), and Heckman (1975) for discussions of sufficiency conditions. 21. Browning, Hansen, and Heckman (1999) survey the estimates of qt obtained from micro and macro data. 22. Note that HS ≡ 1, since individuals do not train if they are not in the labor market yet. 23. The term Ls + Ls/es originates from the fact that more time spent on initial education lowers the time span over which labor can be supplied or leisure can be consumed; see the utility function. LS is associated with the marginal loss of forgone labor earnings, and Ls/es is associated with marginal forgone leisure time when individuals invest more time in initial education. In the absence of endogenous leisure demand during working life, LS = 0, and Ls + Ls/es equals (1 − IS). The same is true if the subutility function over leisure V(LS) is linear, namely when eS = 1. In that case more time spent in initial education reduces the marginal value of working and leisure time during working life equally. 24. Note that IR = 0 at the end of the working life. 25. Again, there is a term (1 + z)LR + (1 + LR)/eR representing the impact of retirement on the time span over which individuals enjoy labor earnings and ordinary leisure. (1 − z)LR corresponds with the marginal loss of forgone labor earnings, which are reduced one for one with the implicit tax on retirement due to actuarially unfair pensions. (1 − LR)/eR measures the marginal value of forgone leisure time when individuals retire earlier. In the absence of an endogenous leisure demand decision and actuarially fair pensions (LR = 1 and z = 0) this term would vanish. Similarly the last term cancels out if the leisure subutility function V(LR) is linear (eR = 1), and pensions are actuarially fair (z = 0). Later retirement then augments the marginal value of working and leisure time equally. 26. Ben-Porath (1967) is a special case of the current model. 27. This analysis mirrors those of Cunha and Heckman (2007) and Heckman (2007). 28. Note that direct costs of training are absent. 29. Gruber and Wise (1999) report the so-called tax force statistic, which corresponds to the sum of marginal tax wedges on retirement while working during ages 55 to 69. Dividing the “tax force” by 15 gives a yearly average marginal tax wedge on retirement during working ages 55 to 69. OECD (2004c) computes marginal tax wedges on retirement that are around 20 percent (40 percent) on average for 55- to 59-year-old (60- to 64-year-old) workers. Duval (2004: 33) calculates that average implicit tax rates in OECD countries are equal to 30 percent.

Policies to Create and Destroy Human Capital in Europe

315

30. Good measures of marginal effective tax rates on savings are difficult to obtain as institutional details are crucial. Taxes on savings can easily be shifted to labor (consumption), which we cannot account for since we use a partial equilibrium setup. Corporate income taxes play a role as well in determining the effective tax burden on savings; for example, see Carey and Rabesona (2004).

References Abbring, J. H., G. J. van den Berg, and J. C. van Ours. 2005. The effect of unemployment insurance sanctions on the transition rate from unemployment to employment. Economic Journal 115: 602–30. Alesina, A. F., E. L. Glaeser, and B. Sacerdote. 2005. Work in the US and Europe: Why so different? NBER Macroeconomic Annuals 2: 1–64. Altonji, J., and T. Dunn. 1996. The effects of family characteristics on the return to education. Review of Economics and Statistics 78 (4): 692–704. Arrow, K. J. 1973. Higher education as a filter. Journal of Public Economics 2: 193–216. Arrow, K. J., and M. Kurz. 1970. Public Investment, the Rate of Return, and Optimal Fiscal Policy. Baltimore: Johns Hopkins University Press. Atkinson, A. B. 2008. The Changing Distribution of Earnings in OECD Countries. Oxford: Oxford University Press. Atkinson, A. B., and W. Salverda. 2005. Top incomes in the Netherlands and the United Kingdom over the twentieth century. Journal of the European Economic Association 3 (4): 883–913. Autor, D. H., L. F. Katz, and M. S. Kearny. 2008. Trends in U.S. wage inequality: Reassessing the revisionists. Review of Economics and Statistics 90 (2): 300–23. Becker, G. S. [1964] 1993. Human Capital: A Theoretical and Empirical Analysis with Special Reference to Education, 3rd ed. Chicago: Chicago University Press. Ben-Porath, Y. 1967. The production of human capital and the life cycle of earnings. Journal of Political Economy 75 (4): 352–65. Bertola, G. 2003. Earnings disparities in OECD countries: Structural trends and institutional influences. Paper prepared for the joint OECD/NDRC Understanding China’s Income Disparities seminar, Paris, October 20–21. Björklund, A., M. Lindahl, and E. J. S. Plug. 2006. The origins of intergenerational associations: Lessons from Swedish adoption data. Quarterly Journal of Economics 121 (3): 999–1028. Blanchard, O. J. 2006. European unemployment: The evolution of facts and ideas. Economic Policy 21 (45): 5–59. Blinder, A. S., and Y. Weiss. 1976. Human capital and labor supply: A synthesis. Journal of Political Economy 84 (3): 449–72. Blundell, R., and T. MaCurdy. 1999. Labor supply: A review of alternative approaches. In O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, vol. 3A. Amsterdam: Elsevier–North Holland, 1559–1695.

316

James J. Heckman and Bas Jacobs

Borghans, L., A. L. Duckworth, J. J. Heckman, and B. ter Weel. 2008. The economics and psychology of personality traits. Journal of Human Resources 43 (4): 972–1059. Bovenberg, A. L. 2006. Tax policy and labor market performance. In J. Agell and P. B. Sørensen, eds., Tax Policy and Labor Market Performance. Cambridge: MIT Press, 3–74. Bovenberg, A. L., J. J. Graafland, and R. A. de Mooij. 2000. Tax reform and the Dutch labor market. Journal of Public Economics 78: 193–214. Bovenberg, A. L., and B. Jacobs. 2003. On the optimal distribution of income and education. Unpublished manuscript. University of Amsterdam/Tilburg. Bovenberg, A. L., and B. Jacobs. 2005. Redistribution and education subsidies are Siamese twins. Journal of Public Economics 89: 2005–35. Bovenberg, A. L., and F. van der Ploeg. 1994. Effects of the tax and benefit system on wage formation and unemployment. Unpublished manuscript. University of Amsterdam/Tilburg. Brandolini, A., and T. M. Smeeding. 2008. Inequality: International evidence. In S. N. Durlauf and L. E. Blume, eds., The New Palgrave Dictionary of Economics, 2nd ed. London: Palgrave Macmillan: 1013–21. Browning, M., L. P. Hansen, and J. J. Heckman. 1999. Micro data and general equilibrium models. In J. B. Taylor and M. Woodford, eds., Handbook of Macroeconomics, vol. 1A. Amsterdam: Elsevier-North Holland, 543–633. Calmfors, L., A. Forslund, and M. Hemström. 2001. Does active labour market policy work? Lessons from the Swedish experiences. Swedish Economic Policy Review 85: 61–124. Cameron, S. V., and J. J. Heckman. 2001. The dynamics of educational attainment for blacks, Hispanics, and whites. Journal of Political Economy 109 (3): 455–99. Card, D. 1999. The causal effect of education on earnings. In O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, vol. 3A. Amsterdam: Elsevier–North Holland, 1801–63. Card, D., and T. Lemieux. 2001. Dropout and enrollment trends in the post-war period: What went wrong in the 1970s? In J. Gruber, ed., An Economic Analysis of Risky Behavior among Youth. Chicago: University of Chicago Press for NBER, 439–82. Carey, D., and J. Rabesona. 2004. Tax ratios on labour and capital income and on consumption. In P. B. Sørensen, ed., Measuring the Tax Burden on Capital and Labor. Cambridge: MIT Press, 213–262. Carneiro, P., F. Cunha, and J. J. Heckman. 2005. Human capital policy for Europe. Unpublished manuscript. University College London/University of Chicago. Carneiro, P., and J. J. Heckman. 2003. Human capital policy. In J. J. Heckman and A. B. Krueger, eds., Inequality in America: What Role for Human Capital Policy? Cambridge: MIT Press, 77–240. Carneiro, P., J. J. Heckman, and E. Vytlacil. 2006. Estimating the rate of return to education when it varies among individuals. Unpublished manuscript. University of Chicago. Carneiro, P., K. Hansen, and J. J. Heckman. 2001. Removing the veil of ignorance in assessing the distributional impacts of social policies. Swedish Economic Policy Review 8: 273–301.

Policies to Create and Destroy Human Capital in Europe

317

Carneiro, P., K. Hansen, and J. J. Heckman. 2003. Estimating distributions of treatment effects with an application to the returns to schooling. International Economic Review 44 (2): 361–422. Carneiro, P., and S. Lee. 2010. Trends in quality-adjusted skill premia in the United States, 1960–2000. American Economic Review, forthcoming. Ciccone, A., and G. Peri. 2006. Identifying human capital externalities: Theory with an application to US cities. Review of Economic Studies 73: 381–412. Cunha, F., and J. J. Heckman. 2007. Identifying and estimating the distributions of ex post and ex ante returns to schooling. Labour Economics 14 (6): 870–93. Cuhna, F., and J. J Heckman. 2007. The technology of skill formation. American Economic Review 97 (2): 31–47. Cunha, F., and J. J. Heckman. 2008. Formulating, identifying and estimating the technology of cognitive and noncognitive skill formation. Journal of Human Resources 43 (4): 738–82. Cunha, F., J. J. Heckman, L. J. Lochner, and D. V. Masterov. 2006. Interpreting the evidence on life cycle skill formation. In E. Hanushek and F. Welch, eds., Handbook of Economics of Education. Amsterdam: Elsevier–North Holland, 697–812. Cunha, F., J. J. Heckman, and S. Navarro. 2005. Separating uncertainty from heterogeneity in life cycle earnings. Oxford Economic Papers 57 (2): 191–261. Cunha, F., J. J. Heckman, and S. M. Schennach. 2010. Estimating the technology of cognitive and noncognitive skill formation. Econometrica 78 (3): 883–931. Davis, S. J. 1992. Cross-country patterns of changes in relative wages. In O. J. Blanchard and S. Fischer, eds., NBER Macroeconomic Annuals 1992. Cambridge: MIT Press, 239–400. De la Fuente, A., and R. Doménech. 2006. Human capital in growth regressions: How much difference does data quality make? Journal of the European Economic Association 4 (1): 1–36. De la Fuente, A., and J. F. Jimeno-Serrano. 2005. The private and fiscal returns to schooling and the effect of public policies on private incentives to invest in education: A general framework and some results for the EU. Working paper 1392. CESifo, Munich. Duval, R. 2004. Retirement behavior in OECD countries: Impact of old-age pension schemes and other social transfer programmes. OECD Economic Studies 37: 7–50. Dynarski, S. 2003. Does aid matter? Measuring the effect of student aid on college attendance and completion. American Economic Review 93 (1): 279–88. Edin, P.-A., and B. Holmlund. 1995. The Swedish wage structure: The rise and fall of wage policy? In R. B. Freeman and L. F. Katz, Differences and Changes in Wage Structures. Chicago: Chicago University Press, 307–43. European Commission. 2005. Study on access to education and training, basic skills and early school leavers (Ref. DG EAC 38/04). Lot 3: Early school leavers final report. European Commission DG EAC, Brussels. Frederiksson, P. 1997. Economic incentives and the demand for higher education. Scandinavian Journal of Economics 99 (1): 129–42.

318

James J. Heckman and Bas Jacobs

Gordon, K., and H. Tchilinguirian. 1998. Marginal effective tax rates on physical, human and R&D capital. Working paper ECO/WKP(98)12. OECD, Paris. Gottschalk, P., and T. M. Smeeding. 1996. Cross-national comparisons of earnings and income inequality. Journal of Economic Literature 35 (2): 633–87. Gruber, J., and D. Wise. 1999. Social Security and Retirement around the World. Chicago: Chicago University Press. Hanushek, E., and D. D. Kimko. 2000. Schooling, labor force quality, and the growth of nations. American Economic Review 90 (5): 1184–1208. Hansen, K., J. J. Heckman, and K. J. Mullen. 2004. The effect of schooling and ability on achievement test scores. Journal of Econometrics 121 (1–2): 39–98. Harmon C., H. Oosterbeek, and I.Walker. 2003. The returns to education: Microeconomics. Journal of Economic Surveys 17: 115–55. Heckman, J. J. 1975. Estimates of a human capital production function embedded in a life-cycle model of a labor supply. In N. E. Terleckyj, ed., Household Production and Consumption. New York: National Bureau of Economic Research. Heckman, J. J. 1976. A life cycle model of earnings, learning and consumption. Journal of Political Economy 4: S11–44. Heckman, J. J. 1993. What has been learned about labor supply in the past twenty years? American Economic Review 83: 116–21. Heckman, J. J. 2000. Policies to foster human capital. Research in Economics 54 (1): 3–56. Heckman, J. J. 2006. Skill formation and the economics of investing in disadvantaged children. Science 312 (5782): 1900–02. Heckman, J. J. 2007. The economics, technology, and neuroscience of human capability formation. Proceedings of the National Academy of Sciences 104 (3): 13250–55. Heckman, J. J., N. Hohmann, M. Khoo, and J. Smith. 2000. Substitution and dropout bias in social experiments: A study of an influential social experiment. Quarterly Journal of Economics 115 (2): 651–94. Heckman, J. J., and B. Jacobs. 2006. Reinventing human capital policy. Paper prepared for the Netspar-SER-CPB conference, “Reinventing the Welfare State,” April 27–28, The Hague. Browning, M., L. P. Hansen, and J. J. Heckman. 1999. Micro data and general equilibrium models. In J. B. Taylor and M. Woodford, eds., Handbook of Macroeconomics, vol. 1A. Amsterdam: Elsevier, 543–633. Heckman, J. J., J. Hsee, and Y. Rubinstein. 2001. The GED is a “mixed signal”: The effect of cognitive and noncognitive skills on human capital and labor market outcomes. Unpublished manuscript. University of Chicago. Heckman, J. J., and P. J. Klenow. 1998. Human capital policy. In M. Boskin, ed., Policies to Promote Human Capital Formation. Palo Alto, CA: Hoover Institution. Heckman, J. J., and P. LaFontaine. 2006. Bias corrected estimates of GED returns. Journal of Labor Economics 24 (3): 661–700.

Policies to Create and Destroy Human Capital in Europe

319

Heckman, J. J., R. J. Lalonde, and J. A. Smith. 1999. The economics and econometrics of active labor market programs. In O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, vol. 3A. Amsterdam: Elsevier–North Holland, 1865–2097. Heckman, J. J., M. Ljunge, and K. S. Ragan. 2006. What are the key employment challenges and policy priorities for OECD countries. Paper presented at the “Boosting Jobs and Incomes” conference, Toronto, June. Heckman, J. J., L. J. Lochner, and C. Taber. 1998. Explaining rising wage inequality: Explorations with a dynamic general equilibrium model of labor earnings with heterogeneous agents. Review of Economic Dynamics 1: 1–58. Heckman, J. J., L. J. Lochner, and P. E. Todd. 2006. Earnings functions, rates of return and treatment effects: The Mincer equation and beyond. In E. Hanushek, and F. Welch, eds., Handbook of Economics of Education. Amsterdam: Elsevier–North Holland, 307–458. Heckman, J. J., L. Malofeeva, R. Pinto, and P. A. Savelyev. 2010. Understanding the mechanisms through which an influential early childhood program boosted adult outcomes. American Economic Review, under review. Heckman, J. J., and D. V. Masterov. 2006. The productivity argument for investing in young children. Discussion paper. Early Childhood Research Collaborative, University of Minnesota. Heckman, J. J., and C. Pagés. 2003. Law and unemployment: Lessons from Latin America and the Carribbean. Working paper 10129. NBER, Cambridge, MA. Heckman, J. J., and C. Pagés. 2004. Law and Employment: Lessons from Latin American and the Caribbean. Chicago: University of Chicago Press. Heckman, J. J., and Y. Rubinstein. 2001. The importance of noncognitive skills: Lessons from the GED testing program. American Economic Review 91 (2): 145–49. Heckman, J. J., J. Stixrud, and S. Urzua. 2006. The effects of cognitive and noncognitive abilities on labor market outcomes and social behavior. Journal of Labor Economics 24 (3): 411–82. Hilmer, M. J. 1998. Post-secondary fees and the decision to attend a university or a community college. Journal of Public Economics 67: 329–48. Jacobs, B. 2004. The lost race between schooling and technology. De Economist 152 (1): 47–78. Jacobs, B. 2005. Optimal income taxation with endogenous human capital. Journal of Public Economic Theory 7 (2): 295–315. Jacobs, B. 2007. Optimal tax and education policies and investments in human capital. In J. Hartog and H. Maassen van den Brink, Human Capital: Moving the Frontier. Cambrige: Cambridge University Press, 212–32. Jacobs, B. 2009a. Is Prescott right? Welfare state policies and the incentives to work, learn and retire. International Tax and Public Finance 16: 253–80. Jacobs, B. 2009b. A life-cycle theory of human capital formation, pension saving and retirement. Unpublished manuscript. Erasmus School of Economics, Erasmus University Rotterdam.

320

James J. Heckman and Bas Jacobs

Jacobs, B., and F. van der Ploeg. 2006. Guide to reform of higher education: A European perspective. Economic Policy 21 (47): 536–92. Kane, T. J. 1994. College entry by blacks since 1970: The role of college costs, family background and the returns to education. Journal of Political Economy 102 (5): 878–911. Kane, T. J. 1995. Rising public tuition and college entry: How well do public subsidies promote access to college. Working paper 5164. NBER, Cambridge. Katz, L. F., and K. M. Murphy. 1992. Changes in relative wages 1963–1987: Supply and demand factors. Quarterly Journal of Economics 107: 35–78. Katz, L. F., and D. H. Autor. 1999. Changes in the wage structure and wage inequality. In O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, vol. 3A. Amsterdam: Elsevier–North Holland, 1463–1555. Killingsworth, M. R., and J. J. Heckman. 1986. Female labor supply: A survey. In O. Ashenfelter and R. Layard, eds., Handbook of Labor Economics, vol. 1. Amsterdam: Elsevier–North Holland, 103–204. Knudsen, E., J. J. Heckman, J. Cameron, and J. Shonkoff. 2006. Economic, neurobiological and behavioral perspectives on building America’s future workforce. Proceedings of the National Academy of Sciences 103 (27): 10155–62. Krueger, A. B., and M. Lindahl. 2001. Education for growth: Why and for whom? Journal of Economic Literature 39: 1101–36. Lalive, R., J. C. van Ours, and J. Zweimuller. 2005. The effects of benefit sanctions on the duration of unemployment. Journal of the European Economic Association 3: 1386–1417. Lalive, R., J. C. van Ours, and J. Zweimuller. 2006. How changes in financial incentives affect the duration of unemployment. Review of Economic Studies 73 (4): 1009–38. Layard, R., S. Nickell, and R. Jackman. 1991. Unemployment, Oxford: Oxford University Press. Lindbeck, A., and S. Nyberg. 2006. Raising children to work hard: Altruism, work norms and social insurance. Quarterly Journal of Economics 121 (4): 1473–1503. Ljunge, M. 2006. Half the job is showing up: Returns to work, taxes, and sick leave choices. PhD dissertation. University of Chicago. Ljungqvist, L., and T.J. Sargent. 1998. The European unemployment dilemma. Journal of Political Economy 106 (3): 514–50. Ljungqvist, L., and T. J. Sargent. 2002. The European employment experience. Discussion paper 3543. CEPR, London. Martin, J., and D. Grubb. 2001. What works and for whom: A review of OECD countries’ experience with active labour market policies. Swedish Economic Policy Review 8 (2): 9–56. Mincer, J. 1974. Schooling, Experience, and Earnings. New York: Columbia University Press/NBER. Murphy, K. M., W. C. Riddell, and P. M. Romer. 1998. Wages, skills, and technology in the United States and Canada. In E. Helpman, ed., General Purpose Technologies and Economic Growth. Cambridge: MIT Press, 283–309.

Policies to Create and Destroy Human Capital in Europe

321

Nahuis, R., and H. L.F. de Groot. 2003. Rising skill premia: You ain’t seen nothing yet? Discussion paper 20. Bureau for Economic Policy Analysis, CPB Netherlands, The Hague. Nickell, S. 1997. Unemployment and labor market rigidities: Europe versus North America. Journal of Economic Perspectives 11 (3): 55–74. Nickell, S. 2003. Poverty and worklessness in Britain. Economic Journal 114: C1–25. OECD. 2003. Education at a Glance. Paris: OECD. OECD. 2004a. Learning for Tomorrow’s World First Results from PISA 2003. Paris: OECD. OECD. 2004b. Benefits and Wages., Paris: OECD. OECD. 2004c. The labor force participation of older workers: The effects of pension and early retirement schemes, Working paper. Economics Department, OECD. OECD. 2005a. Education at a Glance. Paris: OECD. OECD. 2005b. OECD Factbook. Paris OECD. OECD. 2005c. OECD Social Indicators. Paris: OECD. OECD. 2005d. Pensions at a Glance. Paris: OECD. OECD. 2005e. OECD Tax Database. Paris: OECD. OECD. 2006a. OECD Labor Force Statistics Database. Paris: OECD. OECD. 2006b. OECD Social Indicators. Paris: OECD. OECD. 2006c. OECD Education Database. Paris: OECD. Ours, J. C. van, and M. Belot. 2001. Unemployment and labor market institutions: An empirical analysis. Journal of Japanese and International Economies 15 (4): 403–18. Pencavel, J. 1986. Labor supply of men: A survey. In O. Ashenfelter and R. Layard, eds., Handbook of Labor Economics, vol. 1. Amsterdam: Elsevier–North Holland, 3–102. Peracchi, F. 2006. Educational wage premia and the distribution of earnings: An international perspective. In E. Hanushek and F. Welch, eds., Handbook of Economics of Education. Amsterdam: Elsevier–North Holland, 189–254. Piketty, T. 2003. Income inequality in France, 1901–1998. Journal of Political Economy 111: 1004–42. Piketty, T., and E. Saez. 2003. Income inequality in the United States 1913–1998. Quarterly Journal of Economics 118 (1): 1–39. Pissarides, C. A. 1998. The impact of employment tax cuts on unemployment and wages: The role of unemployment benefits and tax structure. European Economic Review 42: 155–83. Phelps, E. S., ed. 2003. Designing Social Inclusion: Tools to Raise Low-End Pay and Employment in Private Enterprise. Cambridge: Cambridge University Press. Ploeg, F. van der. 2006. Do social policies harm employment? Second-best effects of taxes and benefits on labor markets. In J. Agell and P. B. Sørensen, eds., Tax Policy and Labor Market Performance. Cambridge: MIT Press, 97–144.

322

James J. Heckman and Bas Jacobs

Prescott, E. C. 2004. Why do Americans work so much more than Europeans? Federal Reserve Bank of Minneapolis Quarterly Review 28 (1): 2–13. Raudenbush, S. W. 2006. Schooling, statistics and poverty: Measuring school improvement and improving schools. Inaugural lecture. Division of Social Sciences, University of Chicago. Saez, E. 2002. Optimal income transfer programs: Intensive versus extensive labor supply responses. Quarterly Journal of Economics 117: 1039–73. Sianesi, B., and J. van Reenen. 2002. The returns to education: Macroeconomics. Journal of Economic Surveys 17 (2): 1–114. Sørensen, P. B. 1997. Public finance solutions to the European unemployment problem. Economic Policy 12 (25): 221–64. Sørensen, P. B. 1999. Optimal tax progressivity in imperfect labour markets. Labour Economics 6 (3): 435–52. Taber, C. 2001. The rising college premium in the eighties: Return to college or return to unobserved ability? Review of Economic Studies 68 (3): 665–91. Turkheimer, E., A. Haley, M. Waldron, B. D’Onofrio, and I. I. Gottesman. 2003. Socioeconomic status modifies heritability of IQ in young children. Psychological Science 14 (6): 623–28.

10

Market Forces and the Continent’s Growth Problem Gylfi Zoega

10.1

Introduction

Entrepreneurs are the inventors of business ideas and if successful generate followers who imitate their success. The cultural and institutional factors that affect entrepreneurship help explain economic growth and differences in economic performance between countries. But it is also important to study what determines the rate at which these ideas are transmitted from a world leader to each country’s business leader and then from that leader to local followers within the country. The economic performance of the larger continental European economies in recent decades has lagged behind that of the United States in terms of entrepreneurship. Productivity data reveal that the US productivity level fell relative to the average in a group of eighteen countries in the 1960s and the 1970s as Europe and Japan caught up with it. While Europe could benefit from unexploited business ideas in the first two decades following the war and enjoyed growth by learning about and adopting ideas that had been generated by American entrepreneurs in the pre-war decades, the pool of unexploited ideas diminished as the productivity gap between Europe and the United States became smaller. Continental Europe appears to lack dynamism, defined as the social factors that promote entrepreneurship, be they cultural, institutional or market forces. In the model proposed in this chapter, business innovations take place in leading firms in different countries, and these innovations then spread to other domestic and foreign firms. It is assumed that genuinely original ideas do not require much input on behalf of the entrepreneur. Instead, individuals have different intuitions about how the world works and which ideas are likely to generate profits;

324

Gylfi Zoega

these ideas reflect the accumulated experience, education, and lessons learned by individuals, as well as personal attributes and the quality and perspectives of their social circles. A model of this kind may shed light on the causes of the lack of dynamism observed in some of the European economies. The final section explores data on productivity growth and institutions. 10.2

Brand Kindles Brand

Most business people are not the inventors of new business ideas but instead adopt ideas conceived of by others. Managers of businesses spend part of their time supervising and organizing the work of others and part of it learning, adapting and implementing business solutions learned from others. One problem facing the manager is to choose the fraction of time h he spends on actual production using existing knowledge A and the fraction 1 – h spent studying, evaluating and adopting new ideas with the view of maximizing profits. 10.2.1 Local Adoption of Ideas Assume that each firm is owned and operated by a manager who combines business knowledge A with his education E and inputs X—which could be labor or, alternatively, intermediate inputs—in producing output Y. The fraction of his time spent producing is denoted byh, leaving the fraction 1 – h for him to study and adopt new business ideas. Business knowledge is measured by the number of adopted ideas and this determines productivity. The production function for firm i has the Cobb–Douglas form and knowledge is Harrod neutral: Yit = (hEiAit)1−aXait.

(10.1)

Profits P can then be written as Pit = (hEiAit)1−aXait − wxXit,

(10.2)

where wx denotes the (real) price of the input. Profit maximization yields the following first-order conditions with respect to the use of inputs X: a (hEiAit)1−aXa−1 = wx. it

(10.3)

This gives a demand function for inputs: Xit = a1/(1−a)hEiAitwx−1/(1−a).

(10.4)

Market Forces and the Continent’s Growth Problem

325

Combining equations (10.2) and (10.4) gives Pit = Ωhwx−a/(1−a)EiAit,

(10.5)

where Ω = αα/(1−α) − α1/(1−α). Profits are increasing in business knowledge and the education of the manager and decreasing in the price of the inputs. When not producing, a manager spends his time exploring, learning, and adopting new ideas on how to produce more efficiently. There are Bj – Ai locally unexploited ideas that can potentially be adopted by managers, where Bj denotes the number of innovations transferred (adopted) and implemented by the leading local firm in a given country j. However, not all ideas can be adopted in any given period due to information frictions. The matching function (10.6) gives the number of successful adoptions of unexploited ideas. The efficiency of this matching process is captured by the parameter L: Ait − Ait−1 = Λ((1 − h)Ei)bGit1−b,

(10.6)

where Git = Bjt − Ait and 0 ≤ b ≤ 1. The efficiency is determined by such factors as access to information within other firms that is the extent to which these other firms can protect the competitive advantage they have gained from the successful adoption of foreign ideas. The appearance of education in equation (10.6) is in the spirit of Nelson and Phelps (1966). They proposed the idea that individuals gain the ability to learn through education. The ability to learn then determines the rate at which they—and their country—can adapt foreign technologies.1 In the Lucas framework only improvements in the level of human capital can cause growth, whereas in Nelson and Phelps, it is the stock of human capital that determines growth rates. While, equation (10.1) captures the idea that education is a factor of production, equation (10.6) is in the spirit of Nelson and Phelps; education helps managers adopt new ideas. Inserting (10.6) into (10.5) gives Pit = ΩhEi [Ait−1 + Λ (1 − h)bEibGit1−b]w−a/(1−a) . x

(10.7)

We can now address the problem of allocating time between the two tasks performed by managers within the firm, which are managing production and learning about new business ideas. The first-order condition for profit maximization with respect to h is Ph = ΩE [Ait−1 + Λ (1 − h)bEbi Git1−b − hΛ (1 − h)b−1EbGit1−b]wx−a/(1−a) = 0.

(10.8)

326

Gylfi Zoega

Taking logs gives the following expression defining hˆ , the fraction of a manager’s time spent producing:

ηˆ =

1 [log Ait−1 − log Λ − β log Ei − (1 − β ) log Git − log {(1 + β )ηˆ − 1}]. 1− β (10.9)

The fraction of time spent producing h depends on the level of knowledge A, the manager’s education E, and the size of the knowledge gap G. Taking the total differential of equation (10.9) gives partial derivatives of h with respect to the other variables in the equation. The fraction of time spent working h turns out to be increasing in the level of business knowledge A and decreasing in the level of education E, the gap G and the efficiency of the matching function L.2 The manager of a firm approaching the productivity frontier B will hence spend less time studying and more time managing production the closer he gets to the frontier, while an increase in the level of education will make him spend more time studying new ideas. 10.2.2 Adoption of Foreign Ideas New innovations are introduced through transfers to the leading firm—owned and operated by an entrepreneur, indexed by the letter j, with education Ej —from abroad as well as genuine innovations Ij. For simplicity, we assume that there is only one leading firm in each country, hence the index j can also be used for the countries. Productivity growth in the leading firm is described by the following equation *u Bjt − Bjt−1 = Λ[(1 − h)Ej]bGjt*1−b + ImjtBt−1 .

(10.10)

where Gj* = B* − Bj is the gap between the best domestic firm and best practice abroad, with B* denoting the number of successful business innovations in the world. The last term describes genuine innovations in country j, with Ij denoting the number of entrepreneurial ideas that are successful at getting finance and B* the world productivity frontier. The equation implies that new ideas have a greater impact on productivity B the larger is the stock of accumulated knowledge B* in the world. The parameter n ≥ 0 describes the strength of this effect.3 Finally the world frontier moves out when genuine innovations take place in different countries: Bt* − Bt*−1 = ∑ I μjt Bt*−ν1. j

(10.11)

Market Forces and the Continent’s Growth Problem

327

The entrepreneurial firm is also engaged in production, and equations (10.1) through (10.9) describe its decisions—with B now denoting productivity instead of A—when it comes to allocating time between producing and adopting ideas from abroad. What does a simple model of business innovation say about the gradual stagnation of the continental economies? Starting with a large gap in terms of business practices G at the end of the war, these countries could enjoy a rapid rate of growth of A even with a low level of education E and without spending too much time studying business ideas. As the gap diminished and productivity A improved, the optimal response was to increase h and spend less time studying foreign ideas and growth stagnated, both because the pool of unexploited ideas was becoming smaller and because the optimal time spent studying new ideas was falling. 10.2.3 Entrepreneurship Entrepreneurship takes place in leading firms in different countries. Local banks have the capacity to finance Fj entrepreneurial projects in the country of entrepreneur j and the potential number of such projects is related to the creativity of the entrepreneur Cj. In particular, there are Cj entrepreneurial projects or potential innovations. Each potential innovation consists of a genuinely novel business idea and hence embodies a distinct view of the relevant markets. The potential innovations differ along two dimensions. First, the probability of success differs between projects. Take the genuine innovations that take place in country j that we have denoted by Ij. These innovations can then be indexed by t so that t ∈ [0, Ij]. We let the variable b denote the probability of failure, and the expected probability of failure of project t is then denoted by bte. Second, the projects give entrepreneurs nonpecuniary benefits that also differ between projects.4 Some ideas are more fun to carry out than others. As a result entrepreneurs may be willing to go ahead with projects that offer a low expected monetary return. In particular, we let the variable u denote the nonpecuniary benefit so that ut denotes the benefit from idea t. Financing of the different innovations is contingent on the entrepreneur finding a like-minded banker when it comes to the expectations about the probability of success of individual projects since each idea requires, by assumption, one unit of output for its implementation. This is the “innovation market” described by Phelps (2006). The number of such matches is given by the following equation: M = ΓChj F1−h j ,

(10.12)

328

Gylfi Zoega

where G is a measure of the efficiency of the financial system. However, it is not sufficient to find a like-minded banker; the expected return from the idea has to cover the required rate of return, determined by the exogenous world rate of interest r*. The value of a successful project to the entrepreneur—that is one that does not fail—stems from its expected contributions to profits, which analogous to equation (10.2) can be written as 1−a α Pjt = 〈{hEj(Bjt−1 + Λ[(1 − h)Ej]bGjt*1−b + IjtmB*υ X t − wxXt. t )}〉

(10.13)

Solving for X and substituting back into (10.13) gives an equation that is analogous to (10.7): −α/(1−α) Pjt = ΩhEj[Bjt−1 + Λ(1 − h)bEbj Gjt*1−b + IjtmB*υ . t ]wx

(10.14)

From equation (10.14) it follows that the payoff to the entrepreneur from a successful innovation is measured by the derivative of *υ PI = mΩhEjIjtm−1Bt+1 wx−α/(1−α).

(10.15)

The total return from a marginal project t—if financed—can then be written as PI + ut. The interest paid by the entrepreneur is innovation specific, in particular the bank receives rt if it finances a project t. When the entrepreneur has found a like-minded banker who is willing to finance his project, the two have to decide on the terms of their transaction. The interest payment is, by assumption, determined such that the surplus from a successful match between an entrepreneur and a bank is split evenly: VtE = VBt ,

(10.16)

where VtE—how much the entrepreneur values the match—and VtB— how much the bank values it—are given by equations (10.17) and (10.18) below (1 + r*)VtE = (1 − bte)[PI + ut − rt],

(10.17)

(1 + r*)V Bt = (1 − bte )rt ,

(10.18)

where r* is the world rate of interest and be denotes the probability that the project fails.5 This gives the following solution for the interest charged: rτ =

1 [ PI + uτ ] . 2

(10.19)

Market Forces and the Continent’s Growth Problem

329

The number of projects financed is then determined by the condition 1 (1 − bτe ) ( PI + uτ ) ≥ 1 + r * . 2

(10.20)

This implies a lower bound on the sum of the pecuniary and the nonpecuniary benefits from a project to the entrepreneur: PI + uτ ≥ 2

1 + r* . 1 − bτe

(10.21)

Denote the fraction of all entrepreneurial projects that fall below this critical level by H(r*, Ej, wx, B*). It follows from (10.12) and (10.21) that the number of projects financed is Ij = [1 − H(r*, Ej, wx, B*)]ΓChj F1−h j .

(10.22)

The number of projects financed is increasing in the creativity of the entrepreneur Cj, increasing in the supply of loans by the banking system Fj, increasing in the efficiency of the matching process between banks and entrepreneurs G and, finally, increasing in the share of all entrepreneurial projects that offer pecuniary and nonpecuniary benefits above the required rate of return. From equations (10.15) and (10.21) it follows that this is increasing in the world frontier B*, decreasing in the cost of the input wx and the required rate of return r*, and increasing in the level of education E. 10.3

The Flames of Growth

The rate of productivity growth depends on a multitude of market and institutional variables. We can distinguish between domestic and world factors. Combining equations (10.6), (10.10), and (10.22) gives equation (10.23) below: Ait − Ait−1 = Λ[(1 − h)Ej]b{Bit−2 + Λ[(1 − h)Ej]b(B*t−1 − Bjt−1)1−b m u 1−b + [(1 − H(·))ΓChj F1−h j ] B*t−1 − Ait−1}

(10.23)

An improvement in the performance of domestic financial institutions—embodied in an increase in the value of the parameter G—will increase the number of matches between like-minded entrepreneurs and the suppliers of funds, which will raise the rate of growth of leading productivity. A positive domestic shock could take the form of an improvement in the expected profitability of innovations or higher nonpecuniary

330

Gylfi Zoega

benefits from embarking on new entrepreneurial projects and an improvement in the creativity of entrepreneurs, all of which raises the number of innovative business ideas that receive financing in the banking system. The effect of this change would then depend on the efficiency of financial institutions, embodied in the parameter G; only with financial institutions that are able to match entrepreneurs and like-minded financiers do these positive developments have an effect on growth. Education has a positive effect on productivity growth. A higher level of education raises the growth effects of all productivity improvements, whether through domestic adoption, adoption by the leading firm of foreign ideas, or entrepreneurship. Higher education will moreover facilitate learning from the best local firm, as well as learning by the leading firm from foreign firms. Finally education raises the expected profitability of new entrepreneurial projects and hence has the effect of raising the proportion of projects that receive financing from the banks. A rise in the price of inputs would reduce the expected profitability of new technologies and lower the rate of productivity growth. Various institutions affect the cost of labor, such as labor unions, employment protection legislation and the real price of oil. The effect of an increase in interest rates is somewhat more complicated. If world interest rates rise because of a fall in world savings, the consequence would be fewer entrepreneurial projects receiving financing. If, in contrast, the increase was caused by a rise in the world level of entrepreneurial activity, then the effect would be more complex; the world theoretical level of productivity B* would advance—increasing the rate of learning from abroad—but higher interest rates would mean that fewer new business ideas would be funded. 10.4

Empirical Evidence

In this section, data from Australia, Austria, Belgium, Canada, Denmark, Finland, France, Greece, Ireland, Italy, Japan, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, and United States will be analyzed with a view of detecting patterns that relate growth performance to different institutional variables.6 Productivity A is calculated as total factor productivity from a Cobb– Douglas production with capital and employed labor as factors of

Market Forces and the Continent’s Growth Problem

331

production. The capital stock series is calculated using the perpetual inventory method assuming a 6 percent depreciation rate. The value of the stock of capital in year 1949 is first calculated by assuming a steady state in a neoclassical growth model with depreciation 6 percent and a growth rate that equals the average rate of growth of output between 1950 and 1960. The capital stock series 1951 to 2000 is then calculated using investment data and the assumed depreciation rate. Finally the total factor productivity series are derived annually from 1960 to 2000 assuming that labor’s share of output is 0.7. Productivity growth is calculated as the proportional change in the average level of productivity between half-decades—from 1960–64 to 1965–69, and so on—with the last observation on growth rates being the rate of growth between the first half and the second half of the 1990s. The growth rates are shown in appendix A. Total factor productivity in the United States is then taken to be a proxy for the world technology frontier B* and the country frontier A*. In order to explain differences in productivity growth, several variables are explored. They include a measure of education levels, financial market variables, and labor market variables.7 Education is measured as the fraction of the population with some university education. The financial markets variables include deposits (commercial and savings) as a ratio to GDP, the number of listed companies per million inhabitants, and stock market capitalization as a fraction of GDP. The first enters through the supply of capital F in section 10.2—more capital implies that more projects will be financed. The number of listed companies and stock market capitalization are meant to proxy for capital market development. This could be expected to affect the efficiency of the matching process, captured by the parameter G in the model above. The labor market variable is a measure of employment protection and is intended to affect the cost of labor. There have been several attempts at explaining differences in the growth performance of OECD countries using education data. While Benhabib and Spiegel (1994) found a statistically significant effect of human capital on growth in a cross-country regression that included both developing and developed countries,8 Krueger and Lindahl (2001) found that the relationship ceases to be significant once we remove non–OECD countries from the sample. A recent paper by Vandenbussche, Aghion, and Meghir (2006) uses a pooled crosssectional, time-series analysis and finds a statistically significant relationship between the level of education and productivity growth for

332

Gylfi Zoega

Table 10.1 Education and technological progress Variables

(1)

(2)

(3)

log (A(−1)/A*(−1))

−8.18 (2.43)

−21.28 (1.63)

−12.69 (14.80)

E(−1)

−0.005 (0.08)

0.03 (0.29)

0.15 (3.45)

E·log (A(−1)/A*(−1))

0.28 (2.71)

−0.18 (0.87)

0.33 (2.05)

R-squared Observations

0.49 137

0.73 137

0.62 137

Notes: t-statistics in parentheses. Time dummies not reported. Column 1 has no fixed effects; column 2, country fixed effects; and column 3, group fixed effects. The right-had side variables are lagged one period.

the OECD countries by, first, measuring education by the proportion of the population with some university education and, second, interacting the education variable with a variable measuring the (log) difference between US productivity (the frontier) and each country’s productivity. Column 1 in table 10.1 replicates their results when fixed effects are not included using our data set. The estimated coefficients then become statistically insignificant when (country) fixed effects are included as shown in column 2. Vandenbussche et al. (2004) defined group fixed effects and reported a positive effect of tertiary education on growth that becomes stronger the closer a country gets to the productivity frontier (US productivity level). These results are confirmed in column 3. Note that the equation does suffer from a lack of robustness in that it does not survive the inclusion of country fixed effects.9 Experimenting with the other variables included in this study, we found that estimating the equation gave nonrobust results in many cases. One reason for this problem is that the equation has a stationary left-hand side variable, which is the rate of growth of total factor productivity, while the right-hand side has mostly nonstationary variables, such as the level of education, stock market capitalization, the size of deposits, and the number of listed companies. For this reason the regression results are not reported in this chapter. Instead, we focus on broader and more robust patterns in the data set. One robust feature of the data is the slowdown of productivity growth as a country approaches the productivity frontier (the level of

Market Forces and the Continent’s Growth Problem

333

US total factor productivity). When the variable log(A/A*) is included in the equation above it acquires a negative and statistically significant coefficient—such as in the first line of table 10.1—and this result is not sensitive to the inclusion or exclusion of other variables. However, looking at the country data, we find that we can group the countries so that the slowdown in productivity growth is seen in one group and not in the other. The first group has countries from the European continent—Austria, Belgium, France, Greece, Italy, the Netherlands, Portugal, and Spain—in addition to Japan. As shown in figure 10.1 below, the rate of growth of productivity falls as these countries approach the US productivity frontier. Note that productivity growth is measured as the proportional growth of total factor productivity between halfdecades, starting with the growth of productivity between the first and the second half of the 1960s and ending with the growth between the first and the second half of the 1990s. We call these countries the “bad performers” in that they were incapable of maintaining the same rate of growth as the pool of business ideas was gradually depleted—the productivity gap became smaller. Another group did much better in that the rate of productivity growth did not slow down when they approached the frontier. These countries include the Anglo-Saxon countries of Australia, Canada, Ireland, New Zealand, and United Kingdom; the Scandinavian countries of Denmark, Finland, Norway, and Sweden; and finally Switzerland. The relationship between their productivity growth rates and the productivity gap is shown in figure 10.2 below. We call these countries the “good performers.” Before taking a look at the factors that separate the two groups, we focus on two global variables that play a role in the proposed model. These are the real price of oil (rpoil)—which affects the cost of production and hence the expected profitability of innovations—and the world real rate of interest (rworld). Replacing education in table 10.1 with oil prices and interest rates gives the results shown in table 10.2. The only coefficient that is statistically significant is for the lagged productivity gap. However, as can be seen from appendix A, the rise in oil prices in the late 1970s, early 1980 and their fall in the late 1980s coincided with changes in the rate of productivity growth. In column 2 we see that the contemporaneous value of oil prices turns out to have a negative and a significant coefficient. However, changing oil prices can only be a part of the story because both groups of countries faced the same oil prices.

334

Gylfi Zoega

Austria

0.16

Belgium

Growth

0.12

0.14

0.16

0.12

0.14

0.10

0.12

0.08

0.08

0.10

0.06

0.08

0.04 0.00 0.64

0.68

0.72

0.76

0.80

0.06

0.04

0.04 0.55 0.60 0.65 0.70 0.75 0.80

0.02 0.60

A/A*

Growth

0.64

0.68

A/A*

Greece

Italy

0.72 0.76 A/A*

Japan

0.25

0.24

0.25

0.20

0.20

0.20

0.15

0.16

0.15

0.10

0.12

0.10

0.05

0.08

0.05

0.04

0.00 0.45 0.50 0.55 0.60 0.65 0.70

0.00 0.28

0.32

0.36

0.40

0.44

0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85

A/A*

A/A*

Netherlands

A/A*

Portugal

0.10

Spain

0.20

0.20

0.16

0.08

Growth

France

0.18

0.16

0.12 0.06

0.12 0.08

0.04

0.08

0.04

0.02 0.68

0.72

0.76

0.80 A/A*

Figure 10.1 Bad performers

0.00 0.40

0.44

0.48

0.52

0.56

A/A*

0.04 0.50

0.55

0.60

0.65

0.70

A/A*

Market Forces and the Continent’s Growth Problem

Growth

Australia

Canada

0.10

0.12

0.08

0.08

0.06

0.04

0.04

0.00

0.14

0.10

0.70

0.72

0.74 0.76 A/A*

–0.04 0.72

New Zealand

0.08 0.06 0.76

0.80

0.84 A/A*

0.52 0.56 0.60 0.64 0.68 0.72 0.76 0.80

A/A*

United Kingdom

0.08 0.04

Growth

Ireland 0.16

0.12

0.02 0.68

Switzerland

0.12

0.12

0.10

0.08

0.08

0.04

0.06

0.00

0.04

–0.04

0.02 0.62 0.64 0.66 0.68 0.70 0.72 A/A*

–0.08 0.60

0.00 –0.04 –0.08 0.68

0.72

0.76

0.80 0.84 A/A*

Denmark

Finland

0.64

0.68

0.72 0.76 A/A*

Norway

0.12

0.20

0.10

0.10

0.16

0.09

0.12

0.08

0.08

0.07

0.02

0.04

0.06

0.00 0.680 0.685 0.690 0.695 0.700

0.00 0.50

0.08 0.06 0.04

0.55

0.60

A/A*

0.65 A/A*

Sweden 0.16 0.12

Growth

Growth

335

0.08 0.04 0.00 0.63

Figure 10.2 Good performers

0.64

0.65

0.66 0.67 A/A*

0.05 0.44 0.46 0.48 0.50 0.52 0.54 A/A*

336

Gylfi Zoega

Table 10.2 Oil prices, interest rates, and technology growth Variables

(1)

(2)

Constant

−13.50 (5.39)

−2.07 (1.46)

log (A/A*)

−44.61 (10.32)

−32.17 (12.03)

rpoil(−1)

4.71 (2.13)

rpoil(−1)·log (A(−1)/A*(−1))

12.94 (3.66) −7.68 (6.36)

Rpoil rworld(−1)

0.63 (1.72)

rworld(−1)·log (A(−1)/A*(−1))

−0.48 (1.09)

R-squared

0.76

Observations

0.74

137

137

Notes: t-statistics in parentheses. The right-hand side variables in column 1 are lagged one period; oil prices in column 2 are not lagged.

Table 10.3 Education, financial market institutions, and employment protection in 1970 No slowdown

Country

University degrees

Stock market capitalization

Number of listed companies

Size of bank deposits

Employment protection

Australia

21.5

0.76

93.72

0.38

0.5

Canada

20.4

1.75

55.20

0.37

0.3

Denmark

15.5

0.17

52.14

0.25

0.98

Norway

7.4

0.23

37.9

0.49

1.55

Sweden

8.3

0.14

13.18

0.50

0.23

Switzerland

9.0

0.50

58.72

0.69

0.55

United Kingdom

7.9

1.63

47.22

0.22

0.21

Average

12.86

0.74

51.15

0.41

0.62

Notes: University degrees measures the fraction of population with some tertiary education; stock market capitalization is the ratio of the aggregate market value of equity of domestic companies to GDP; the number of listed companies is the number of domestic companies whose equity is publicly traded in a domestic stock exchange divided by the population in millions; size of bank deposits measures the ratio of commercial and savings deposits to GDP.

Market Forces and the Continent’s Growth Problem

337

Table 10.3 reports the value of our explanatory variables—tertiary education, financial market variables and employment protection for the two groups of countries. A stark difference emerges in that the good performers have a higher level of university education, larger stock market capitalization and a large number of listed companies. In addition the cost of labor should be lower because of less stringent employment protection. The only variable that does not differ much between the two groups is the size of bank deposits. These results fit well with the observation that the continental European economies benefited from being far behind in terms of productivity levels at the beginning of the postwar period, which made it possible to sustain high growth rates despite relatively low levels of university education, an underdeveloped stock market, and labor market rigidities. However, when they started to close in on the US productivity frontier and the pool of unexploited businesses ideas started to dry up, growth could not be sustained due to a combination of a low fraction of the population having entered university, capital market being not sufficiently developed, and the labor market infested with rigidities making labor a more expensive factor of production.

Growth slowdown

University degrees

Stock market capitalization

Number of listed companies

Size of bank deposits

Employment protection

2.6

0.09

12.05

0.31

0.65

Austria

5.2

0.23

38.39

0.40

1.24

Belgium

3.0

0.16

15.98

0.33

0.68

France

2.6

0.14

2.46

0.54

1.99

Italy

5.5

0.23

15.19

0.33

1.4

Japan

7.2

0.42

15.95

0.26

1.35

Netherlands

3.7

0.17

25.20

0.53

2

Spain

4.26

0.21

17.89

0.39

1.33

Average

338

10.5

Gylfi Zoega

Concluding Remarks

Business innovations play a fundamental role in economic growth. However, traditional models of endogenous growth emphasize technical innovations. The discussion of this chapter is an attempt to focus on the process of growth through business innovations and to study the role education and market forces play in this regard. Growth was shown to depend crucially on the ability of managers to study, understand, and adopt innovations already adopted by the local leader, as well as the ability of the local leader to learn from foreign business practices and the creativity of local entrepreneurs, and the ability of the local financial system to separate good business ideas from bad ones. The empirical results suggest that the continental European economies could sustain growth in the first decades following the war because of the large productivity gap that existed between them and the United States. However, their lack of dynamism became all too apparent as the pool of unexploited ideas was gradually depleted when they caught up with the United States. Appendix A: Total Factor Productivity Growth

3

6

7

8

5

5

0.16

0.02

0.04

0.06

0.08

0.10

0.02

1

2

3

5

1

1

1

2

2

2

6

4

5

6

3

3

5

4

5

Sweden

4

6

6

Netherlands

3

5

France

4

7

8

7

7

7

8

8

8

6

7

8

8

0.12

0.14

0.16

1

1

2

2

3

3

5

4

5

Ireland

4

6

6

7

7

8

8

0.16

0.20

0.24

0.00

4

5

6

7

7

8

8

0.12

–0.08

–0.04

0.00

3

4

5

6

0.02

0.04

0.06

0.08

Switzerland

3

0.04

2

2

–0.08

–0.04

0.00

0.04

0.08

0.10

1

1

Norway

0.06

0.08

0.12

0.05

0.06

0.07

0.08

0.09

0.10

0.00

1

1

2

2

4

5

6

3

4

5

6

United Kingdom

3

New Zealand

7

7

8

8

0.00

0.04

0.08

0.12

0.16

0.00

0.04

0.08

0.12

0.16

0.20

0.04

0.08

5

7

0.12

4

6

0.02

0.04

0.06

0.08

0.10

0.12

0.08

3

5

Greece

4

Canada

0.05

2

3

–0.04

0.00

0.04

0.08

0.12

0.10

1

2

Belgium

0.10

0.15

0.20

0.25

1

Figure 10A.1 Growth rates of total factor productivity over half-decades since 1960

4

0.00

8

8

8

0.04

7

7

7

0.04

6

6

6

0.08

Spain

4

Japan

4

0.08

3

3

3

0.12

2

2

2

0.04

0.06

0.08

0.10

0.12

0.14

0.12

1

1

1

Finland

5

0.16

0.20

0.00

0.05

0.10

0.15

0.20

0.25

0.00

0.04

0.08

0.12

0.16

0.20

4

0.04

0.00

0.02

2

0.08

0.04

0.04

1

0.12

0.08

0.20

0.06

Austria 0.16

0.16

0.12

Australia

0.08

0.10

1

1

1

1

2

2

2

2

5

5

4

5

Portugal

4

Italy

4

6

6

6

3

4

5

6

United States

3

3

3

Denmark

7

7

7

7

8

8

8

8

Market Forces and the Continent’s Growth Problem 339

340

Gylfi Zoega

Appendix B: The Data Table 10A.1 Productivity

TFP calculated using data on investment, labor force participation, and unemployment rate and assuming a factor share of 0.7 for labor

Penn World Tables

Schooling

Fraction of population with Some tertiary education

Barro and Lee (2000)

Employment protection

Index of employment protection.

OECD

Deposits

The ratio of commercial and savings deposits to GDP

Rajan and Zingales (2001)

Stock market capitalization

Number of listed companies

Rajan and Zingales (2001)

Number of listed companies per million people

The number of listed companies per million people is the number of domestic companies whose equity is publicly traded in a domestic stock exchange divided by the population in millions.

Rajan and Zingales (2001)

Oil prices

The real price of oil

Andrew Oswald

Real interest rates

The average real rate of interest in the G7 (GDP used as weights)

IMF and Penn World Tables

Notes 1. This contrasts with the later model of Lucas (1988) who emphasizes human capital accumulation as a source of growth. 2. For a solution we need h ≥ 1/(1 + β), which is also the condition necessary for the derivatives to have the signs described in the text. 3. See also Benhabib and Spiegel (1994). 4. See Hamilton (2000). 5. Dunne et al. (1988) used the Census of Manufacturers to calculate that on average 61.5 percent of firms disappear in their first five years and 79.6 percent in the first ten years. 6. Germany is excluded because of the effect of its unification in 1990 on average productivity. Data on financial market variables are available for Australia, Austria, Belgium, Canada, Denmark, France, Italy, Japan, the Netherlands, Norway, Spain, Sweden, Switzerland, the United Kingdom, and the United States. In addition productivity performance is explored for Finland, Greece, Ireland, New Zealand, and Portugal. 7. See appendix B for a description of variables and their sources.

Market Forces and the Continent’s Growth Problem

341

8. They did not find support for the hypothesis that changes in human capital cause growth. However, Temple (1999) finds support for the Lucas model when controlling for outliers. 9. Moreover the country grouping used for the group fixed effects is quite nonintuitive; see Vandenbussche et al. (2006). The groups are (1) Belgium, France, Italy, Netherlands; (2) Denmark, Finland, Norway, Sweden, Austria, United Kingdom, Switzerland; (3) Canada, United States; (4) Australia, New Zealand; (5) Portugal, Spain; (6) Greece; (7) Ireland.

References Barro, R. J., and J.-W. Lee. 2000. International data on educational attainment: Updates and implications. Working paper 7911. NBER, Cambridge, MA. Benhabib, J., and M. M. Spiegel. 1994. The role of human capital in economic development: Evidence from aggregate cross-country data. Journal of Monetary Economics 34: 143–73. Dunne, T., M. J. Roberts, and L. Samuelson. 1988. Patterns of firm entry and exit in U.S. manufacturing industries. RAND Journal of Economics 19 (4): 495–515. Hamilton, B. H. 2000. Does entrepreneurship pay? An empirical analysis of the returns to self-employment. Journal of Political Economy 108 (3): 604–31. Krueger, A. B., and M. Lindahl. 2001. Education for growth: Why and for whom? Journal of Economic Literature 39: 1101–36. Lucas, R. E. 1988. On the mechanics of economic development. Journal of Monetary Economics 22: 3–42. Nelson, R. R., and E. S. Phelps. 1966. Investment in humans, technological diffusion, and economics growth. American Economic Review 56 (2): 69–75. Phelps, E. S. 2009. Toward a model of innovation and performance along the lines of Knight, Keynes, Hayek, and M. Polanyí. In Entrepreneurship, Growth, and Public Policy. Cambridge: Cambridge University Press. Rajan, R., and L. Zingales. 2001.The great reversals: The politics of financial development in the 20th century. Working paper 8178. NBER, Cambridge, MA. Temple, J. 1999. A positive effect of human capital on growth. Economics Letters 65: 131–34. Vandenbussche, J., P. Aghion, and C. Meghir. 2004. Growth, distance to frontier and the composition of human capital. Journal of Economic Growth 11: 97–12

11

Controversies about Work, Leisure, and Welfare in Europe and the United States Robert J. Gordon

11.1

Introduction

As is widely known and further documented here, Europe (the fifteen EU members prior to the 2004 enlargement, hereafter the EU-15), has a lower standard of living than the United States as measured by PPPadjusted real GDP per person. This per capita real income shortfall, which has persisted for the last four decades, combines a deficit both in output per hour and in hours per capita. This chapter asks whether Europe is really as poor as is suggested by comparative data on real GDP per capita. How much can the low Europe-to-United States (EU/ US) ratio of real GDP per capita be supplemented by a careful analysis that adds to European welfare the value of leisure implied by shorter work hours, as well as subtracting components of US GDP that do not represent higher welfare. The position that Europe has simply chosen equal-valued leisure hours over work hours is most strongly proposed by Blanchard (2004: 4), who writes that “The main difference is that Europe has used some of the increase in productivity to increase leisure rather than income, while the United States has done the opposite.” An alternative interpretation is that the rise in the EU/US productivity ratio was artificial, as Europe made labor expensive through high labor taxes, high minimum wages, and tight labor and product market regulations. As a result firms were forced to slide northwest up their labor demand curves, retaining high-productivity workers while forcing low-productivity workers into unemployment or out of the labor force entirely. Under this interpretation the decline in hours per capita is largely involuntary and does not represent unmeasured welfare. The “unmeasured leisure” hypothesis has only a grain of truth and fails for three reasons. First, many of those “not-worked” hours

344

Robert J. Gordon

represent involuntary unemployment and involuntary low participation. Second, the counterpart of low European hours spent in market work is only partly leisure with a remaining sizable portion of home production. Third, to explain away the low level of European real GDP per capita, the entire gap of low hours per capita in Europe not only would have to take the form of voluntarily chosen leisure, but in addition that leisure would need to be valued not at the after-tax real wage but at before-tax real GDP per hour worked—which is a problem of valuation. In Europe the after-tax real wage is only a third of GDP, meaning that leisure has a low value even before we recognize that most of the reduction in European work hours was involuntary and represents a shift into home production rather than pure leisure. A general point to emerge from this analysis is that work-leisure decisions are made based on the after-tax wage while tax revenue depends on the before-tax wage that in Europe may be twice as high as the after-tax wage in countries with effective tax rates of 50 percent. This chapter extends the distinction between before-tax GDP per hour and the after-tax wage per hour. Each hour of reduced market work in Europe carries with it a “tax multiplier effect” by spreading the fixed costs of European government and welfare systems across fewer hours of work. A 25 percent reduction of work hours per capita implies a 33 percent increase in taxes per work hour in order to maintain a fixed level of government expenditure. When the reduction in work hours takes the form of early retirement the tax multiplier effect is increased by the need to increase government spending on pensions. Thus there is a mutual path of causation between tax rates and work hours, the tax rates to hours comparison emphasized by Prescott (2004) and the reverse causation of hours to tax rates suggested here. This chapter differs from most of the previous literature in this field by emphasizing the turnaround of labor market behavior in Europe after 1995. Until 1995 productivity in Europe converged toward the US level while hours per capita fell relative to the US level. After 1995 there was a simultaneous turnaround, with productivity falling relative to the US level while hours per capita began a substantial recovery. This chapter examines the recent literature on the sources of this post-1995 turnaround that simultaneously raised European hours per capita relative to the United States but also reduced its relative level of output per hour. The second aspect of the welfare comparison concerns not the interpretation of hours in Europe but rather output, the numerator of

Controversies about Work, Leisure, and Welfare

345

both the output per capita and output per hour ratios. Is the translation of output to welfare different in the United States than in Europe? The comparisons undertaken in this chapter are rarely discussed. The claim that US GDP is overstated for welfare comparisons begins with its harsh climate that requires higher expenditures on energy to achieve a given level of interior comfort. Another portion of US GDP goes to maintaining an enormous prison system that currently incarcerates two million Americans, mostly for minor drug offenses. A more controversial claim is that longstanding US policies have encouraged inefficient low densities of metropolitan areas, adding to traffic congestion, commuting times, and air pollution. A final element that is both the largest and perhaps easiest to quantify is the enormous waste of resources involved in the dysfunctional American medical care system, amounting to perhaps 3 percent of US GDP. The chapter includes a speculative section that addresses the distain of Europeans for the insecurity of the American way of life. Rather than place an explicit value on greater security in Europe than in America, it is assumed that high taxes in Europe buy something of value, that is the social welfare system, and accordingly this chapter’s point of departure for all comparisons is before-tax income per capita rather than after-tax income per capita. 11.2 The Evidence: Productivity Almost Converges but Per Capita Income Does Not Examined in this section are the basic data on output per capita and output per hour.1 The data on both labor productivity (Y/H) and real GDP per capita (Y/N) come from the Groningen data bank, which provides cross-country comparisons with two base years and the weighting schemes.2 An averaging of these two data sources shows that Europe’s (EU-15) productivity level by 1995 had reached 91.7 percent of that of the United States and then by 2008 fell back to 83.0 percent. Three European countries exceeded the US level for a few years during the 1990s: France was at 103.6 percent in 1995 and 95.2 percent in 2008, Belgium was at 108.1 percent in 1995 and 96.4 percent in 2008, the Netherlands was at 105.0 percent in 1995 and fell back to 97.1 percent in 2008. However, none of these countries came close to catching up with the level of US real GDP per capita. In the 1995 to 2008 period, when the

346

Robert J. Gordon

same two data sources were averaged, France achieved no better than 76.1 percent of the US level in 1995, and this ratio fell back to 73.1 percent by 2008. Thus the Y/N ratio for France was 30 points below its Y/H ratio in 1995 and 20 points in 2008. By definition, this discrepancy is due to a precipitous decline in hours per capita in France relative to the United States over the past four decades. For the EU-15 the same discrepancy was 10 percentage points in 2008, with a productivity ratio to the United States of 83 percent and an income per capita ratio of 71 percent. Figure 11.1 provides a display of the percentage per capita income (Y/N) and productivity (Y/H) ratios of Europe to the United States and reveals a dramatic contrast in the timing and magnitude of changes of these ratios. To allow for the initial stage of rapid postwar reconstruction in Europe, each of the graphs of figure 11.1 begins in 1960 rather than 1950. The productivity ratio rises steeply until 1995, holds at a plateau near 90 percent until 2000, and then enters into a decline during 2001 to 2008. In contrast, the per capita income ratio first reaches 70 percent in 1973 and then fluctuates in a narrow range around 70 percent. The 1982 peak in this ratio is artificial, as it reflects the US recession of that year rather than progress for Europe. 110 100 Output per hour

Percent

90 80 Output per capita 70 60 50 40 1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

Figure 11.1 Ratio of Europe-15 to the United States, output per capita and output per hour, 1960 to 2008

Controversies about Work, Leisure, and Welfare

347

11.2.1 Decomposition of the Decline in Europe/US Hours per Capita By definition, real output (Y), population (N), hours of work (H), and employment (E), are related as Y/N ≡ Y/H * H/E * E/N,

(11.1)

which states that output per capita equals labor productivity times annual hours per employee, times employment per capita. Equation (11.1) will be used to further subdivide changes in the E/N ratio into its two components, the employment rate (E/L) and the labor force participation rate (L/N). E/N ≡ E/L * L/N = (1 + U/L) * L/N,

(11.2)

where U/L is the unemployment rate. In figure 11.2 the dashed gray line is the ratio of the two lines in figure 11.1, namely the EU/US ratio of output per capita divided by the EU/US ratio of output per hour. By definition, the dashed gray line equals hours per capita and is labeled as such in figure 11.2. This shows a decline from almost 126 percent in 1960 to 108 percent in 1970 to 77 percent in 1995 and then a substantial recovery to 85 percent in 2008. 130 120 110

Percent

Employee to population ratio

100 Hours per employee

90 80 Hours per capita

70 60 1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

Figure 11.2 Ratio of Europe-15 to the United States, hours per capita, hours per employee, and employees per capita, 1960 to 2008

348

Robert J. Gordon

By definition, any changes in hours per capita (H/N) must be explained by changes in the same direction in the product of the hours/ employee and employment/population ratios, as shown in figure 11.2 by the solid gray and solid black lines, respectively. An important finding is that the decline in the hours per capita ratio has been explained more by the decline in the employee to population ratio than by the hours to employee ratio. Thus Blanchard’s (2004) overly facile explanation can be rejected, as quoted above, that the differential behavior of European productivity to European per capita income is simply a matter of the voluntary choice of shorter hours. Also there are two interesting aspects of timing to note here that may help distinguish alternative hypotheses. First, much of the decline in the employee to population ratio had already occurred by 1970, whereas the decline in the ratio for hours per employee was more gradual. Second, there was a distinct turnaround in the employee to population ratio after 1995 but not in the hours to employee ratio. The time-series plots of the five ratios in figures 11.1 and 11.2 are summarized in table 11.1, which gives both the levels and growth rates for 1960, 1970, 1995, and 2008. Because the intervals are of different lengths, the focus here is on the growth rates shown in the bottom three lines of table 11.1. Column 1 shows that the European catch-up to the American level of real GDP per capita halted after 1970, with a growth rate of a mere 0.1 percent since then. In contrast, the European catch-up to the US productivity level shows a rapid growth rate of 2.7 percent Table 11.1 Levels and growth rates of output and labor utilization: Ratio of Europe-15 to United States, 1960 to 2008 Output per capita

Output per hour

Hours per capita

Hours per employee

Employees per capita

Levels 1960

60.1

48.6

126.0

109.8

114.8

1970

68.1

63.9

108.0

103.6

104.2

1995

70.4

91.7

77.3

89.6

86.3

2008

71.2

83.0

85.2

88.9

95.9

Annual growth rates 1960–1970

1.2

2.7

−1.5

−0.6

−1.0

1970–1995

0.1

1.4

−1.3

−0.6

−0.8

1995–2008

0.1

−0.8

0.8

−0.1

0.8

Controversies about Work, Leisure, and Welfare

349

per annum for 1960 to 1970, followed by a decline in the convergence growth rate by half in 1970 to 1995 to 1.4 percent per annum, and then a turnaround to a declining relative growth rate in 1995 to 2008. Column 3 shows a steady decrease in hours per capita at an annual rate of −1.5 percent for 1960 to 1970 and −1.3 percent for 1970 to 1995, followed by a turnaround to a positive growth rate of 0.8 percent for 1995 to 2008. The mirror-image behavior after 1970 noted in columns 2 and 3 has elicited interest in the idea of a trade-off between hours and productivity, as will be explored further below. As shown in column 4, hours per employee also declined relatively steadily from 1960 to 1995—with 1960 to 1970 and 1970 to 1995 growth rates of −0.6 percent—and followed by a much slower rate of decline of −0.1 percent per annum after 1995. The “residual,” employment per capita, declined steadily in 1960 to 1970 and 1970 to 1995 at respective growth rates of −1.0 and −0.8 percent, followed by a sharp turnaround after 1995 to +0.8 percent. This turnaround in the hours and employment per capita may be helpful in assessing alternative hypotheses to explain Europe’s low hours per capita. 11.2.2 The Time Series of Hours per Employee and the Employment/Population Ratio The EU/US ratios corresponding to equation (11.1) are quite revealing in the magnitude and timing of the changes. However, we can gain additional insights by looking at the raw numbers for Europe and the United States separately. As shown in figure 11.3, hours per employee in 1960 were higher in Europe, 2,148 hours per year compared to 1,956 hours in the United States. From 1960 to 1975 hours in Europe declined slightly faster than in the United States, in 1975 reaching 1,850 for Europe and 1,826 for the United States. After 1975 there was a sharp divergence, so that by 2008 hours in the United States had barely declined, from 1,826 to 1,775, whereas the decline in Europe was much more significant, from 1,826 to 1,578. Those like Prescott (2004) who attribute the entire decline in European hours to higher taxes must show that tax rates in Europe steadily increased during 1975 to 2008 at a pace corresponding with the decline in hours per employee. Even more interesting are the results shown separately in figure 11.4 for the E/N ratio in Europe and the United States. In the United States the sharp increase in this ratio occurred between 1965 and 1985 with the entry of females into the labor force. Over the period plotted in

350

Robert J. Gordon

2,200 2,100

Hours per year

2,000 1,900 United States 1,800 1,700 Europe - 15 1,600 1,500 1,400 1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

2000

2005

Figure 11.3 Hours per employee, Europe-15 and United States, 1960 to 2008

55

50 United States

Percent

45

40

Europe - 15

35

30 1960

1965

1970

1975

1980

1985

1990

1995

Figure 11.4 Employment-to-population ratio, Europe-15 and United States, 1960 to 2008

Controversies about Work, Leisure, and Welfare

351

figure 11.4, the US ratio increased from 37.9 percent in 1960 to 48.4 percent in 1990 and then flattened out to 47.8 percent in 2008. In contrast, the European ratio fell from 43.5 percent in 1960 to 40.8 percent in 1983, and this was followed by a small recovery to 42.9 percent in 1991 and then a substantial revival to 45.9 percent in 2008. Why did the entry of females into the labor force in Europe not generate the same rise in the E/N ratio in Europe as in the United States in the 1965 to 1985 period? The explanation may be, in part, the sharp increase in European unemployment that appeared over the same time interval. The increase in the unemployment rate reduced the ratio of employment to the labor force apparently by enough to offset the role of females who would have been expected to increase the labor force participation rate. Another explanation may be that the trend to earlier retirement age pushed down the labor force participation rate by enough to offset the increase in the female labor force participation rate. It is worth noting that fertility rates in the United States are substantially higher than in Europe, which suggests that more European women have time free from raising children and so would be expected to have higher labor force participation than in the United States. 11.3

Interpreting Changes in Hours per Capita

Thus far we have examined time-series changes in the key components of hours per capita in Europe compared to the United States. The pattern of changes over time may be more consistent with some types of explanations than others, helping us discriminate among them. Age is another dimension that may help with this discrimination; for instance, an explanation for falling hours per capita in Europe based on higher labor taxes would consider the impact on workers of all ages up to retirement age rather than disproportionately one age group or another. 11.3.1 The Age Distribution of Unemployment and Labor Force Participation Unemployment rates by five-year age groups are shown for the EU-15 and United States in figure 11.5. The data refer to the year 2007, chosen deliberately to represent a relative prosperous period prior to the 2008 to 2009 recession. The European unemployment rate is uniformly higher across all age groups except for the age group 65 and above. These differences can be assessed using absolute or relative differences.

352

Robert J. Gordon

25

20

Percent

15

10 Europe - 15 5 United States 0 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74

Figure 11.5 Unemployment rates by age group, Europe-15 and United States, 2007

For teenagers the European rate is 19.1 percent compared to 15.7 for the United States, an absolute gap of 3.4 points and a relative gap that is 22 percent of the US rate. The lowest absolute gap is for age group 45 to 49, where the European rate is 5.1 and the US rate is 3.3, for an absolute gap of 1.8 points and a relative gap of 55 percent. Because the absolute and relative differences occur for all age groups (except 65+), this evidence would seem to support a single explanation such as high labor taxes. However, as shown in figure 11.6 the behavior of the labor force participation rate (LFPR) is different. For the four prime age groups from 30 to 49, the rates in Europe are slightly higher than in the United States, 86.0 compared to 83.7 percent, for a difference of 2.3 percentage points. The big differences that drag down the overall LFPR for the EU-15 are for the young and particularly for the older age groups. The absolute shortfall for Europe is 11.7 points for ages 15 to 19, 7.6 points for ages 20 to 24, 8.3 points for ages 55 to 59, and a huge 21.0 points for ages 60 to 64 and 65 to 69. These differences are not compatible with Prescott’s (2004) labor tax explanation but are compatible with the Alesina et al. (2006) emphasis on the political process that put pressure on pension schemes to encourage early retirement. The low participation for the older groups in Europe may also be

Controversies about Work, Leisure, and Welfare

353

90 80 United States 70 60

Percent

Europe - 15 50 40 30 20 10 0 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74

Figure 11.6 Labor force participation rates by age group, Europe-15 and United States, 2007

compatible with the Ljundqvist and Sargent (2006) claim that European social welfare policies have a stronger effect than labor tax rates on retirement age. By definition, the employment rate (E/L) times the LFPR (L/N) equals the employment to population ratio (E/N), for which we have already examined time-series changes in figure 11.2 and table 11.1. Figure 11.7 shows the E/N ratio by age group, as combined with the age pattern of unemployment in figure 11.5 and the age pattern of the LFPR in figure 11.6. The overall pattern for the E/N ratio in figure 11.7 is similar to that for the LFPR in figure 11.6, with parity for the prime age groups and larger absolute and relative differences for the youngest and oldest age groups. The aggregation of the group-specific unemployment rate and the LFPR depends on the relative size of each group. As shown in figure 11.8, Europe’s population structure is more heavily weighted to the older age groups, as would be expected with lower fertility, higher life expectancy, and a smaller flow of immigration. All the European age groups, starting with ages 60 to 64, have a higher weight than in the United States, and all younger age groups have a higher weight in the United States between ages 15 and 29. When population weights are applied to the LFPR data plotted in figure 11.6, the overall LFPR of 70.7

354

Robert J. Gordon

90 80 70 United States 60

Percent

Europe - 15 50 40 30 20 10 0 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74

Figure 11.7 Employment-to-population ratios by age group, Europe-15 and United States, 2007

12

10

8

Percent

Europe - 15 6 United States 4

2

0 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74

Figure 11.8 Share of population aged 15 to 74 by age group, Europe-15 and United States, 2007

Controversies about Work, Leisure, and Welfare

355

in the United States exceeds the 64.4 rate in the EU-15 by 6.4 percentage points. The respective E/N ratios are 67.5, 59.9, and 7.6 percentage points.3 11.3.2 Summary of Findings on Changes in Hours per Capita Changes in hours per capita in Europe compared to the United States can be divided into two categories: changes in hours per employee (H/E) and changes in employment per capita (E/N). The latter can be further subdivided into changes in the employment rate (E/L) and changes in the labor force participation rate (L/N). We learned from figure 11.2 and table 11.1 that the post-1960 period can be divided into two distinct phases split at 1995. Between 1960 and 1995, fully twothirds of the decline in hours per capita was accounted for by the employment ratio (E/N) and only one-third by hours per employee (H/E). This provides useful dose of skepticism for Blanchard’s previously cited view that Europeans used their high productivity to purchase more leisure; leisure in the form of shorter hours per employee were only one-third of the story through 1995. The two-thirds of the decline in hours per capita consisting of higher unemployment and lower labor force participation is not an outcome of voluntary choice. An additional dimension of evidence in table 11.1 is that the EU/US ratio for employment per capita (E/N) turned around after 1995 while the hours per employee ratio, while declining more slowly, did not turn around. All this suggests that a different set of factors may have been driving changes in the hours per employee ratio from the employment per capita ratio. While we do not have graphs on the time-series behavior of the split of the employment ratio between the unemployment rate and the labor force participation rate, we can calculate the importance of each of these components for a single year, 2002. Using US population weights as in figure 11.10 to aggregate across age groups, with EU unemployment and LFPR’s the EU/US employment ratio (E/N) would have been 86.2 percent. Continuing with US population weights, with US agespecific unemployment rates that E/N ratio would have risen to 90.8 percent, and obviously to 100.0 percent with US age-specific unemployment rates and labor-force participation rates. Thus we conclude that in 2002, of the gap of 13.8 percent between the European and US E/N ratio, less than one-third (4.4/13.8) is explained by higher European age-specific unemployment rates and more than two-thirds (9.2/13.8) by lower European age-specific labor force participation rates.

356

11.4

Robert J. Gordon

Alternative Hypotheses to Explain Declining Hours per Capita

In recent journal and conference discussions most of the attention has focused on single-cause explanations of the secular decline in hours per capita in Europe, such as Prescott’s labor taxes or Alesina’s politically powerful unions. However, my examination of the data suggests that more nuanced multiple causes might provide a better explanation, including the post-1995 turnaround in the EU/US employment per capita (E/N) ratio and the sharp differences in the EU/US ratios of labor force participation by age group. Among the alternative hypotheses should also be the welfare implications of extra hours per year spent by Europeans in nonmarket work instead of market work. Conventional economic analysis values leisure at the marginal after-tax wage. If a single cause like higher labor taxes causes a substitution from work to leisure, the value of the extra leisure consumed would be measured by the area under the labor supply curve in a diagram like figure 11.9. Since Prescott’s approach to the analysis of labor taxes assumes that there is no income effect, because tax revenues are rebated to the population through government expenditures and transfers, the effect of taxes is to create a pure substitution effect. Presumably, if we imagine an upward7 6

High-cost labor supply curve

Labor demand curve

5

Real wage

4 (W/P)0

(W/P)1

A

3

Low-cost labor supply curve

B

2 1 0 Downward shift in labor supply curve reduces real wage and productivity

–1 –2 1

2

3

4

5

6

N0

N1

Labor input Figure 11.9 Labor demand and supply

7

8

9

10

11

Controversies about Work, Leisure, and Welfare

357

sloping labor supply curve extending between the 2004 European H/E annual total in of 1,550 hours and the US ratio of 1,811 hours, then the average value of the extra leisure in Europe would be halfway between the marginal after-tax wage that Europeans receive today and the higher marginal after-tax wage that Europeans would receive in a hypothetical world in which taxes are levied at American rates. A basic question, of course, is whether this valuation of leisure should be applied to the entire reduction in hours per capita that includes the effects of higher unemployment and lower labor force participation rates, or only to the one-third of the drop in European hours per capita consisting of lower hours per employee, meaning vacations and shorter work weeks. 11.4.1 Prescott on Labor Taxes Prescott (2004) claims that the entire difference between Europe and the United States not just for hours per employee but for hours per capita can be explained by higher tax rates on labor. The key to this demonstration, as explained by Alesina et al. (2006: 13), is that Prescott chooses a functional form that delivers a very high elasticity of labor supply, that is, a response of around −0.8 in logs to 1/(1 − t), where t is the tax rate on labor income. Alesina and coauthors show that the data require an even higher elasticity of −0.92, which is the ratio of the −29.7 percent log difference between European and American hours, divided by the 32.4 percent log difference in the marginal tax rate expressed as 1/(1 − t). They reject the Prescott assumptions after reviewing the micro labor supply literature that shows uncompensated labor supply elasticities for men that are close to zero. While labor supply elasticities for married women are high enough so that European tax rates could explain the entire EU/US difference for women, averaging the zero response for men and the large response for women leads the authors to conclude that tax rate differences can explain at best half of the hours per capita difference. A further weakness in the Prescott argument comes from the times series evidence. Most of the increase in tax rates occurred between the 1960s and mid-1980s, whereas the decline in hours continued at least through 1995. As we have noted, after 1995 the decline in hours per employee continued at a slower rate whereas the decline in employees per capita turned around into an increase. A final problem is that high tax rates may be standing as a proxy for a whole range of variables that differ between Europe and the United States but are not included

358

Robert J. Gordon

explicitly in cross-country correlations between tax rates and hours per capita, namely “generous welfare systems, workplace regulations, unemployment compensation programs, powerful unions, generous social security systems” (Alesina et al. 2006). Subsequently we report on regression analysis that joins together with taxes other sources of the pre-1995 decline in European hours or employment per capita (H/N and E/N), namely labor and product market regulation, generosity of unemployment insurance, and union density. 11.4.2 The Welfare State Some critics, in particular, Ljungqvist and Sargent (2006), criticize Prescott’s assumption that labor taxes are entirely redistributed to households as lump-sum transfers that are valued as if they were privately purchased goods and services. It is this device that allows Prescott to ignore income effects, and in turn to overstate the portion of changing work hours attributable to changing tax rates. These authors also criticize Prescott for ignoring the fact that in the early 1970s tax rates in France and Germany were already ten points higher than in the United States but hours per capita were basically the same, as shown above in figure 11.2. Ljungqvist and Sargent (2006: 43–44) emphasize the different welfare implications of the “national family perspective” implicit for Prescott, in which the entire population is viewed as a set of representative agents. When higher taxes reduce labor force participation, there are voluntary transfers between working and nonworking members of the “national family.” In reality, however, most nonemployed heads of households in Europe are not supported by voluntary intra–family transfers but rather by welfare systems that not only support reduced hours per capita but also “strain social insurance systems and government finances.” These authors argue that reforming European welfare systems would raise hours per capita more than cutting labor tax rates. They support their view in part by pointing to the fact that Europeans worked as much as Americans in the early 1970s despite higher labor tax rates, because Prescott’s hypothetical costless lump-sum redistribution within the national family was not in fact available. “Tax revenues were funneled to public goods and government expenditures that were poor substitutes for private consumption. The negative income effect of taxation worked in favor of sustaining high employment in the European welfare states” (Ljungqvist and Sargent 2006: 45).

Controversies about Work, Leisure, and Welfare

359

An additional consequence of generous welfare benefits is to encourage workers to remain unemployed for long periods of time after negative demand or productivity shocks. With heterogeneous workers who have previously accumulated skills, there will be a loss of those skills over prolonged spells of unemployment. The skill set of workers will no longer be high enough to warrant their high reservation wage, and they “become discouraged and are likely to fall into long-term unemployment or end up in other government programs, such as disability insurance and early retirement” (Ljungqvist 2006: 75). Figure 11.10 illustrates the sharp contrast between the EU-15 and the United States in the duration of unemployment over the interval 1975 to 2008. The initial decade of 1975 to 1985 witnessed a sharp decline in EU short-term unemployment (less than three months) and a mirrorimage increase in EU long-term unemployment (greater than one year). During the subsequent period from 1985 to 2008, an average of 69.9 percent of the US unemployed had durations of less than three months compared to only 19.7 percent in the EU-15. The corollary was that 47.0 percent of the EU unemployed had durations greater than one year, compared to only 9.0 percent in the United States. Thus the European Union had only about one-quarter the incidence of short-term unem90 80

US < 3 months

70

Percent

60 50 EU > 1 year

40 30 EU < 3 months

20 10

US > 1 year

0 1975

1980

1985

1990

1995

2000

2005

Figure 11.10 Percentage of unemployed population by duration of unemployment, 1975 to 2008

360

Robert J. Gordon

ployment and five times the incidence of long-term unemployment. My subsequent evaluation of the welfare benefits of low European hours per capita will return to these stark differences in the type of unemployment. 11.4.3 Unionization and Regulation Alesina and coauthors make much of the higher penetration of unions in Europe than in the United States. As is well known, in the United States the unions had a negligible role prior to the 1930s. Unions were legitimized by New Deal legislation, reached their peak of influence in the 1940s and 1950s, and began to decline in importance from the late 1960s. Some authors, including Goldin and Margo (1992), have stressed the role of unions in helping achieve the “great moderation” of income inequality in the 1940 to 1970 era. More recently Gordon and DewBecker (2007), among others, have emphasized the role of the reversal of union penetration to help explain the downward pressure on wages in the bottom 50 percent of the income distribution and the corollary that the share of the top 10 percent has increased substantially relative to the bottom 50 percent. In contrast, “union strength reached a peak in most European countries in the late 1970s and ear 1980s” (Alesina et al. 2006: 29). These authors trace two channels between high unionization and lower hours. First, unions keep wages artificially high and thus restrict employment, and in this sense labor demand is just like a labor tax. Second, unions may pursue a political agenda to reduce work hours per employee in order to force firms to hire more unionized workers to achieve the assumed fixed total of aggregate work hours. They derive several propositions from a simple model (1) that regulations limiting work hours will decrease productivity per worker but will raise productivity per hour, (2) that total hours worked under unionization will be lower and productivity per hour will be higher, and (3) that unions impede the reallocation of labor in response to sectoral shocks and can cause a decrease in overall hours worked, in comparison to an increase in hours worked in response to sectoral shocks in a competitive economy. The authors support their emphasis on unions by demonstrating a negative correlation between union coverage and hours of work that they claim is at least as high as between marginal tax rates and hours of work.4 However, this kind of cross-sectional evidence is fragile, both because of the large size of the outliers and because there is no attempt to model the time-series properties of unionization against the pattern

Controversies about Work, Leisure, and Welfare

361

of European hours per capita. Neither the Alesina et al. (2006) paper nor the comments by its discussants recognize the sharp turnaround in the EU/US employment population ratio after 1995. This turnaround provides an opportunity to weigh the explanatory power of alternative hypotheses. Overall, the authors’ analysis provides little insight beyond the general idea that unions push the labor market northwest along the labor demand curve, thus reducing hours per capita and raising the marginal and average product of labor. Alesina et al. (2006) go beyond a reliance on union density to provide numerous examples in individual European countries in which unions promoted policies like “work less, work all,” which reflected the belief that an enforcement of regulations that reduced work hours would create more jobs. Since this political pressure required that wages per job remain fixed, it forced upward the wage per hour and pushed hours per capita lower than otherwise. Examples are given for France, Germany, and Italy of union political involvement not only in shortening work hours without pay reductions, but also in “promoting and defending the welfare state in general and public pension systems in particular” (Alesina and Glaeser 2004). They cite the role of unions in the push for early retirement and in negotiating early retirement schemes for older workers in cases where the closing of a large plant might otherwise cause unemployment. Alesina and Glaeser attribute the concern of unions with early retirement to the political power of older workers within the union hierarchy. Recall from figure 11.6 that the most important single source in Europe’s reduction in hours per capita relative to the United States is early retirement, as shown by the age-specific labor force participation rates in the figure. Thus the key differences espoused by the leading authors is that Blanchard implicitly assumes that early retirement has been voluntary, Prescott assumes that early retirement is an endogenous response to high labor taxes, and Alesina et al. regard early retirement as the outcome of a political process led by unions who were involved in a political philosophy of work sharing regardless of whether workers actually want to stop working and live off pension income. Illustrated below is an example of the enormous cost to any society of early retirement. 11.4.4 Empirical Evidence on the Tax Hypothesis Prescott’s focus on taxation as the only cause of low European hours per capita relies on a model with parameters that are assumed rather

362

Robert J. Gordon

than estimated. A more convincing demonstration of the importance of tax effects is provided by Davis and Henrekson (hereafter D-H 2004: 37–38) who find a multidimensional tax impact: Higher tax rates on labor income and consumption expenditures lead to less work time in the market sector, more work time in the household sector, a bigger underground economy, and smaller value added and employment shares in industries that rely heavily on low wage, low skill labor inputs.

The D-H empirical work extends only to 1995 and thus has nothing to say about the post-1995 turnaround in European hours per capita. Regressions are run across countries for four years (1977, 1983, 1990, 1995) in which the dependent variable is alternatively the H/N, H/E, or E/N ratio, and the explanatory value is the sum of the employer labor tax rate, employee income tax rate, and indirect tax rate on consumption. In contrast to Prescott’s assumed elasticity of −0.92 on the H/N ratio (as discussed above), the D-H estimated elasticity for 1995 is −0.47.5 Thus the D-H paper arrives at the same conclusion as Alesina et al. reached by a different route, that Prescott overstates the tax effect by a factor of roughly two. Subsequently we will summarize the results of Basannini and Duval (2006) and Dew-Becker and Gordon (2008), which use richer and updated data sets going through 2003 to explain the pre-1995 decline in E/N and the post-1995 recovery of European E/N. An important point discussed by D-H relates to the emphasis by Ljundqvist and Sargent (2006) on effects of the welfare state. D-H admit that their tax elasticities do not reflect simply the impact of taxes; they reflect also the disincentive effects of the welfare state on hours per capita, due to the fact that taxes are used primarily to finance the welfare state so that countries with high taxes also have high levels of welfare support. Finally D-H cite the work of Olovsson (2004) showing that higher taxes shift work from the marketplace to home production, and this leads to large reductions in market work time with much smaller reductions in total work time as the time devoted to home production increases. Indeed the empirical results of Freeman and Schettkat (2005) cited below show that home production in Europe more than offsets the low level of market work, so leisure is lower in Europe than in the United States for both men and women. Another study of tax effects by Warren and Worthington (2004) is limited only to changes in hours per employee, not hours per capita, and thus misses the bulk of the reduction in European H/N, which takes

Controversies about Work, Leisure, and Welfare

363

the form of lower E/N. The results are hard to interpret because the authors include both an income tax variable and a separate “tax wedge” variable that includes the income tax. Nevertheless, a rough guess from the author’s reported results is that the elasticity of hours per employee to an increase in both the income tax and in the tax wedge is roughly −0.25, and this is consistent with a further −0.25 effect coming from the tax effect on the E/N ratio. The best recent work on the response of E/N to the tax wedge, with data that extends to 2003, is by Bassinini and Duval (2006). They include numerous additional control variables, among these some unique variables that control for the costs of employment by women. In DewBecker and Gordon (2008) we replicated and extended their results by confining the sample to the EU-15 (they include the United States, Canada, and Japan) and dropping some of their particular specification choices. Fortunately, our tax wedge coefficients are very close to theirs, −0.37 for us compared to their −0.30 for males, and a tax wedge coefficient of −0.4 compared to their −0.5 for females. Overall, the large literature on tax effects seems to be converging to a significant impact with an elasticity of about −0.4, less than half of Prescott’s conjectured elasticity. 11.4.5 Further Evaluation One line of criticism of the Alesina emphasis on unions is that the timing is wrong. As shown by Rogerson (2006: 83), union density averaged over 19 European countries rose through the late 1970s and fell until 1995, reaching a level that was little different than the starting value of 1960. If unions became strong and then became weak, why was their political influence still strong enough to explain low European hours per capita in 2004? Rogerson supports his skepticism by showing that union density and a measure of employment protection have very little explanatory power for changes in European hours per capita. This criticism falls into the trap of simple correlation and ignores inertia in the political process. It is possible that Europe could still be suffering from legislation that unions successfully pushed when they were strong in the 1980s but which opposing political forces have thus far been unable to overturn. The demonstrations in Paris in October 2010 against modest reforms in retirement ages suggest the power of such political inertia. None of the explanations reviewed from the recent literature has any explanation of the post-1995 reversal in the ratio of the EU/US employ-

364

Robert J. Gordon

ment to population (E/N) ratio. Most observers are startled to find that employment has grown faster relative to population in Europe than in the United States, where hours of work per capita in 2007 were still 4.3 percent their peak level in 2000 while the corresponding EU-15 figure increased from 2000 to 2007 by 2.3 percent. There is a chicken-and-egg aspect to this phenomenon of growing work hours in Europe and shrinking work hours in the United States since 2000. Is the phenomenon to be explained an autonomous shift in the incentives for work hours in Europe compared to the United States, as is implicitly assumed by most of the literature reviewed above, or is the behavior of work hours a by-product of differences in productivity growth in Europe compared to the United States that emerge from a totally different set of factors? Simple single-cause explanations of falling hours in Europe, such as “higher taxes,” “welfare state,” and “unions” appear to have missed completely the post-1995 turnaround and the related chickenegg question. Dew-Becker and Gordon (2008) are among the first to provide a quantitative explanation of the post-1995 turnaround of E/N growth in Europe. They use their regressions that explain annual changes in E/N across the fifteen EU countries over 1978 to 2003 and create counterfactual simulations that assume no changes in the explanatory variables after 1995. They attempt to explain an increase of 8.2 percent (in logs) of E/N in the EU-15 between 1995 and 2003. Their regression equations, following Basannini–Duval, allow for both country fixed effects and time effects, which measure those changes in the dependent variable E/N that cannot be explained by the five policy/institutional variables and the cyclical variable, the output gap. They find that the most important contribution is made by the time effects, explaining 5.4 of the 8.2 point increase in E/N, and they interpret this as largely due to increases in EU-15 female labor force participation arising from a shift in cultural acceptance of females in the workplace. They conclude that such time effects cannot be explained by the policy and institutional variables. This leaves 2.8 percent of the overall change to be explained by the explanatory variables other than the time effects. Of this, about two-thirds is explained by declining tax rates, and the remaining third is explained by a combination of lower values of the product and labor market regulation indexes, lower union density, and a decline in the business cycle variable measured by the output gap. Overall, much of the post-1995 revival in the growth of E/N in Europe seems to involve

Controversies about Work, Leisure, and Welfare

365

female labor force participation, not the turnaround of the tax, welfare, and policy variables that have dominated the debate in the literature. 11.5

Welfare Implications of the Decline in Hours per Capita

Why does it matter whether the decline in the EU/US ratio of hours per capita is mainly caused by higher labor taxes, employment and product market regulations, generous unemployment benefits, by other aspects of the welfare state, or political pressure engineered by unions? In each case the decline in European hours per capita is involuntary, in contrast to Blanchard’s interpretation that Europeans value leisure more than Americans. The Prescott tax story has Europeans pushed northwest along the labor demand curve, voluntarily choosing to reduce labor hours subject to the constraint of a large tax wedge between before-tax and after-tax labor income. Prescott’s interpretation includes the assumption that the high taxes buy high welfare benefits that are valued by citizens as much as the same monetary value of market consumption, thus eliminating any impact of the income effect and making the labor response into a voluntary movement up or down the labor supply curve. Employment and product market regulations and generous unemployment benefits also push European labor markets northwest along the labor demand curve. These explanations, which play an important role in the regressions of Basannini–Duval (2006) and Dew-Becker– Gordon (2008), are complementary with the Alesina et al. (2005) emphasis on the role of unions in the political process that created the regulations and the benefits. The Alesina political explanations imply that European households are not receiving leisure that they value as highly as in the standard economics textbook analysis. Ljundqvist– Sargent (2006) claim that the high labor taxes in Europe buy welfare benefits that are valued less than the equivalent monetary value of market consumption. Alesina assumes that unions and left-wing political parties push the labor market equilibrium away from that which would have been voluntarily chosen. We can provide additional insight by looking more closely at the nuances of how those spending time in unemployment, home production, and early retirement value leisure. How valuable is the leisure that Europeans gain from their shorter working hours due to higher unemployment and lower labor force participations?

366

Robert J. Gordon

11.5.1 The Welfare Cost of Higher Unemployment An elementary textbook analysis would value hours spent in unemployment just as in any other kind of leisure by multiplying each hour by the after-tax real wage. This overstates the value of unemployed time by ignoring the slope of the labor supply function. Only the marginal hour of leisure is worth the after-tax real wage; each additional hour of leisure (each hour less of work) is valued at less than the aftertax real wage. If the normal work week is 40 hours, 80 hours are spent in leisure (ignoring for now home production), and the remainder in sleep, then reducing work to 20 hours and raising leisure to 100 hours adds extra hours of leisure that are valued less than the after-tax real wage, due to the diminishing marginal value of leisure. In parallel, the 60th hour of leisure is worth more than the 80th, which is why workers receive premia for working overtime. Gordon’s (1973) analysis of the welfare cost of higher unemployment begins with the fundamental distinction between a temporary increase in unemployment caused by a short recession that pushes the actual unemployment rate above the natural rate of unemployment, as contrasted with a permanent increase in unemployment caused by a higher natural rate of unemployment. Europe’s transition from 2 percent unemployment before 1973 to 9 or 10 percent in the 1990s represents a permanent shift. Nevertheless it is interesting briefly to review the temporary case, where Gordon includes the welfare costs associated with temporary recessions that cause large welfare losses beyond the time use of the workers who shift from work to officially defined unemployment. These include the value of the lost work hours of those who leave the labor force and of those who work a shorter work week, as well as the value of the lower productivity of the remaining work hours. Lost hours and productivity are valued not at the after-tax real wage but at private output per hour because society loses all the output produced by the lost hours, including that which would otherwise go to indirect taxes, capital taxes, after-tax income to capital, and taxes on labor income. Society also loses the extra unemployment compensation that is paid to the unemployed. Consider the valuation of the time spent in search or at home by the unemployed. Gordon places a value on this time that for adult males is only about one-tenth private output per hour, with a higher fraction for females and teenagers. Part of the argument is that the estimated labor supply curve for adult males is nearly vertical, implying a zero value of leisure for those hours that are normally spent at work. A

Controversies about Work, Leisure, and Welfare

367

related argument, based on a survey of blue-collar workers, relates to a question regarding a hypothetical government payment. Would these workers require a government payment higher or lower than their present wage to stay at home rather than working? Seventy-five percent of the males responded that they would require a higher payment, 25 percent “the same,” and nobody said “less.” This reverses the normal textbook analysis that assumes that leisure raises utility and work reduces it, at least for adult males during the hours of the normal work week. Overall, a 1.0 percent temporary increase in the unemployment rate is associated with a 2.7 percent decline in market output when no value is imputed to nonmarket activity of those who shift from work to home activity, and this is reduced only from 2.7 to 2.3 when an appropriate price is applied to nonmarket activity (Gordon 1973: 162–64). The welfare effects of a permanent 1.0 percent increase in the unemployment rate are less than in the temporary case, largely because the procyclical movements of hours per employee and of productivity are absent. The reduction in labor input is assumed to be accompanied by a long-run unit elastic reduction in capital input, leaving the capital to labor ratio unchanged. However, in the permanent case the welfare analysis of the time value of unemployment for adult males remains the same, and there is only a minor offset of lost output by the value of leisure time. Gordon’s value of lost output associated with a permanent one percentage point increase in unemployment is 0.7 percent, as contrasted with 2.3 percent for the temporary case. 11.5.2 Early Retirement and the Valuation of Leisure Perhaps the most convincing aspect of the Alesina approach is the interplay between the political process and early retirement in Europe. If individual households in a welfare state are given the option of a defined benefit government-funded pension plan that allows them to retire at nearly full pay at age 58, they would be crazy to turn down the option of receiving the same income for not working as they would receive for working. Nevertheless, the survey cited above suggests that 75 percent of male respondents would need retirement pay higher than their current wage to consider retiring. The costs of early retirement to society can be illustrated by a simple example. Consider an economy that initially has people work from ages 20 to 65 and then retire from ages 65 to an assumed age of death of 84. There is no private saving. A 30 percent tax finances pay-as-you-

368

Robert J. Gordon

go pensions with a balanced government budget. This tax finances a level of consumption during the 20 years of retirement equal to consumption during the 45 years of work. Now let the politicians reduce the retirement age from 65 to 55. Instead of 45 years of work financing 20 years of retirement, now 35 years of work finances 30 years of retirement. The tax rate must increase from 30 to 45.6 percent. Even ignoring the Prescott-like withdrawal of work hours by people of working age that reduces market GDP, there is a 25.1 percent decline in consumption during both work years and retirement years. In short, a 22.3 percent reduction in total work effort (from 45 to 35 years) generates a 25.1 percent decline in consumption. With a few additional assumptions we can translate this decline in market consumption into a welfare measure. Let us ignore for this purpose the Gordon (1973) argument that adult males place little value on leisure time during the normal work week; shifting to those assumptions would strengthen the argument of this section that the leisure value of early retirement is a minimal offset to the lost market consumption caused by early retirement. Let us assume instead that hours that are normally spent by current workers in leisure-time activities, namely on weekday evenings and on weekends, are valued at 4/3 of the after-tax market wage but that hours switched from work to weekday leisure as a result of early retirement are valued at 2/3 of the after-tax market wage. Total welfare is market consumption plus the total value of leisure. The early retirees continue to enjoy high-valued weekday evening and weekend leisure but switch from market consumption to low-valued weekday daytime leisure. A simple simulation shows that as a result of the decline in market consumption of 25.1 percent determined above, total welfare declines by 22.6 percent, and the value of extra leisure as a result of early retirement offsets only 10 percent of the loss of market consumption that results from early retirement. Part of the reason for this is the 2/3 value of leisure for the work hours transferred into retirement leisure, but another less recognized part is that the increase in the tax rate from 30 to 45 percent required to finance existing government spending with a reduced number of work years reduces in proportion the after-tax wage and thus the value of both types of leisure, weekday and weekend. The time-study research by Freeman and Schettkat (2005: table 3) provides another qualification regarding the value of leisure time gained by those who are not working. They find that the time allocation of men across market work, home production, leisure, and personal

Controversies about Work, Leisure, and Welfare

369

time (mainly sleep) is quite similar in Europe and the United States, but the story is very different for women. On average, European women spend eight hours per week less in market work than American women and ten hours more in home production, and actually experience three hours less of leisure (the remaining small difference is in personal time). European mothers cook more at home; American mothers more frequently go out to eat as they spend their higher market income on market consumption. Higher labor force participation in the United States brings not only the benefits of higher market incomes that allow the substitution of restaurants and hired help for household drudgery but also provides for greater socialization as people remain in an organized social context during the workday in contrast to loneliness at home. As interpreted by Mees (2006), Europe has fallen into a “leisure trap” in which both the best educated and least educated women are out of the workplace. The Freeman–Schettkat evidence fits nicely with the D-H result showing that a substantially smaller share of work hours and GDP in Europe occurs in the trade and service sectors. As Mees interprets the transatlantic divide: Instead of performing these household jobs themselves, Americans pay other people to do them. Americans eat more often in restaurants, make ample use of laundry, dry-cleaning and shopping services, and hire nannies to take care of young infants. Indeed, in the US, one finds all kinds of personal services that do not exist on a similar scale in Europe. A manicure, carwash, or a massage is often only a stone’s throw from one’s home. Doorman buildings provide round-the-clock service to residents and dog-walkers look after pets during the workday. . . . By contrast, European women work less and have less money to spend on services. In their “free time,” European women are busy cleaning the house and looking after the children. On balance, therefore, European and American women work about the same amount of hours.

As we have seen, Freeman–Schettkat emerge with the amazing result that leisure is no different in Europe than in the United States. The extra hours of market work by Americans are completely balanced by extra hours of household production by Europeans. The Freeman–Schettkat evidence blends nicely with the Nordhaus (2006) discussion of time use summarized below—in several surveys work is actually viewed as more “enjoyable” than several aspects of home production in which European women participate more than American. The Mees list of service occupations that are much more common in America than in Europe echoes Gordon’s (1997) recitation

370

Robert J. Gordon

of four low-skilled jobs that have long been common in the United States but barely exist in the rich countries of Europe—grocery baggers, busboys in restaurants, parking lot attendants, and valet parkers. As a result of the post-1995 turnaround in Europe’s hours per capita, some of these contrasts are lessening, as witness the profusion of voituriers (“valet parkers”) in 2010 Paris. 11.5.3 Idle European Youth We learned from figures 11.5 and 11.6 that a major contributor to lower labor force participation in Europe compared to the United States is not only early retirement but also lower participation and higher unemployment among youth aged 15 to 29. The French riots of the banlieue in 2005 and riots in southern Italy in January 2010 remind us that many European youth are marginalized from contact with the market economy. Are unmarried Italian 30-year-old males sitting at home, insisting that their mothers cook for them and do their laundry, because they have a special taste for leisure or because the economy and society do not provide sufficiently rewarding jobs for them?6 Differences in the economic environment of American and European youth are pervasive. Because of the flexibility of American labor markets, American high-school students easily find after-school jobs in fast-food restaurants and other service outlets. Instead of receiving government-funded tuition grants for college, American youth are expected both by their parents and by colleges to work part-time during the school year and full-time during the summer. They adopt early a culture of work rather than idleness, and this continues after graduation from college. In contrast, judging from the low employment to population ratios for Europeans aged 15 to 29, much of the time in this European age group is wasted, especially when we recognize the larger share of American youth compared to European youth going to college and hence removed from the E/N ratio. 11.5.4 Does Conventional Economics Miss the Welfare Valuation of Work versus Leisure? Gordon’s (1973) reported survey result showed that adult male bluecollar workers prefer work to staying at home during the normal work week. Nordhaus (2006: 156–57) reports on a much more extensive set of evidence, the compilation by Robinson and Godbey (1997: 243) of several surveys of US households about their degree of “enjoyment” of different activities.7 The most striking result is that while some types

Controversies about Work, Leisure, and Welfare

371

of leisure activities are more enjoyable than work, in turn work is more enjoyable than other types of leisure and most types of home production. The many activities ranked in the compilation can be sorted into four groups that have roughly equal evaluations of enjoyment, ranked on a scale of 0 to 10. At the top in order with enjoyment scores between 7.8 and 6.3 are stereo, conversations, child play, sleep, eating, and cultural events. In the second group with scores between 5.8 and 6.0 are socializing, grooming, hobbies, child care, religion, reading, sports, and work. In the third group with scores of 5.0 to 5.4 (and thus less enjoyable than work) are relaxing, cooking, TV, education, and work commute. In the bottom group with scores of 4.3 to 4.8 are paperwork, organizations, maintenance, grocery shopping, cleaning, and other shopping. Nordhaus provides several interpretations of this surprising evidence that are relevant to the discussion of this chapter. First, the survey results may refer to average rather than marginal evaluations. People want to have some contact with “work,” possibly because of its social aspects, but at the margin work has sufficient disutility that few people choose to work in second or third jobs. Second, most workers cannot choose their hours and effectively have a marginal wage of zero, explaining why enjoyment from work does not differ markedly from many other nonwork activities. Third, and consistent with our discussion above, which places a different value on leisure hours during the normal work week from leisure hours in the evening and on weekends, is that people place different values on different times of the day and different days of the year. Fourth, Nordhaus considers as inconclusive the criticism that the survey results have methodological flaws. Nordhaus also raises the issue of simultaneous activities. In his example many home activities mix home production and leisure (cooking while watching TV or socializing with friends and family). To provide a more relevant example, office workers not only work, but they socialize in the cafeteria or near the vending machines, they use their high-speed Internet connections to shop on the Web, and some office workers (including two of my former secretaries) play games installed on their office computers. It is possible that the social aspects of work help explain the paradoxical result that work is as enjoyable as some other activities traditionally considered as leisure. But also it is possible that, at least for adult men, the absence of a job carries with it a social stigma. We have all read anecdotes about unemployed American men in the Great Depression or Japanese men in the 1990s who

372

Robert J. Gordon

would dress up and leave the house in the morning so that the neighbors would think they still had jobs, then frittering away the day across town until it was time to return home. 11.6

By How Much Does American GDP Overstate Welfare?

Up to this point the chapter has been about welfare interpretations of the decline in European hours per capita relative to the United States. By definition, this decline explains why Europe performs much better in comparisons with American productivity than in comparisons with American market output per capita. This section addresses several issues that concern the numerator of the productivity and output per capita ratios, namely real GDP itself. How much does measured real GDP with typical PPP exchange rate translations exaggerate or understate welfare in Europe compared to the United States? 11.6.1 Housing A considerable part of the US advantage in cross-country comparisons of living standards must stem from the much larger size of average American dwelling units, both their internal dimensions and the amount of surrounding land. Fully three-quarters of the American housing stock consists of single-family detached and attached units. The median living area in the detached units is 1,720 square feet, with an average acreage for all single-family units of 0.35 (equivalent to a lot size of 100 by 150 feet or 1,394 square meters). Another figure that must seem unbelievable to Europeans is that fully 25 percent of American single-family units rest on lots of one acre or more, equivalent to 4,052 square meters. Available data, although spotty for Europe, suggest that the average American dwelling unit is at least 50 to 75 percent larger than the average European unit.8 Since construction of new units and imputed rent on old units are included in GDP comparisons across countries, our EU/US ratio of per capita output in figure 11.3 already incorporates the superiority of the US housing stock (as long as the cross-country PPP-based price indexes make adequate allowance for housing quality). 11.6.2 Energy and Metropolitan Dispersion Yet a European might retort that while the gap between US and European standards already includes the housing difference, it also includes activities that are not welfare enhancing. A significant fraction of GDP

Controversies about Work, Leisure, and Welfare

373

in the United States does not improve welfare but rather involves fighting the environment whether created by nature or human-made decisions. The American climate is more extreme than in Europe (excluding the ex-USSR), and this means that some of GDP is spent on larger airconditioning and heating bills than in Europe to attain any given indoor temperature. The harsh American climate introduces additional issues in welfare comparisons beyond the narrow calculations of energy use. Americans in most regions of the country are afflicted by meterological events that rarely occur in Europe, including hurricanes, tornadoes, and forest fires. The world learned in early September 2005 that the Katrina hurricane could not only devastate a region but also a culture, could send residents to temporary homes hundreds of miles away, and also could reveal squalor and inequality that may not exist to the same degree in most of the EU-15 nations. The US GDP includes a sizable share of the insurance industry and also self-financed reconstruction that in some regions echoes the destruction of central Europe during World War II. The harsh climate itself does not actually represent much of an economic burden measured as a share of GDP. One source cumulates the total costs of insured disaster losses over the period 1986 to 2005 were $289 billion in 2005 dollars, a mere 0.15 percent of GDP over those twenty years measured in 2005 prices. Even if the losses are doubled to take account of uninsured losses and supplemented again by the costs of administering the insurance, the marginal cost of the severe US climate in terms of direct damage is unlikely to exceed 0.5 percent of GDP.9 Some of US GDP is spent on extra highways and extra energy to support the dispersion of the American population into huge metropolitan areas spreading over hundreds or even thousands of square miles, in many cases with few transport options other than the automobile. European real GDP is held down by the correctly measured high price of petrol, but sufficient credit is not given for convenience benefits from frequent bus, subway, and train (including TGV) public transit. High taxes in Europe provide not only the benefits of the welfare state but also large subsidies to allow high-speed rail and urban public transport to coexist with motor expressways of similar density to those in the United States. However, to the extent that tax-financed subsidies are the major source of the greater quality and density of public transportation in Europe, these transportation benefits (just as the entire European tax-financed welfare and medical-care system) are

374

Robert J. Gordon

already included in Europe’s real before-tax GDP that is the basis for the comparisons in table 11.1 and figure 11.1. While an economist’s first reaction is that the dispersion of US metropolitan areas must be optimal, since people have chosen to buy houses in the outer suburbs, a more careful reaction would be to view the American dispersion as related to public policy in addition to private choice, a point related to Alesina’s emphasis on political decisions that do not necessarily reveal voter preferences. These policies include subsidies to interstate highways in vast amounts relative to public transport, local zoning measures in some suburbs that prohibit residential land allocations below a fixed size, such as two acres, and the infamous and politically untouchable deduction of mortgage interest payments from income tax. Europeans enjoy shopping from small individually owned shops on lively central city main streets and pedestrian arcades, and recoil with distaste from the ubiquitous and cheerless American strip malls and “big-box” retailers—although Carrefour, Ikea, and others, provide American-like options in some European cities. To counter the effects of American land-use regulations that create overly dispersed metropolitan areas, European regulation includes land-use rules that preserve greenbelts and inhibit growth of suburban and exurban retailing and have indirectly prevented Europeans from enjoying either the low prices or high productivity growth of American big-box retailers. Tastes are in part the result of circumstances and habit, and to the European critique many Americans would deliver a counter-retort. An American mother of two small children wants nothing to do with schlepping those kids through endless tunnels while making connections on the London or Paris subways, or with waiting in the rain for the next bus, or with shopping for groceries more often than once per week. The three-quarters of American households living in singlefamily units treasure their backyards, decks, and barbeques and do not want to be forced to go to a public park for outdoor recreation—whose barbeque grill would they use, and why should they have to compete with others for a limited supply of public picnic tables? European land-use planning that restricts the growth of Americanstyle big-box retailing is considered by many analysts to be the single most important reason for the slump in European productivity growth shown in figure 11.1 (e.g., see Inklaar, O’Mahony, and Timmer 2005; McGuckin, Spiegelman, and van Ark 2005). Reform of European land-

Controversies about Work, Leisure, and Welfare

375

use planning is the first-listed recommendation for structural reforms in the comprehensive recent evaluation of the European economy by Baily and Kirkegaard (2004: 8). In fact, consideration of American big-box retailing reveals it not only to be a source of rapid productivity growth in retailing since 1990 but also a source of unmeasured US growth in output per capita. As shown by Hausman and Leibtag (2005), Wal-Mart reduces retail food prices by 25 percent, of which 20 percent is the direct effect of its own low prices and the other 5 percent represents the reduction of prices by competitive stores. Because the US Consumer Price Index “links out” price differences among outlets for the same products (the so-called outlet substitution effect, the impact of Wal-Mart and other big-box retailers in lowering prices is ignored by the CPI and as a result in the deflators for US personal consumption expenditures and GDP. Putting it simply, European restrictions that protect inner-city pedestrian shopping districts create value for the rich, who live inside Europe’s cities, while the restrictions that make it hard for low-priced big boxes to establish in the city and suburbs hurt the poor.10 European regulations also restrict the sale of nonprescription drugs in self-service aisles; in many countries routine nonprescription drugs are still dispensed on a one-by-one basis to individual customers by pharmacists. This reduces retail productivity and drives up the cost of living for many Europeans. This European set of policies that favor the rich and hurt the poor may offset some of the increased skewness of the American income distribution, discussed below. Even if part of American energy use is not welfare-enhancing, either because it offsets the harsh climate or politically motivate “excess dispersion” of American metropolitan areas, how much could this possibly be worth? Figure 11.11 displays the time path of energy consumption per dollar of GDP in the United States and Europe since 1980. Despite the continuation of low gasoline taxes in the United States, the gap between American and European energy use has narrowed and now amounts to no more than 2 percent of GDP. If we take half of that gap as welfare enhancing (the value of heating large interior spaces and driving larger cars and trucks), and the other half as non– welfare enhancing (offsetting the harsh climate and unnecessary driving caused by excess dispersion and the lack of public transit), the energy story emerges with an overstatement of US welfare by only 1 percent of GDP. Other US expenditures, including keeping 2 million people in prison, might add another 1 percent of GDP in non–welfareenhancing activities.

376

Robert J. Gordon

16,000 14,000

United States 12,000 10,000 8,000

Europe - 15 6,000 4,000 2,000 0 1980

1985

1990

1995

2000

2005

Figure 11.11 BTUs of energy consumption per dollar of GDP, Europe-15 versus United States, 1980 to 2006

11.6.3 Insecurity and Inefficiency Finally some attention must be paid to the widespread European distaste for American institutions as providing insufficient welfare benefits, job security, and inadequate medical care. Louis Uchitelle (2006) documents the “human damage” inflicted by layoffs of the low-skilled blue-collar worker and high-skilled professional alike. However, to make a separate allowance for the benefits of the European welfare state would amount to double counting. We have compared real GDP per capita between the European Union and United States on a beforetax, not after-tax basis. Thus (as assumed also by Prescott 2004) we implicitly assumed that high taxes in Europe are buying government expenditures that are valued by households as highly as an equal dollar of private consumption. Ljundqvist–Sargent (2006) question this assumption as ignoring the inherent inefficiency in government spending. I would offset this alleged inefficiency of higher European government expenditure against the notable inefficiency of the American medical care system, which spends a much larger share of GDP and yet produces mediocre outcomes in life expectancy and medical care inputs. As much as 3 percent of US GDP may be wasted in excessive

Controversies about Work, Leisure, and Welfare

377

administrative expenditure by the private insurance companies that run the American medical care system. This is clearly a component of GDP that does not raise the welfare of American consumers of its dysfunctional medical care system and should be added to our final tally of the amount by which market GDP comparisons overstate US welfare. 11.6.4 Immigration and the Black Economy Three final issues must be considered in an evaluation of European and American welfare. The first two issues involve immigration and the black economy. As many as 11 million Americans are illegal immigrants. To the extent that they are working for cash and do not pay taxes, their contribution to GDP is missed, and American GDP is understated. Since most of the illegal immigrants are picked up in the population census, American GDP per capita is understated. Because of tighter border controls, fewer immigrants in the United Kingdom and continental Europe are illegal. However, any unmeasured GDP in the United States should be offset by the incentive in Europe of high labor taxes to enlarge the underground economy. Without definitive information on the role of illegal immigrants in the United States as compared to the black economy in Europe, we consider this issue to be a toss-up and do not include it in our final score-keeping in table 11.2. 11.7

A Summary of the Welfare Adjustments

Table 11.2 summarizes the results of the chapter and asks the question, how much of the measured shortfall of European relative to American income per capita is eliminated by adding the value of extra leisure in Europe or subtracting non–welfare-enhancing components of American GDP? In the top row there is copied from table 11.1 the 2008 EU/ US ratio of 71.2 percent for real GDP per capita and in row 8 the 2008 ratio of 83.0 for real GDP per hour. How much can we add to the initial ratio of 71.2 percent? Are the additions sufficient to exceed the 83 percent EU/US productivity ratio in 2008? The most important addition reflects the value of extra leisure in Europe as a result of declining European work hours per employee. However, the leisure gained from each employee working fewer hours than in the United States is worth surprisingly little because the value of the after-tax wage in Europe is so low. The 2004 value of the beforetax share of employee compensation in EU-15 GDP was only 55 percent,

Market PPP ratio of Y per capita (table 11.1)

Add: 1/2 times 2/3 of 1/3 ratio of real after-tax wage to GDP to allow for EU/US difference in hours per employee (table 11.1)

Add: 1/10 of difference in employment per capita (4.1 percentage points from data underlying figure 11.4)

Add: Half of energy use difference

Add: Prisons and other

Add: Medical care inefficiency

Sum of market PPP ratio and above additions (sum rows 1–6)

Market PPP ratio of Y per hour (table 11.1)

Percent productivity gap explained (row 7 minus 1 divided by row 8 minus 1)

Percent total gap explained (row 7 minus 1 divided by 100 minus row 1)

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Table 11.2 Summary of adjustments to the EU/US ratio of per capita income, 2008

22.9

55.9

83.0

77.8

71.2

EU/US ratio of real GDP per capita

0.4

1.2

Adjustment to leisure component of hours

3.0

1.0

1.0

Adjustment to GDP

378 Robert J. Gordon

Controversies about Work, Leisure, and Welfare

379

about six points below labor’s share in the United States. To calculate the after-tax wage, we multiply the 55 percent labor’s share by unity minus the labor tax rate of 39 percent based on population-weighted labor tax rates across the EU.11 This brings the after-tax wage down to only 33 percent of real GDP. But we have argued above that converting hours from mid-week work time to leisure is worth less than the value of leisure at the margin, say 2/3 the real after-tax wage. This brings the value of leisure converted from work to vacations or short hours per week to 2/3 times 1/3 of GDP, or 2/9 (22.2 percent). But this still exaggerates the value of European leisure, since we have cited sources that show that Europeans do not enjoy more leisure than Americans, with extra home production more than offsetting shorter hours of work. Our inclination is to split the difference and to credit Europe with half of the 22 percent of GDP that the leisure would be worth if every hour of reduced work was converted into leisure and none into home production. While the Freeman–Schettkat evidence suggests that Europeans do not have more leisure to enjoy than Americans, Europe should still be given some credit because having longer vacations is surely valuable and because some aspects of home production are combined with leisure, as in watching TV while cooking. Thus in row 2 of table 11.2 we have added 11 percent of the 11.1 point gap between Europe and the United States in lower hours per employee. However, for the reduction in the E/N ratio, we view this as largely involuntary and, using our example of early retirement, providing a relatively small value of additional leisure, roughly one-tenth of the value of output that could have been produced by those extra hours. This adds another 0.4 percent, equal to one-tenth of the 4.1 percentagepoint difference between employment per capita in Europe compared to the United States. The three adjustments to real GDP add, first, 1.0 points for excess US energy use. Second, another 1.0 percent is added to reflect the wasted resources created by excess incarceration and the creation of a gigantic prison population of 2 million people who in their future life are deprived of educational and job opportunities as a result of their prison records. Third, 3.0 percent of US GDP is added to reflect the excess administrative costs of the American medical care system, with its battalions of clerks employed by insurance companies to dispute claims submitted by doctors and hospitals, and the countervailing battalions of extra clerical personnel employed by doctors and hospitals to pursue their half of the never-ending battle.

380

Robert J. Gordon

Adding together these supplements to the European standard of living raises the ratio from the initial 71.2 percent at the top of table 11.2 to a more robust 77.8 percent, and this explains more than half of the initial 11.8 point gap between the EU/US ratio of output per capita to output per hour. However, the full gap between Europe and the United States to be explained is between Europe’s 71.2 percent of US per capita GDP and the US level of 100 percent, or a gap of 28.8 percent. The adjustments in table 11.2 eliminate only 23 percent of the total EU/ US income per capita gap. 11.8

Conclusion

This chapter has examined two classes of arguments implying that standard PPP-based ratios of European output per capita relative to that of the United States understate true European welfare. As is documented in the chapter, the EU/US ratio of real GDP per capita has languished at close to 70 percent since 1975, while the same ratio for productivity (output per hour) reached 92 percent in 1995 before falling back to 83 percent in 2008. By definition, the gap between the income per capita ratio and the productivity ratio represents the influence of the decline in European hours per capita relative to the United States until 1995 and its partial recovery since then. A novel contribution of this chapter has been to show that even if all the decline in European hours per capita represented a voluntary transfer of work hours to pure leisure, that leisure is not worth much. The traditional valuation of an extra leisure hour is the after-tax wage per hour. But in Europe, with labor’s income share only 55 percent and 39 percent of pre-tax labor income removed by taxation, the after-tax wage per hour is only 33 percent of GDP per hour. But this is still an overstatement of the value of leisure, for three reasons. First, the declining marginal value of any commodity applies to leisure; each extra hour taken by Europeans in longer vacations has a value below the after-tax real wage. This chapter suggests that the value of leisure hours taken on weekends (which are work-free both in America and Europe) may be worth say 4/3 of the after-tax wage, while leisure hours obtained by transferring mid-week hours of work to leisure may be worth only two-thirds of the after-tax wage. Second, most of the decline in European hours per capita has taken the form of higher unemployment, particularly of youth, and lower

Controversies about Work, Leisure, and Welfare

381

labor force participation of both young and old people. Ample evidence suggests that European youth to a much greater extent than in the United States are detached from the labor market, experiencing not just low participation but spells of unemployment that are much longer than in the United States. For all age groups the average European unemployed person is five times more likely to be unemployed more than one year and only one-quarter as likely to experience an unemployment spell less than three months. Third, time-use studies reported by Freeman–Schettkat (2005) and others turn the standard European interpretation of “less work, more leisure” on its head. Hours transferred from work in Europe are not spent in leisure activities but rather in household production. Americans use their higher market income to buy services much more often than in Europe with its proportionately smaller retail service sector. Europeans cook more and devote more time to household chores and child-rearing, while Americans go out to eat, hire cleaning services, and buy child-care services. The Freeman–Schettkat evidence concludes that Europeans work less than Americans but do not enjoy more leisure, despite their fabled long vacations.12 In the summary table 11.2 modest credit is given to Europeans for their extra leisure despite the Freeman– Schettkat evidence, both because longer vacations must involve at least some valuable leisure and because some household chores can be combined with entertainment such as watching TV. The “tax multiplier” emphasized in this chapter is often neglected. A reduction in work hours reduces income. To the extent that government expenditures remain fixed, the tax rate on each remaining labor hour must increase, and to the extent that the lower work hours are caused by earlier retirement, the resulting increase in government pension expenditures raises the tax rate on the remaining work hours even more. In a simple example, a reduction in the retirement age from 65 to 55 requires an increase in the average tax rate from 30 to 45 percent to maintain average consumption in retirement at the same (reduced) level as during work years. This chapter has provided a review of debates involving four leading interpretations of the relative decline in European hours per capita. These are that most or all of the difference represents a different taste for leisure in Europe, that all of the difference reflects high taxes on labor in Europe, that much of the difference represents the effects not of high taxes but of an overly generous welfare state, and finally that

382

Robert J. Gordon

hours per capita have been driven down not by voluntary choices but by political pressure initiated by unions that have promoted productmarket and labor-market regulations leading to high labor costs, high unemployment, a shorter work week, long vacations, and early retirement. In sorting through the debate about these explanations, we examined data that allows us to make three distinctions that rarely appear in the literature. First, the time-series evidence shows that from 1960 to 1995 only one-third of the relative decline in European hours per capita was due to a decline in hours per employee, namely due to the famous European long vacations and short work week. The remaining two-thirds was divided into roughly two-thirds due to falling labor force participation and one-third to rising unemployment, both corrected for differences between the United States and Europe in the composition of the working-age population by age group. Second, the time-series data showed a distinct turnaround after 1995. While hours per employee continued to fall in Europe relative to the United States, albeit at a slower rate, there was a complete turnaround in the behavior of employment per capita, from 35 years of steady decline to 9 years since 1995 of steady increase. None of the recent literature on European hours, at least that cited here, calls attention to this turnaround nor provides any explanation of this phenomenon. Third, our examination of European and US unemployment rates and labor force participation rates by age group showed another little discussed contrast. The unemployment rate is higher across the board in every European age group. But for labor force participation the pattern is completely different. Among prime-age workers (aged 30 to 44) European participation rates are identical to those in the United States, whereas participation rates are much lower in the 15 to 29 and the over 60 age groups. These patterns make it unlikely that a single explanation of lower European hours per capita can suffice. For instance, if high labor taxes are the dominant cause of falling European hours per capita, why did this not affect the labor force participation rate of prime-age Europeans at all? From the econometric research that we cite there emerges a surprising conclusion. Fully two-thirds of the growth revival in European employment per capita after 1995 is due not to policy responses such as lower taxes and loosened regulations but rather to an unexplained

Controversies about Work, Leisure, and Welfare

383

“time effect” that has raised E/N over time. We interpreted this as reflecting a cultural change favoring the movement of females from home production into market employment. Of the remaining increase in employment per capita not explained by the time effect, a reduction in the tax wedge explains fully two-thirds. The chapter then turned to possible dimensions in which measured PPP GDP overstates welfare in the United States compared to Europe. The easiest case to make is that the United States has a harsher climate and so some of the extra energy consumption in the United States (measured relative to GDP) is not welfare enhancing. A more debatable position is that the United States has long instituted policies that have created overly dispersed metropolitan areas with few public transit options, also leading to excess energy use. However, the extra use of energy in the United States compared to Europe is currently worth only around 2 percent of GDP, so any allowance for “excess” energy use could at most account for only 1 percent of GDP. Our discussion of GDP overstatement also made an allowance of 1 percent of GDP for excessive incarceration in prisons. A brief discussion of insecurity, inefficiency, immigration, and the black economy identified only one further source of overstatement of US GDP, and this is the estimated 3 percent of GDP that is lost to the inefficiency of the dysfunctional US medical care system. Almost everything discussed in this chapter is debatable. Some of the adjustments are subjective. But this chapter is the first to put the issues in the terms of the matrix format of table 11.2. Future research will need to address table 11.2, as to whether additional line items should be added, and what way individual items should be changed, should they be moved higher or lower? This chapter has begun the process not just of debating the causes of relatively low hours per capita in Europe but also of rethinking the translation of real GDP into welfare comparisons across countries and regions. Notes 1. As indicated before, all data on GDP, population, and hours come from the Groningen economywide database, which has assembled data for many countries going back to 1950. 2. These are the “G-K” (Geary–Khamis) weights calculated in 1990 dollars and the “E-K-S” (Eltetö, Köves, and Szulc) weights calculated in 2008 dollars. All data on productivity, income per capita, and hours per capita come from the Groningen economywide database http://www.conference-board.org/economics/database.cfm.

384

Robert J. Gordon

3. The ratios here are higher than in the official US data (e.g., 66.0 percent LFPR for 2007) because our data exclude the population aged 75 and over. 4. This comparison is not appropriate, because the measure of hours in the tax correlation is H/N but is H/E in the union correlation. 5. This can be calculated as a response of H/N of 122 fewer hours divided by a 1995 total of 1,067 hours per adult, to a change in the combined tax rate of 12.8 percentage points on a base of 53.7 points in 1995. See Davis and Henrekson (2004: 38) for the changes and table 11.1 for the 1995 base values. 6. Roughly 52 percent of Italians between the ages of 20 and 34 live at home with their parents (Rhoads 2002). 7. Five surveys were taken from 1965 to 1985 with samples sizes ranging from 133 to 2,500. 8. The average estimated useful floor space of dwellings in 1997 or 1998 was 2,058 square feet for the United States and 995 for the average of Austria, Denmark, Finland, and Switzerland (none of the large European countries are listed). For newly constructed dwellings, “average living floor space” for Germany and Italy was 969. See United Nations, Annual Bulletin of Housing and Building Statistics for Europe and North America 2000, pp. 21 and 24, obtained from www.unece.org/env/hs/bulletin/00pdf/h10.pdf. An alternative measure for the United States in 2003 is a median square footage of all existing single detached and mobile homes occupied year-round (71 percent of all occupied yearround housing units) equal to 1,756. For all newly constructed privately owned singlefamily houses in 2004, the median was 2,140 and the average was 2,349. See Statistical Abstract of the United States: 2006, tables 951 and 932, respectively. The former table is the source of the average lot size data given in the text. All available data for the United States seem to refer only to single-family units and omit apartments in multifamily units, which presumably are smaller in size. 9. The total of $289 billion in 2005 prices comes from slide 8 of nhc2007-1.ppt, a Powerpoint presentation available by googling “ISO on historic catastrophe losses.” 10. I owe this connection between retail regulation and the income distribution to Ian Dew-Becker. 11. The tax rates come from oecd.org/document/4. 12. Even the extent of European long vacations has been exaggerated. It is not five weeks for Europeans and two weeks for Americans. Mercer Human Resources Consulting reports a total of 33.7 annual vacation and paid holidays per year in Europe compared to 25 in the United States.

References Alesina, A., and G.-M. Angeletos. 2005. Fairness and redistribution: US versus Europe. The American Economic Review 95 (September): 913–35. Alesina, A., and E. Glaeser. 2004. Fighting Poverty in the US and Europe: A World of Difference. Oxford: Oxford University Press. Alesina, A., E. Glaeser, and B. Sacerdote. 2006. Work and leisure in the United States and Europe: Why so different? NBER Macroeconomics Annual 2005 20: 1–64.

Controversies about Work, Leisure, and Welfare

385

Alesina, A., and E. La Ferrara. 2005. Preferences for redistribution in the land of opportunities. Journal of Public Economics 89 (5–6): 897–931. Baily, M. N., and J. F. Kirkegaard. 2004. Transforming the European Economy. Washington, DC: Institute for International Economics. Bassanini, A., and R. Duval. 2006. Employment patterns in OECD countries: Reassessing the role of policies and institutions. Social, Employment, and Migration working paper 35. OECD, Paris. Blanchard, O. 2004. The economic future of Europe. Journal of Economic Perspectives 18 (4): 3–26. Daveri, F., and C. Jona-Lasinio. 2005. Italy’s decline: Getting the facts right. Working paper 301. IGIER, Milan. Davis, S. J., and M. Henrekson. 2004. Tax effects on work activity, industry mix, and shadow economy size: Evidence from rich-country comparisons. Working paper 10509. NBER, Cambridge, MA. Dew-Becker, I., and R. J. Gordon. 2008. The role of labour-market changes in the slowdown of European productivity growth. Working paper 13840. NBER, Cambridge, MA. Freeman, R., and R. Schettkat. 2005. Marketization of household production and the EU-US gap in work. Economic Policy 20 (41): 6–50. Goldin, C., and R. A. Margo. 1992. The great compression: The wage structure in the United States at mid-century. Quarterly Journal of Economics 107 (February): 1–34. Gordon, R. J. 1973. The welfare cost of higher unemployment. Brookings Papers on Economic Activity 4 (1): 133–95. Gordon, R. J. 1997. Is there a tradeoff between unemployment and productivity growth? In D. Snower and G. de la Dehesa, eds., Unemployment Policy: Government Options for the Labour Market. Cambridge: Cambridge University Press, 433–63. Gordon, R. J., and I. Dew-Becker. 2007. Selected issues in the rise of income inequality,” Brookings Papers on Economic Activity, vol. 38, no. 2, pp. 191–215. A longer and more complete version is NBER Working Paper 13982, May 2008. Hausman, J., and E. Leibtag. 2005. Consumer benefits from increased competition in shopping outlets: Measuring the effect of Wal-Mart. Working paper 11809. NBER, Cambridge, MA. Inklaar, R., M. O’Mahony, and M. P. Timmer. 2005. ICT and Europe’s productivity performance: Industry-level growth account comparisons with the United States. Review of Income and Wealth 51 (4): 505–36. Ljungqvist, L. 2006. Comment on “A. Alesina, E. Glaeser, and B. Sacerdote, Work and leisure in the United States and Europe.” NBER Macroeconomics Annual 2005 20: 65–77. Ljungqvist, L., and T. J. Sargent. 2006. Indivisible labor, human capital, lotteries, and personal savings: Do taxes explain European employment? Paper presented at NBER Macroannual conference, April 7–8. McGuckin, R. H., M. Spiegelman, and B. van Ark. 2005. The retail revolution: Can Europe match the US productivity performance? Perspectives on a Global Economy. New York: Conference Board.

386

Robert J. Gordon

Mees, H. 2006. Europe’s leisure trap. Project Syndicate. www.project-syndicate.org (accessed July 8). Nordhaus, W. D. 2006. Principles of national accounting for nonmarket accounts. In D. W. Jorgenson, J. Steven Landefeld, and W. D. Nordhaus, eds., A New Architecture for the US National Accounts. Chicago: University of Chicago Press, 143–60. Olovsson, C. 2004. Why do Europeans work so little? Working paper. Stockholm School of Economics. Paramio, J. L., and J. L. Zofio. 2005. Labor market duality and leisure industries in Spain: Quality of life versus standard of living. Working paper. Universidad Autonoma de Madrid. Prescott, E. C. 2004. Why do Americans work so much more than Europeans? Federal Reserve Bank of Minneapolis Quarterly Review 28 (1): 2–13. Ramey, V. A., and N. Francis. 2006. A century of work and leisure. Working paper 12264. NBER, Cambridge, MA. Rhoads, C. 2002. Short work hours undercut Europe in economic drive. Wall Street Journal, August 8, p. A1. Robinson, J., and G. Godbey. 1997. Time for Life: The Surprising Ways Americans Use Their Time. University Park: Pennsylvania State University Press. Rogerson, R. 2006. Comment on “A. Alesina, E. Glaeser, and B. Sacerdote, Work and leisure in the United States and Europe.” NBER Macroeconomics Annual 2005 20: 79–95. Smeeding, T. M. 2006. Poor people in rich nations: The United States in comparative perspective. Journal of Economic Perspectives 20 (1): 69–90. Sonin, K. 2003. Why the rich may prefer poor protection of property rights. Journal of Comparative Economics 31: 715–31. Uchitelle, L. 2006. The Disposable American: Layoffs and their Consequences. New York: Knopf. Warren, S., and S. Worthington. 2004. The impact of taxes, laws, and inequality on hours worked. Working paper. University of Puget Sound.

12

Revisiting the Nordic Model: Evidence on Recent Macroeconomic Performance Jeffrey D. Sachs

12.1

Introduction

For twenty-five years the US political system has been in the midst of a heated debate over taxes and growth. US supply-siders have argued that high rates of taxation cripple economic growth and lower living standards, and that tax cuts are indeed sufficiently stimulating to economic activity as to be self-financing. From the supply-sider point of view, low tax rates in the United States explain US dynamism, while high European tax rates explain European stagnation. More generally, Europe’s tax-financed social welfare state is seen as undermining incentives both due to high tax rates and the high levels of social spending that they finance. Europeans too are in a long-standing debate about the social welfare system. With chronically high unemployment in several European countries, many “Euro-pessimists” are calling for significant cuts in taxation and social expenditures, often with the same argumentation as the US supply-siders. The claim is that the social welfare state is no longer affordable, especially in a globally interconnected economy. As often happens, these debates have shed much more heat than light, since the use of evidence has been wildly selective. This is ironic, since the cross-country evidence actually gives us a generation of macroeconomic experience operating under very different forms of capitalism. The range of social spending and taxation relative to national income across the OECD countries is very large and relatively stable. To the extent that high taxation and social expenditures cripple economic activity, we should have plenty of evidence over a considerable number of years to show that that this has been the case. Considered in this chapter are three groups of high-income countries. The first group is the Nordic (N) economies of Denmark, Finland,

388

Jeffrey D. Sachs

Norway, and Sweden. All of these countries maintain very high levels of social expenditure as a share of GDP. The second group of countries includes the core European continental (EC) countries of the European Union: Austria, Belgium, France, Germany, Italy, and the Netherlands.1 The third group of countries includes the English-speaking (ES) countries: Australia, Canada, Ireland, New Zealand, United Kingdom, and United States, all of which have a lower share of social spending in GDP than the European continental states and the Nordic states. Excluded from the analysis are the very small countries of the OECD (Iceland and Luxembourg), the poorer southern European states (Greece, Portugal, and Spain), Switzerland (as a non-EU country distinct from other continental economies), and the developing-country members of the OECD. The evidence reviewed in this chapter, on the economic and political performance of these countries in view of their vastly different systems of social protection, is with regard to the question: Is there evidence that the high rates of taxation have caused low rates of economic growth, low levels of income per person, major disincentives to work, and perhaps even a diminution of freedom (in line with the Von HayekFriedman argument that state intervention leads to a loss of freedom)? The short answer is no. Despite the vigor of supply-sider arguments in the United States and the corresponding euro-pessimist arguments in Europe, the Nordic countries have not suffered obvious liabilities regarding economic growth, standards of living, labor force participation, or political freedoms. If anything, the data suggest in fact the opposite. Before proceeding, it is important to make two further points. First, like the Anglo-Saxon economies, the Nordic economies are overwhelmingly private-sector owned, open to trade, and oriented to international markets. Financial, labor, and product market forces operate powerfully throughout non-state sector. In short, these are capitalist economies. Moreover they are far from rigid. Industrial change is accepted, even encouraged pro-actively. For example, change in the productive economy (both the creation of new sectors and the “creative destruction” of declining industries) is encouraged through active labor market policies, public-sector commitments to higher education, retraining, and R&D, and other institutions. Second, there is no single Nordic model, and still less, an unchanging Nordic model. What has been consistently true for decades is a high level of public social outlays as a share of national income, and a

Revisiting the Nordic Model

389

sustained commitment to social insurance and redistributive social support for the poor, disabled, and otherwise vulnerable parts of the population. The details of those policies—including labor market regulations, incentive structures, and coverage rates—have evolved over time, sometimes in the face of financial difficulties, high unemployment, or institutional failures. Nordic governance, in other words, has generally been active and alert to the need for change. 12.2 Levels of Social Outlays Since the supply-side critique focuses on the alleged high costs of taxation for economic well-being, I start with a comparison of the total tax take across these groups of countries. As can be seen in figure 12.1a, the ratio of total government receipts (taxes plus other receipts) to GDP varies from an average of 56 percent of GDP in the Nordic countries to 47 percent of GDP in the EC countries, and 38 percent of GDP in the ES countries. Figure 12.1b shows these ratios by individual country. Notice that Japan and the United States are at the far low end of the scale, with government receipts of around 32 percent of GDP, roughly half of the level in the Nordic countries, which are grouped at the high end of the scale.2 (In figure 12.1 all group averages are simple unweighted averages.) The ratio of government receipts to GDP is, of course, very highly correlated with the ratio of government outlays to GDP (r = 0.85, spearman rank correlation = 0.85). Figure 12.2a shows that total government outlays averaged 52 percent of GDP in the Nordic countries, 49 percent of GDP in the EC countries, and 38 percent of GDP in the ES countries.3 Once again, when we look at the individual countries in figure 12.2b, we see that the United States is at the very low end of the scale while the Nordic countries are still grouped at the top of the scale. The key difference in fiscal outlays across countries is in the levels of social expenditures to GNP, including both cash transfers and the public provision of social services. The average level of social outlays in GDP in the three groups of countries is shown in figure 12.3a and for individual countries in figure 12.3b.4 The correlation between total outlays to GDP and social expenditures to GDP is r = 0.85, and the Spearman rank correlation is 0.84. In simple terms, the major fiscal differences across these countries, both in total tax collections and public outlays, lie mainly in regard to how these countries deal with social expenditures. The Nordic

390

Jeffrey D. Sachs

(a)

Percent of GDP

60

40

20

0

English

Europe

Nordic

(b)

Percent of GDP

60

40

20

0

US IR AU CA UK GE IT NE NZ AR BE FR FI DE NO SW

Figure 12.1 Government receipts as share of GDP: (a) high-income country groups; (b) individual high-income countries compared. Nordic economies are Denmark, Finland, Norway, Sweden; European continental countries are Austria, Belgium, France, Germany, Italy, Netherlands; English-speaking countries are Australia, Canada, Ireland, New Zealand, United Kingdom, United States

Revisiting the Nordic Model

391

(a)

50

Percent of GDP

40

30

20

10

0 English

Europe

Nordic

(b)

Percent of GDP

60

40

20

0 IR AU US NZ CA UK NO NE GE IT

BE AR FI FR DE SW

Figure 12.2 Government outlays as share of GDP: (a) high-income country groups; (b) individual countries compared. Nordic economies are Denmark, Finland, Norway, Sweden; European continental countries are Austria, Belgium, France, Germany, Italy, Netherlands; English-speaking countries are Australia, Canada, Ireland, New Zealand, United Kingdom, United States

392

Jeffrey D. Sachs

(a)

Percent of GDP

30

20

10

0 English

Europe

Nordic

(b)

Percent of GDP

30

20

10

0 IR US CA AU NZ NE UK NO IT

FI AR BE GE FR SW DE

Figure 12.3 Public social expenditures as share of GDP: (a) high-income country groups; (b) individual countries compared. Nordic economies are Denmark, Finland, Norway, Sweden; European continental countries are Austria, Belgium, France, Germany, Italy, Netherlands; English-speaking countries are Australia, Canada, Ireland, New Zealand, United Kingdom, United States

Revisiting the Nordic Model

393

countries are high-tax countries, financing a high level of social expenditures. The English-speaking countries are low-tax countries, financing a much lower level of social expenditures. The European continental countries fall in the middle. 12.3 Characteristics of Social Outlays Public sector social outlays are divided between cash transfers, direct government provision of services, and active labor market policies (e.g., job training and government hires under jobs programs). Cash transfers include transfers to retirees (pensions and survivor benefits) plus cash transfers to working-age households on the other. Government social services are divided between health and non–health services (e.g., child care and disability care). The breakdown of these main categories of social outlays is as shown in table 12.1. I will refer to the sum of the first two categories (cash transfers plus direct government provision of services) as direct public social outlays. These plus spending on active labor market programs equal total public sector outlays. We see that the Nordic countries are distinctive not only in their overall high level of social expenditures but also in their high direct provision of non–health social services, such as child care. These directly provided services are important not only for the services themselves but also for the public employment positions that they represent. The Nordic countries hired many otherwise hard to employ individuals into the government social sectors in the past twenty years as part of their labor market strategy. Table 12.1 Public-sector social outlays (share of GDP)

Country English-speaking

Cash transfers

Direct provision of services

Active labor market policies

Total public sector social outlay

9.8

7.2

0.4

17.4

Europe

16.8

8.0

1.0

25.8

Nordic

14.2

11.4

1.2

26.8

7.9

6.7

0.2

14.8

United States

Source: OECD, Social Expenditure Database, 1980–2001 (2004: www.oecd.org/els/social/ expenditure). Note: Values cited are for 2001.

394

Jeffrey D. Sachs

The low level of public social expenditures in GDP in the United States compared with the other countries is offset in part by a higher level of private social expenditures in GDP. The United States has given greater weight to private over public outlays in several categories of social spending, including health care, pensions, and child care. US private outlays include 5.0 percent of GDP for health expenditure and 4.7 percent of GDP for private pension savings, compared with just 0.1 and 0.9 percent of GDP, respectively, for the Nordic countries.5 One of the important features of private outlays, of course, is that they are not redistributive in nature. They contribute little if at all to poverty reduction. Another feature seems to be the low efficiency of private spending on health, for reason of market failure in the health sector first elaborated by Arrow (1962). We will note from the evidence below that the United States does not get much “bang for the buck” out of its heavily private health spending. When public and private outlays are added together, the total US social outlay is 25.1 percent of GDP (14.8 percent public and 10.3 percent private), which is still considerably below the level of the European continental countries and the Nordic countries6. 12.4

Social Outlays and Poverty

The first important observation about the Nordic social welfare systems is that they succeed in reducing poverty. The OECD collects comparative data on inequality in three relevant ways: (1) the share of the population living at less than 50 percent of the average household income (standardized by household size), (2) the share of disposable income (after-tax and transfer) received by the bottom 20 percent of the population, and (3) the Gini coefficient on income distribution.7 As shown in table 12.2, on all accounts the Nordic countries rank as the most equal Table 12.2 Inequality and poverty indicators Share of disposable income to lowest quintile

Gini coefficient

Country

Poverty rate

English-speaking

12.6

7.3

32.0

9.0

8.4

28.0

Europe Nordic United States

5.6

9.7

24.7

17.1

6.2

35.7

Revisiting the Nordic Model

395

of the three groups of countries, with the difference in means of these measures across the groups highly significant. The average poverty rate in the Nordic countries in 2004 was just 5.6 percent of households, compared with 9 percent in Europe and 12.6 percent in the Englishspeaking countries. The United States, among the richest of all the countries in per capita GDP, has also by far the highest poverty rate, at 17.1 percent of households. The cross-country evidence suggests that public social spending does reduce poverty. A simple regression of the poverty rate on public social outlays and private social outlays shows the negative effect of public social outlays on poverty (table 12.3, regression 1). Next we separate public social outlays into direct public social outlays and active labor market expenditures, and find that both the direct outlays and the active labor market policies have a significant negative effect on the poverty rate while once again the private social expenditures have no statistically significant effect (regression 2). When we divide the direct public outlays into cash transfers and the public provision of Table 12.3 Regression results (1)

Dependent variables

Poverty rate

(2)

(3)

(4)

(5)

Poverty rate

Poverty rate

Disposable income of bottom quintile

Disposable income of bottom quintile

−0.09 (−0.42)

−0.07 (−0.30)

0.16 (0.23)

−0.01 (−0.08)

Independent variables Public social outlays

−0.63** (−4.61)

Private social outlays

−0.19 (−0.76)

Direct public social outlays

−0.39** (−2.76)

Active labor market expenditures

−4.47** (−3.25)

0.09* (2.00) −4.08** (−2.69)

1.56** (3.29)

1.41** (3.04)

Cash transfers

−0.34* (−2.09)

0.51 (0.97)

Public services

−0.57* (−1.86)

0.21** (2.41)

Notes: t-Statistics listed in parentheses. * Indicates significance at 10 percent level. ** Indicates significance at 5 percent level.

396

Jeffrey D. Sachs

services, both show up as reducing the poverty rate (regression 3). When we regress the disposable income share of the bottom quintile on public and private outlays, we again find that the public outlays and active labor market policies boost the share of income in the bottom quintile, while private social outlays have no statistically significant effect on the income share of the poorest 20 percent (regression 4). We also find that the direct provision of public services has a larger effect than cash transfers on the income of the bottom quintile (regression 5). Figure 12.4a shows the added-variable plot of the poverty rate against the share of public sector outlays in GDP, based on regression 1 in table 12.3. The strong negative effect of public social outlays on poverty is clearly evident. Figure 12.4b shows the same for private social outlays, which clearly have no strong effect in reducing poverty rates. 12.5

Labor Market Outcomes of the Nordic Social Welfare Policies

We now turn to the possible adverse consequences of the high levels of social expenditure in the Nordic states. This section examines the patterns of employment, unemployment, and hours of work per employee. Later sections examine other possible adverse effects, on the standard of living, fiscal policy, and other indicators of economic performance and well-being. In general, we find no evidence for any significant adverse economic effects of high levels of public social outlays and government revenue collection. The employment rate is the number of employed people divided by the population of working age, taken to be ages 15 to 64. The surprising fact, to be seen in table 12.4, is that the Nordic countries have maintained an even higher employment rate than the English-speaking countries.8 The ES countries in turn have a higher employment rate than the EC countries, which is as expected given the chronic concerns about labor market rigidity and high unemployment in continental Europe. The main message is not to lump the Nordic states and the continental states with regard to employment outcomes. The very high employment rate of the Nordic states would appear to reflect two important facts about Nordic economic policies. First, especially during the past ten years, social support for the working-age population has been tied to active labor market policies whereby recipients of social support are required to seek employment. Second, the state has acted as an important employer of last resort. Many older,

Revisiting the Nordic Model

397

(a)

e( pov2000 | X )

5

0

-5 –10

–5

0

5

10

e( psocexp | X ) coef = –0.62530884, se = 0.13564464, t = –4.61

(b) 4

e( pov2000 | X )

2

0

–2

–4 –4

–2

0

2

4

6

e( prsocexp | X ) coef = –0.19458068, se = 0.25752539, t = –0.76

Figure 12.4 Share of public and private sector outlays in GDP: (a) strong effect of public sector outlays on reducing poverty rate; (b) weak effect of private sector outlays on reducing poverty rate

398

Jeffrey D. Sachs

Table 12.4 Employment Region

Employment rate (share of working age population)

English-speaking

72.4

Europe

68.9

Nordic

73.7

Source: OECD Factbook, 2006.

lower skilled, and partially disabled workers are employed in the public sector, and especially by local governments, in the provision of public-sector social services, including day care, health care, and support for the disabled population.9 The rise of public employment is exemplified by the case of Sweden, where the public employment swelled from around 22 percent of the employed in 1970 to around 38 percent of the employed today, with most of that increase taking place in local government positions in the social sector. By contrast, the US public employment rose only slightly, from around 22 percent of the employed in 1970, roughly the same as in Sweden, to around 26 percent today. Unemployment rates are shown in figure 12.5a and b. The European continental states have had chronically high unemployment rates, as shown. The Nordic unemployment rates are lower than in the EC countries, and slightly higher than in the ES countries. This is due solely to Finland, where unemployment rates remain relatively high. The unemployment rates in Denmark, Norway, and Sweden are comparable to the low rates in the ES countries. The high EC unemployment rates are generally attributed to relatively fixed real wages, often supported by generous unemployment benefits but without work requirements. Unlike the Nordic states, its appears that the EC states are less rigorous in enforcing work requirements on hard to employ workers, and are less willing or able to hire such workers directly into the public sector.10 The main place where the labor market outcomes differ between the lower taxed ES and the higher taxed EC and Nordic states is in hours of work per employee. The English speaking countries work roughly 200 hours per year more than their counterparts, and this does seem to be statistically related to the level of social expenditures and taxation. The long working hours of the United States are not an obvious outlier

Revisiting the Nordic Model

399

(a)

Percent of labor force

8

6

4

2

0

English

Nordic

Europe

(b)

10

Percent of labor force

8

6

4

2

0

NZ NO IR NE UK AR DE AU US SW CA BE IT

FI GE FR

Figure 12.5 Standardized unemployment rate: (a) high-income country groups; (b) individual highincome countries compared. Nordic economies are Denmark, Finland, Norway, Sweden; European continental countries are Austria, Belgium, France, Germany, Italy, Netherlands; English-speaking countries are Australia, Canada, Ireland, New Zealand, United Kingdom, United States

400

Jeffrey D. Sachs

compared with the other English-speaking countries; the pattern of greater working hours is common throughout the English-speaking countries.11 The difference in working hours results mainly from differences in three areas: paid vacation time—which averages around six weeks in Europe compared with two weeks in the United States—paid maternity leave—which is typically twelve weeks in Europe and without public guarantee in the United States—and hours of work per week— which are typically under forty hours in most of Europe. These differences are probably due in significant part to the higher effect rates of taxation on labor income in Europe, though other factors are also important (unionization, public policy views on maternity, etc.).12 The welfare consequences are complex. On the one hand, neoclassical theory suggests a straightforward deadweight loss to the shift out of labor in to leisure as the result of high rates of labor taxation. On the other hand, the provision of paid maternity leave may, in the same manner as mandatory primary education, provide a social guarantee of well-being for newborns. Similarly societywide choices on vacation time and weekly working hours can prevent the “rat race” syndrome of overwork that results from a zero-sum (and hence inefficient) competition among workers with each trying to earn a higher income than their peers. Finally socially agreed reductions in work hours and increases in vacation time may facilitate a socially optimal coordination of increased leisure time in a way that market forces alone may be unable to accomplish, when the value of leisure time of each individual depends on the leisure time taken by others. 12.6 Social Spending and Per capita Income As a matter of simple averages, the Nordic countries have a higher average per capita GDP than the English-speaking countries, with the European continent countries coming in third. Norway is the highest income country in the sample, followed by the United States. One could rightly argue that Norway’s hydrocarbon income make it something of a special case, giving an artificial boost to the income of the Nordic countries, but one could make similar cases about the especially favorable endowments of many of the countries in the sample, including the United States. What is true, at the least, is that the per capita income of the Nordic countries is not in any obvious sense “crippled” by the high social expenditure, and per cepita income may even be

Revisiting the Nordic Model

401

higher than lower taxed English-speaking economies. It is interesting to note that if both Norway and the United States are removed from the sample as high-income special cases, then the remaining three Nordic countries still have a slightly higher average per capita income ($47,008) than the remaining five English-speaking countries ($46,279), though the difference is not statistically significant. Despite almost 20 percentage points of difference in tax take as a share of GDP, GDP per working-age population is essentially the same! The standard measures of GDP per person perhaps skew the deck against economic well-being in the Nordic countries, since GDP does not count the value of the greater leisure time. As a rough measure, let us calculate a “full GDP,” inclusive of the value of leisure. As in Gordon (2006), we put the value of an hour of leisure time as equal to two-thirds of the hourly labor compensation, which we estimate as equal to threefourths of the value of GDP per labor hour.13 These assumptions give us the value of an hour of leisure as equal to one-half of GDP per labor hour. The total value of leisure time in the economy is then equal to 0.5 times GDP times (leisure hours/labor hours). This sum is added to total GDP to get the full GDP, which is then divided by the working-age population. This correction, of course, widens the lead of the Nordic countries. Indeed, even the European Continental countries now outpace the English-speaking countries in full GDP per working-age population. The United States still ranks above almost all the other countries whether the measure is GDP or full GDP per working-age population. Indeed, the United States ranks second only to Norway. One might argue therefore that the US “system” is more productive than the alternatives. This does not square easily with the fact that other relatively liberal economies (e.g., Australia, Canada, New Zealand, and United Kingdom) do not obviously outpace their counterparts.14 The US advantage, just as Norway’s advantage, may have to do with a combination of extraordinary resource endowments (land, energy, minerals, etc.) as well as its economic system. 12.7 Per capita Income of the Rich and Poor Average per capita income should not be the beginning and end of welfare comparisons. With very different patterns of income distribution, it is important to compare the average income levels at various points of the income distribution as well. A simple way to do this, without access to extensive and comparable household data sets, is to

402

Jeffrey D. Sachs

Table 12.5 Income distribution Region

Bottom quintile

Middle quintiles

Top quintile

English-speaking

7.3

54.1

38.6

Europe

8.4

54.8

36.8

Nordic

9.6

56

34.4

Source: Forster and Mira d’Ercole (2005). Note: Most data cited here are for 2000.

approximate per capita income for different quintiles of the population. Household disposable income for the top 20 percent, middle 60 percent, and bottom 20 percent of the households is shown in table 12.5 for the three groups of countries, as usual taking simple averages of the income shares of the countries in the respective categories. The per capita income level of the bottom quintile is approximately equal to the economywide average per capita income level multiplied by income share of the bottom quintile divided by 20. Thus, in the English-speaking countries, with an average per capita income per working-age population of $48,456, the average per capita income of the bottom quintile is approximately $17,686 (=7.3/20 × $48,456). (The approximation arises, among other things, because the income quintiles are for all ages, not only working ages.) The bottom quintile comes out way ahead in the Nordic countries. The average per capita income of the working age population in the bottom quintile, estimated as just described, comes out to be $24,465 for the Nordic countries, $19,066 for the European continental countries, and just $17,553 for the English-speaking countries. Once again we see that the English-speaking countries are indeed tough places to be poor! Moreover the United States, with the second-highest GDP per working-age population, comes out in the middle of the pack in terms of per capita income of the bottom 20 percent ($18,395), behind all four of the Nordic countries. 12.8 Are There Other Adverse Consequences of the Social Welfare State? Here are some additional alleged disadvantages of the social welfare state, at least as argued by US supply-siders and by the laissez faire tradition.

Revisiting the Nordic Model

403

The Hayek–Friedman freedom hypothesis Hayek raised the idea that large-scale state involvement in the economy would be the road to serfdom, namely to the diminution and even collapse of political liberties.15 Friedman has written about taxation as coercion and therefore antithetical to a free society. Fiscal distress It has been hypothesized that the social welfare state would lead to chronic fiscal crisis, evidenced by chronic fiscal deficits and high levels of public indebtedness relative to GDP. Household saving rate It is argued that high rates of taxation reduce the incentives to save and invest. Innovation It has been asserted that high tax rates will stifle entrepreneurship and innovation, with a consequent decline in the rate of total factor productivity growth, patenting, and other outcomes of innovative activity. Let us consider the basic evidence. The Hayek–Friedman hypothesis has been disproved by experience. All the Nordic countries remain vibrant democracies, and all score very well on every widely known indicator of governance. As shown in table 12.6, the countries tend to be less corrupt on average (as scored by the Transparency International corruption perceptions index) than the other countries. The English-speaking countries rank second, and the European continental countries rank third. The Nordic countries similarly score better than the other groups on various measures of Table 12.6 Governance indicators Global Competitiveness Reportb

Region

Transparency International CPIa

Overall score

Contracts

Public institutions

English-speaking

8.4

13.5

13.2

12.5

Europe

7.6

25.8

23.7

21.5

Nordic

9.3

4.3

7.5

7.5

Source: Transparency International (2005); GCR from Lopez-Carlos et al. (2006). Notes: The TI scores are up to 10 (with higher signifying less corrupt). The Global Competitiveness Report numbers indicate the ranking of each country in the group, with rank equal to 1 being the best score, so that lower numbers signify better performance. a. Index is scored from 0 to 10 (higher score indicates less corruption). Values cited are for 2005. b. Average ranking. See Lopez-Carlos et al. (2006).

404

Jeffrey D. Sachs

public institutions in the World Economic Forum’s Global Competitiveness Report. On average the Nordic countries tend to have the best governance scores or ranks, followed by the English-speaking countries, and then the continental countries, though the differences in means are not significant. There is no evidence that higher levels of taxation as a share of GDP or higher social outlays as a share of GDP are conducive to higher corruption or weaker property rights. The fiscal distress hypothesis is also decisively rejected by the evidence. Table 12.7 shows why this is so. The countries are ordered, lowest to highest, regarding the net liabilities of the public sector as a share of GDP. Amazingly, three of the four Nordic countries have positive government financial net worth (i.e., negative net liabilities), and the fourth, Denmark has only small net liabilities as a share of GDP (12 Table 12.7 Fiscal distress Country

Net government liabilities (percentage of GDP, 2004)

Current fiscal imbalance (percentage of GDP, 2004)

Norway

−111.2

11.7

Finland

−47.5

1.9

Sweden

−5.7

1.6

Australia

0.7

1.3

New Zealand

4.8

5.9

Ireland

12.0

1.6

Denmark

12.1

1.7

Switzerland

25.4

−1.1

Canada

31.1

0.7

Spain

33.6

−0.2

United Kingdom

36.9

−3.3

Netherlands

37.8

−2.1

Portugal

40.3

−3.2

Austria

41.7

−1.2

France

44.4

−3.7

United States

45.1

−4.7

Germany

54.5

−3.7

Japan

82.2

−6.3

Belgium

86.5

−0.1

Italy

97.7

−3.5

Greece

97.9

−6.9

Source: OECD Economic Outlook, 2006.

Revisiting the Nordic Model

405

percent). The English-speaking countries, generally speaking are in the middle of the pack, while the European continental countries have very large net financial liabilities in many cases, notably Belgium and Italy. The same holds true regarding current fiscal imbalances, shown in the second column of the table. All four of the Nordic states had budget surpluses in 2004, while all of the continental states have deficits. The English-speaking countries are again in the middle, with four out of six in surplus and two in deficit (with the United States running the largest deficit as a share of GDP among the English-speaking countries).16 It is interesting that while both the Nordic countries and the European continental countries have high levels of social spending, only the continental countries have chronic fiscal distress. The Nordic states have matched their social ambitions with their tax collections, while the European continental countries have not. Social spending is nearly the same across the two groups, but the EC countries have government receipts to GDP that average 17 percentage points lower than those in the Nordic countries.17 It seems that the EC countries are conflicted about where to turn: they are pulled to high levels of social spending, but then pressed for tax cuts. With regard to household saving the evidence runs overwhelmingly against the hypothesis that high rates of taxation and social outlays diminish the rate of household saving. While there are no doubt some important measurement issues here, to get a precise comparative account, the overwhelming evidence is that the English-speaking countries have the lowest household saving rates, while both the Nordic and EC countries have similar and higher net household saving rates. It would be therefore interesting to learn whether the free-wheeling competitive environment of the English-speaking countries actually contributes culturally to the credit-card, debt-ridden society. But this is not the place to decide such a weighty issue; we can seek only to refute the common claim that tax cutting is somehow conducive to higher household saving rates. One of the striking facts about the Nordic states is their very high rate of technological excellence. Sweden and Finland, of course, prosper on their ICT sectors, notably led by Ericsson and Nokia, respectively. Table 12.8 shows the ranking of countries in the World Economic Forum Technology Index, which is built on evidence of innovation, R&D, and mobilization of information and communications technology. The Nordic countries score exceptionally high on the technology index.

406

Jeffrey D. Sachs

Table 12.8 Technology and R&D Region

World Economic Forum Technology Indexa

English-speaking

16.2

Europe

24

Nordic

6

Source: Lopez-Carlos et al. (2006). a. Average ranking.

They are heavy investors in both R&D and higher education, and they have very high rates of patents per capita as well.18 12.9 Is the Nordic Model Transferable and Sustainable? The Nordic commitment to the social welfare state is long-standing, and dates back at least to the post–World War II political scene. Social Democrats have governed in northern Europe for a preponderance of years since 1950. Social spending as a percentage of GDP has been relatively high in the Nordic countries for at least forty years. In this sense, there is a long-standing Nordic model of social democracy. We have shown that the Nordic model of high social spending has not led to long-term political or economic deterioration. The Nordic countries tend to outperform most of the other countries on most economic and governance indicators. There are still important questions regarding the transferability of the Nordic model. There is probably little room for doubt that Nordic ethnic homogeneity has been an important enabling social factor in the success of the social welfare state. In a wonderful series of articles, Alesina and colleagues have shown that social spending tends to be highest where social and racial cleavages are the smallest. This is true across US states and apparently across countries as well. White Americans living in states with higher proportions of AfricanAmericans, for example, seem to be much less likely to support high levels of social spending (Alesina et al. 2001). The authors summarize matters as follows: Racial discord plays a critical role in determining beliefs about the poor. Since minorities are highly over-represented amongst the poorest Americans, any income-based redistribution measures will redistribute particularly to

Revisiting the Nordic Model

407

minorities. The opponents of redistribution have regularly used race based rhetoric to fight left-wing policies. Across countries, racial fragmentation is a powerful predictor of redistribution. Within the U.S., race is the single most important predictor of support for welfare. America’s troubled race relations are clearly a major reason for the absence of an American welfare state. (2001: 4)

The Nordic model may itself come under stress if inward migration into these countries, and a relatively high fertility rates of immigrant populations, lead to a sharp rise in nonnative populations in the Nordic countries. This important issue, however, is well beyond the scope of this chapter. Globalization too could undermine the ability to levy high tax rates on a sustainable basis. The Nordic countries have wisely kept the rate of taxation on capital relatively low, garnering the bulk of tax revenues through the value-added tax, and other taxes on goods, services, and wage income. Tax policy favorable toward capital has enabled the Nordic states to combine high rates of taxation with internationally open capital markets. Still, we can wonder whether increased labor migration and the high share of taxation that falls on human capital, if not on corporate capital, will eventually undermine the ability of the Nordic states to collect the requisite levels of GNP needed to sustain the social welfare state. Such doubts have been around for a long time, and have not really materialized. Yet increasing globalization could still undermine the tax base of the Nordic states. 12.10

Conclusions

The comparisons among Nordic, European continental, and Englishspeaking countries offered in this chapter made no attempt to formulate precise models of the Nordic social welfare state, or to estimate precise parameters regarding labor force participation, hours worked, poverty rates, and other critical variables. The aim was much simpler: to show that highly ideological claims made against the social welfare states of Scandinavia are simply off the mark. The “euro-pessimism” in many parts of the continental Europe, and the claim that AngloSaxon liberalization is crucial to economic well-being, is belied by the persistent high performance of the Nordic economies. For decades these economies have maintained high levels of GNP per worker, low rates of poverty, high rates of innovation, and high levels of labor force participation.

408

Jeffrey D. Sachs

The continental European countries seem to be caught, in some ways, between two urges: the social welfare impulse of northern Europe, and the liberalizing influence of the Anglo-Saxon model. The result may actually be the worst of both worlds. The continental European countries spend like the Nordic states but do not generate the tax revenues to support that high level of social spending. They instead run chronic fiscal deficits. Moreover their labor market policies seem less geared to supporting the labor force participation of the lessskilled workers, as in the Nordic countries, and more geared toward simpler (and inefficient) labor market transfers, which raise unemployment rather than employ the lower skilled workers. The evidence would at least suggest that countries such as France and Germany have a true choice: to raise tax collections further in order to sustain the social welfare state, or to cut spending in order to emulate the Anglo-Saxon model. Both paths seem viable, not only the liberalizing course favored by many of euro-pessimists. Of course, the longterm viability of the Nordic model (and of all other models) is open to question. The Swedish voters have just replaced the long-standing Social Democrats with a new Moderate government, one that is intent at least at trimming some aspects of the social democratic state. Still, to paraphrase Mark Twain, rumors of the death of the social welfare state have been greatly exaggerated. Notes This chapter was originally prepared for the CESifo/Center for Capitalism and Society Venice Summer Institute, Venice International University, July 21–22, 2006. The author would like to thank Samuel Freeman for excellent research assistance. 1. Netherlands in fact lies somewhere between the four Nordic states and the rest of the continental countries in social expenditures, and might plausibly have been grouped with the Nordic states. 2. OECD, Economic Outlook, no. 79. Numbers cited are for 2004. 3. OECD, Economic Outlook, no. 79. Numbers cited are for 2004. 4. OECD, Social Expenditure Database 1980–2001, 2004 (www.oecd.org/els/social/ expenditure). Values cited are for 2001. 5. Numbers are for 2001 and are based on Adema and Ladaique (2005). 6. OECD (2004), Social Expenditure Database 1980–2001 (www.oecd.org/els/social/ expenditure). Numbers cited are for 2001. 7. Data on all three indicators are taken from Förster and Mira D’Ercole (2005). 8. OECD Factbook 2006: Economic, Environmental and Social Statistics (OECD 2006b).

Revisiting the Nordic Model

409

9. See, for example, “Disability programs in need of reform,” OECD Policy Brief, March 2003. 10. Data on unemployment rates from OECD in Figures, 2005 edition. Accessible online at http://www.oecd.org/infigures/. Numbers cited are for 2004. 11. For data on average working hours, see OECD Productivity Database, January 2006. 12. For a discussion of the differences in working hours, vacation, and maternity leave between Europe and the United States, see Alesina, Glaeser, and Sacerdote (2005), Meyers and Gornik (2004), United Nations (2006), and Expedia.com (2006). 13. Suppose that the labor share of GDP is 0.75. Then wL = 0.75 × GDP, or w = 0.75 × GDP/L, where L is total labor hours in the economy. 14. For data on labor productivity, see OECD Productivity Database, 2006 (OECD 2006c). 15. In the original edition of The Road to Serfdom, the argument centered mainly on government ownership of industry and industrial planning. By 1976, however, von Hayek wrote the following in a forward to a reprinting of the book: “[S]ocialism has come to mean chiefly the extensive redistribution of incomes through taxation and the institutions of the welfare state. In [this] kind of socialism the [totalitarian] effects I discuss in this book are brought about more slowly, indirectly, and imperfectly. I believe that the ultimate outcome tends to be very much the same. ...” (pp. xx-xxi) 16. Data on net financial liability from OECD Economic Outlook 79 (OECD 2005). Numbers cited are for 2004. 17. The simple average of government receipts as a percentage of GDP in 2004 was 46.7 for the EC countries, and 56.4 percent for the Nordic countries. Data on government receipts available from OECD Economic Outlook, No. 79 (OECD 2004). 18. Data on R&D expenditure, investment in knowledge, and tertiary attainment are available from the OECD Factbook 2006: Economic and Social Statistics (OECD 2006a).

References Adema, W., and M. Ladaique. 2005. Net Social Expenditure, 2005 edition: More Comprehensive Measures of Social Support. Social, Employment and Migration Working Paper 29. OECD, Paris Alesina, A., and G.-M. Angeletos. 2003. Fairness and redistribution: U.S. versus Europe. Working paper 9502. NBER, Cambridge, MA. Available online at http://www.nber.org/ papers/w9502. Alesina, A., R. Di Tella, and R. MacCulloch. 2001. Inequality and happiness: Are Europeans and Americans different? Working paper 8198. NBER, Cambridge, MA. Available online at http://www.nber.org/papers/w8198. Alesina, A., E. Glaeser, and B. Sacerdote. 2005. Work and leisure in the U.S. and Europe: Why so different? Working paper 11278. NBER, Cambridge, MA. Available online at http://www.nber.org/papers/w11278. Alesina, A., E. Glaeser, and B. Sacerdote. 2001. Why doesn’t the US have a European-style welfare system? Working paper 8524. NBER, Cambridge, MA. Available online at http:// www.nber.org/papers/w8524.

410

Jeffrey D. Sachs

Baker, D., A. Glyn, D. Howell, and J. Schmitt. 2004. Unemployment and labor market institutions: The failure of the empirical case for deregulation. September. Available at http://www.ilo.org/wcmsp5/groups/public/—dgreports/—integration/documents/ publication/wcms_079135.pdf. Boarini, R., A. Johansson, and M. Mira d’Ercole. 2006. Alternative measures of well-being. Social, Employment and Migration Working Paper 33. OECD, Paris. Daugaard, S. 2002. Enhancing expenditure control with a decentralised public sector in Denmark. Working paper 320. Economics Department, OECD, Paris. Engström, P., and B. Holmlund. 2006. Tax evasion and self-employment in a high-tax country: Evidence from Sweden. Working paper 12. Department of Economics, Uppsala University. Available online at http://www.nek.uu.se. Expedia.com. 2006. International Vacation Survey. Forslund, A., D. Froberg, and L. Lindqvist. 2004. The Swedish Activity Guarantee. Social, Employment and Migration Working Paper 16. OECD, Paris. Förster, M., and M. Mira d’Ercole. 2005. Income distribution and poverty in OECD countries in the second half of the 1990s. Social, Employment and Migration Working Paper 22. OECD, Paris. Gordon, R. J. 2006. Issues in the comparison of welfare between Europe and the United States. Prepared for the Venice Summer Institute, organized by CES-Ifo and the Center on Capitalism and Society, Venice International University, San Servolo, July 21–22. Hayek, F. A. von. 1944. The Road to Serfdom. Chicago: University of Chicago Press. Heckman, J. J., and B. Jacobs. 2006. Policies to create and destroy human capital in Europe. Prepared for the Venice Summer Institute, organized by CES-Ifo and the Center on Capitalism and Society, Venice International University, San Servolo, July 21–22. Howell, D. R., D. Baker, A. Glyn, and J. Schmitt. 2007. Are protective labor market institutions really at the root of unemployment? A critical perspective on the statistical evidence. Capitalism and Society 2 (1). Available at http://www.bepress.com/cas/vol2/iss1/art1/. Immervoll, H., P. Marianna, and M. Mira d’Ercole. 2004. Benefit coverage rates and household typologies: Scope and limitations of tax-benefit indicators. Social, Employment and Migration Working Paper 20. OECD, Paris. Lopez-Claros, A., M. E. Porter, X. Sala-i-Martin, and K. Schwab. 2006. Global competitiveness report 2006–2007. World Economic Forum. Available at: http://us.macmillan.com/ theglobalcompetitivenessreport20062007. Miles, M. A., K. R. Holmes, M. A. O’Grady, A. I. Eiras, A. B. Kim. 2006. 2006 Index of Economic Freedom. Washington, DC: Heritage Foundation. Myers, M. K., and J. C. Gornik. 2004. The European model: What we can learn from how other nations support families that work. The American Prospect (November): A21–22. OECD. 2006a. OECD Economic Outlook, no. 79, May. OECD. 2006b. OECD Factbook 2006: Economic and Social Statistics. OECD. 2006c. OECD Productivity Database, 2006. January.

Revisiting the Nordic Model

411

OECD. 2005. OECD in Figures, 2005 edition. Available online at http://www.oecd.org/ infigures/. OECD. 2004. OECD Social Expenditure Database 1980–2001. Available online at http:// www.oecd.org/els/social/expenditure. OECD. 2003. Disability programs in need of reform. OECD Policy Brief, March. Pearson, M., and J. Martin. 2005. Should we extend the role of private social expenditure? Social, Employment and Migration Working Paper 23. OECD, Paris. Pissarides, C. A. 2006. Unemployment and hours of work: The North Atlantic divide revisited. Prepared for the Venice Summer Institute, organized by CES-Ifo and the Center on Capitalism and Society, Venice International University, San Servolo, July 21–22. Roseveare, D. 2002. Enhancing the effectiveness of public expenditure in Sweden. Working paper 345. Economics Department. Sinn, H.-W. 2006. The welfare state and the forces of globalization. Prepared for the Venice Summer Institute, organized by CES-Ifo and the Center on Capitalism and Society, Venice International University, San Servolo, July 21–22. Transparency International. 2005. International Corruption Perceptions Index. Berlin. United Nations Statistics Division, Demographic and Social Statistics. 2005. Statistics and Indicators on Women and Men. Available online at http://unstats.un.org/unsd/ Demographic/products/indwmn/ww2005/tab5c.htm (accessed July 24, 2006).

13

The Welfare State and the Forces of Globalization Hans-Werner Sinn

13.1 Globalization, the Fall of the Iron Curtain, and Factor Price Convergence The fall of the Iron Curtain was an unparalleled shock to Western economies in general and Europe in particular. Suddenly the 28 percent of humankind who previously had lived in the Communist states began to participate in the market game in the same way as the 15 percent OECD people had done before. And, coincidentally, India also decided to participate, which increased the new entrants into the world market economy to 45 percent. While this development is likely to produce gains from trade for most countries involved and substantially raise the standard of living of a substantial part of mankind, it is not without problems for the West. The main difficulty is the process of factor price convergence. Gains from trade always accompany the process of price and wage convergence, because the original differences in prices and wages initiate the arbitrage and specialization processes that are the reason for these gains. Thus there will always be groups in society that lose, despite the gains to be achieved in the aggregate. In theory, if we assume no frictions and transportation costs, the aggregate gains from trade reach a maximum when all wages and prices are equal and arbitrage ceases to be profitable. In practice, given the frictions, prices and wages will not and should not become identical. However, given the huge differences that we see today, they will converge over time. For the next half century it will hardly make any difference whether the wage convergence between China and the West eventually lead to equal wages or to wage gaps of, say, 30 percent as are found among the countries of the West.

414

Hans-Werner Sinn

In euros Denmark West Germany Norway Switzerland Netherlands Sweden Austria France United States United Kingdom Ireland Japan Italy Spain Korea Portugal Hungary Czech Republic Estonia Slovakia Poland Lithuania Latvia Romania Bulgaria China

28.14 27.60 27.31 25.31 23.74 23.32 21.50 20.74 19.91 19.89 18.79 17.95 17.24 16.59 10.00 7.21 4.53 4.49 3.88 3.61 3.29 3.03 2.52 1.78 1.45 1.10

0

5

10

15

20

25

30

Figure 13.1 Hourly labor costs in 2004. The asterisk indicates average labor costs in the manufacturing industry.

Figure 13.1 shows how far the ex-Communist countries will have to go until their wages match those of the West. Currently the average wage cost per hour of the eight east European EU accession countries is 13 percent of the west German wage cost, for example, and the Chinese wage is only 1/27. Closing the wage gap will be a positive experience for the ex-Communist countries, but the adjustment processes will be severe and painful for the West. Some policy makers have downplayed the difficulties by pointing out how easy it was for the European Union to manage Western enlargement by Spain and Portugal in the mid-1980s. However, these developments are not comparable, as the two countries were much more advanced than the eastern European countries. Instead of the 13 percent ratio of the west German wage, their wages stood at 50 percent at the time of EU accession. How long will convergence take? Estimates vary. If one believes in the convergence figure of 2 percent that was determined by Barro and

The Welfare State and the Forces of Globalization

415

Sala-i-Martin (1995), implying a half time of 35 years, east European wages will have reached 50 percent of German wages by 2035. However, during the last decades, west European convergence has just been 1.1 percent p.a., implying a half time of 63 years. If east-west convergence were to proceed at the same speed as the internal convergence in the western part of the European Union, the wages of the east European member countries would reach 50 percent in western Germany’s wages by 2054. 13.2 The Forces of Factor Price Convergence There are various economic forces bringing about factor price convergence. The first is the spillover of technological knowledge. The transfer of scientific knowledge through the scholarly media as well as the transfer through observation and imitation can be powerful equalizing forces. It is true that some of this transfer can be prevented by patents and other intellectual property rights. However, as patents usually expire after thirty years, the still backward countries can immediately leapfrog to the state of western knowledge of thirty years ago. Moreover most of the technological knowledge of this world is not protected, and much knowledge is too diffuse to be patented. When a European company opens a factory in China, there is little to prevent the Chinese from building a nearly identical plant on the green field next door. The second force is capital flows. Capital moves from high-wage Western countries to low-wage Eastern countries because the rate of return is higher there. The capital creates jobs in the East and increases the demand for labor there, which raises the wage, and the reverse process takes place in the West. Thus wages converge. Germany is a country that is currently very strongly affected by this process. At a net investment share of only 2.9 percent of net domestic product (NDP), it currently ranks lowest among all OECD countries. In 2005 net investment was only 60 billion euros, or 40 percent of aggregate savings, which in turn stood at 150 billion euros. Sixty percent of German savings, or 90 billion euros, were sent abroad as net capital exports. The third force is migration. As people migrate from low-wage to high-wage countries, they make labor scarcer in the former and more abundant in the latter, which tends to reduce the international wage differences. In Europe, migration has already been strong in the

416

Hans-Werner Sinn

past decades and, in all likelihood, it will continue to be strong in the future. Figure 13.2 gives an overview of the shares of the foreign-born population in those countries for which the OECD has provided data plus Germany, for which the Federal Statistical Office published such data for the first time in June 2006. It shows that the Netherlands, Austria, Sweden, and Germany have immigration figures that resemble and, in the latter two cases, even exceed those of the United States. Obviously western Europe is currently facing a mass immigration wave of historical dimensions. To properly interpret the immigration figures, note that they cannot be equated with the shares of foreigners, since the former also reflect the countries’ naturalization policies. Where available, the shares of foreigners have also been indicated in the diagram. In the case of the Netherlands, which has a very liberal naturalization policy, this share is only 4.3 percent; in the case of Germany it is 8.9 percent. The figures also cannot be equated with the share of the population with a migration background, as the children of the immigrants born in 25

20

18.2

15 10.4

11.1

11.0

11.5

12.6

10

Au

St at es * Sw ed en G er m an y* C an ad N a ew Ze al an d

st ria U

st ra lia

5.3

Au

k ar m

un

ga

an H

nl Fi

ay N et he rla nd s

1.3

8.9

4.3

N or w

2.0

5.0

ry

3.0

d

0

2.8

en

5

9.4

6.9

ni te d

6.0

D

Percent of total population

21.5 19.5

Figure 13.2 Stock of foreign-born population, 2000 to 2005. Dotted areas indicate share of foreigners, where data were available: most countries for 2001, USA for 2000, Germany for 2005. (sources: OECD 2004; Trends in International Migration: Sopemi, 2003 ed.; Statistisches Bundesamt 2006; Leben in Deutschland—Ergebnisse des Mikrozensus 2005; Eurostat, Population and social conditions; Statistcs in Focus 8/2006)

The Welfare State and the Forces of Globalization

417

the host country are not included. In the case of Germany, immigrants— including their children—currently account for 18.5 percent of its population. A fourth force of factor price convergence is specialization. As the capital-rich West encounters new trade possibilities with the low-wage countries of the East, it retreats from labor-intensive production processes and specializes in capital-intensive ones, where labor typically is unskilled labor and capital includes human capital in the sense of skilled labor. Conversely, the East, which is rich in labor, specializes in labor-intensive production. Both specialization processes also tend to reduce the wage gaps, as the demand for unskilled labor in the West falls while it rises in the East. Specialization toward capital-intensive sectors can take place in the horizontal and in the vertical direction. Examples of horizontal specialization are the retreat from textile and leather industries and the expansion of the high-tech areas like chemistry or automobiles. Examples of vertical specialization are outsourcing and offshoring activities. Outsourcing and offshoring have become particularly strong since the mid-1990s because that was the time when the eastern European EU accession countries had overcome their transformation crises and the decisions on EU membership were made. West European firms have increasingly cut off the more labor-intensive upstream parts of their production chains, shifting them to low-wage countries in the East. Either they established plants there (offshoring) or they gave up part of their own intermediate production, buying the respective parts instead from other companies located in low-wage countries (international outsourcing). Western firms thereby created jobs in the East and destroyed them in the West, pulling up wages there and depressing them here, thus contributing to the process of factor price convergence. The shift to outsourcing has been observable in most EU countries. Practically everywhere has production depth declined in the sense that the share of the manufacturing sector’s value added in its own output has fallen. Due to its geographical and cultural proximity to the East and its own high wages, Germany has been affected particularly strongly. There is hardly any German car whose domestic share in production cost reaches 50 percent. Even exports as such, including the exports of nonmanufactured goods, increasingly consist of imports. From 1991 to 2001 the share of imported intermediate goods in exports rose from 27

418

Hans-Werner Sinn

to 39 percent, and at the margin this share is already 53 percent. From each additional euro that German companies earn in exports, they need 53 cents to buy the additional imports needed to produce these exports. Germany is developing toward what I have called a bazaar economy (see Statistisches Bundesamt 2004; Sinn 2004, 2006). The labor market implications of these and other flight reactions have been enormous, as can be seen by the sharp decline in manufacturing employment since the collapse of Communism. From 1991 to 2003 manufacturing employment declined by 11 percent in the Netherlands, about 12 percent in France, 23 percent in Great Britain, 26 percent in Japan, and 27 percent in Germany, the dubious OECD champion in this regard. 13.3 Gains from Trade and Specialization: Theory and Reality Despite the obvious hardship resulting from job losses, these losses cannot alone be considered a sign of welfare losses for the West. After all, they could be the necessary counterpart of a gradual process of sector shift that normally is associated with gains from trade, specialization, and an improved international division of labor. There could be an equal number of jobs created elsewhere in the economy, compensating for the losses. This at least is the typical economics textbook view. Unfortunately, however, reality does not always fit the textbook view. One example of where it definitely does not fit is Germany. As shown in table 13.1, full-time equivalent employment in German manufacturing declined by 1.21 million in the ten years from 1995 to 2005. However, in the whole rest of the economy, including services, construction and high-tech areas, no jobs were created in net terms that could have compensated for these losses. On the contrary, even these sectors lost 150,000 jobs in full-time equivalents. So where did the manufacturing workers go? There is only one possibility left. They Table 13.1 Improvement in the division of labor? The German case, 1995 to 2005, million full-time equivalents Manufacturing industry

−1.21

Rest of the economy

−0.15

Into nonemployment

1.36

The Welfare State and the Forces of Globalization

419

went to the welfare state, into state-financed unemployment. That was the improvement in the division of labor à l’Allemande. The dirty industrial jobs were abolished, but nothing was created in exchange! This development pattern is so remote from what the textbook predicts that it would be absurd to interpret it as an improvement in the international division of labor. It rather seems to be accompanied by welfare losses. As more and more people stop working, they stop making contributions to GDP and national income. Small wonder that Germany was the European growth laggard in the period 1995 to 2005—next to Italy. While no comparable data for other European countries are available, it may be suspected that the problems described for Germany may also be occurring there. After all, even western Europe on average had a miserable growth performance during the last ten years. There was no comparable region and no continent in the whole world whose growth performance was similarly disappointing. From 1995 to 2004 the world economy grew by 30 percent, east Asia grew by 87 percent, the United States by 34 percent, Latin America by 26 percent, and even Africa grew by 31 percent, while the old EU countries grew only by 23 percent. 13.4

Rigid Wages and the Welfare State

What went wrong? That is the obvious question one must ask after realizing the problematic reactions of the labor market. The answer is that the European labor market is rigid, impeding the textbook-type wage adjustments, and with rigid wages, unemployment results. The rigidity stems, to some extent, from the strong unions that have not allowed wage reductions despite the fact that the participation of the ex-Communist countries has reduced the equilibrium wage of unskilled labor in Europe. To another extent, it stems from the welfare state that offers generous wage replacement incomes. All western European countries have social systems that are based on the idea of the state paying money in the case no labor income is available. Whether one thinks of social assistance, unemployment compensation, or early retirement benefits, the rule is always that the government provides an income under the condition that people do not work and withdraws this income to the extent they do. The government acts as a competitor to private business in

420

Hans-Werner Sinn

the labor market, paying a wage for doing nothing, a replacement wage. This replacement wage creates a minimum wage demand that the private sector must exceed in order to find people who are willing to work. If, however, the labor productivity of the people involved is not high enough to make the employer pay this wage, unemployment results. Evidently the gradual expansion of replacement incomes as part of the expansion of the welfare state in recent decades contributed significantly to the rise in structural unemployment (Phelps 1994). This is a particular problem for the unskilled, whose wage is significantly pushed above the market-clearing level by this effect. However, it extends also to more skilled wage categories because certain natural distances between the categories have to be maintained. As a result of replacement incomes, the whole wage distribution is compressed from below like a harmonica held with the left hand and pushed upward with the right. Unemployment is created up to the middle income ranges, though at progressively lower rates. Small wonder that in nearly all EU countries unemployment among the unskilled is higher than that among medium-skill levels, which in turn exceeds that among people with a university education. Wage rigidity is a clue to understanding the difference between the textbook prediction of gains from trade and the reality of increasing unemployment and also for evaluating the adjustments one can observe. There are three different conceptual levels from which the structural changes can be assessed. One is the business perspective. From a business point of view, the economic reactions to trade and international wage differences obviously make sense, for otherwise they would not be taken. This is a trivial truth that no one can deny. Take the German automobile industry, for example. Because it can have its parts cheaply produced in eastern Europe, it manages to stay competitive and make profits. Another level of judgment is economic second best. Given that the wages are cut in stone and do not react to the forces of globalization, it is probably good that firms react the way they do, since otherwise they would go bankrupt, unable to maintain any employment. Capital deepening and sector shifts toward capital-intensive production, including outsourcing and offshoring the upstream parts of the production chain, are welfare-improving reactions to the forces of global-

The Welfare State and the Forces of Globalization

421

ization when wages cannot be changed. Theses reactions go along with unemployment, but there would be even more unemployment in their absence. The last and most important level of judgment is economic first best, and this is the textbook view discussed above, claiming that trade generates gains with all countries involved thanks to an improved international division of labor. This view obviously is not correct for a country like Germany, as was demonstrated above. In summary, one can therefore say that the European economy’s reactions to international low-wage competition are efficient from the perspective of firms and from an economic second-best perspective, given the rigidities of the welfare state. 13.5

Pathological Overreactions

Many politicians and even some economists face substantial difficulties when they are asked to evaluate the European economy’s reactions to the globalized world, including the new trade possibilities with the ex-Communist countries. These difficulties result from the fact that the distortions in economic reactions resulting from rigid wages resemble efficient economic reactions in qualitative terms, but go too far quantitatively if judged from the perspective of the economic first best. As the direction of economic reactions fits what the trade textbook predicts, many do not think there is reason to worry. But they are wrong because they overlook the fact that the reactions are pathologically overdrawn. Let us once again consider the forces of factor price convergence mentioned above to see why this is so. Consider migration first. If the borders between a high-wage and a low-wage region are opened, people migrate from the low- to the highwage region. In principle, such migration is efficient because it is driven by wage differences that in turn reflect marginal productivities of labor in different locations. With flexible wages, the stock of east European immigrants living and working in the West is determined such that the last migrant earns a wage increase and induces a corresponding output increase that just matches his migration cost. Thus the joint GDP of the countries involved net of migration costs is maximized. The nationals of both countries share in this welfare gain by experiencing a higher income net of migration costs than before. There are obvious gains from trade.

422

Hans-Werner Sinn

When Western wages are rigid, above their market-clearing level, this is not quite true, however. On the one hand, the high wages induce more than efficient migration. On the other hand, wage rigidity prevents the creation of new jobs for the immigrants. The result is immigration into unemployment. Immigration into unemployment is only indirect, then. It is not the immigrants who become unemployed but the nationals. The immigrants have low reservation wages reflecting their low wages at home, and they are not entitled to social benefits in the host country before they have worked there. Domestic workers, by contrast, have high reservation wages reflecting the high replacement incomes offered by the welfare state. Thus the domestic workers are the marginal suppliers in the labor market determining the wage and the employment level, and the immigrants are inframarginal low-wage suppliers that simply crowd out the domestic suppliers one by one (Sinn 2005). The next topic is capital exports. Capital exports from rich western Europe to the poor countries of the East can, in principle, be seen as welfare-increasing intertemporal trade among nations. The rate of return to capital in the rich countries as well as the rate of time preference are above the respective values in the poor countries. Thus both countries gain if the rich countries lend some of their capital to the poor countries, be it for the purpose of investment or for the purpose of consumption smoothing over time. However, when wages in the rich western countries are kept rigidly above their market-clearing levels, the rate of return to capital in these countries is artificially reduced and more than the efficient amount of capital leaves the country while unemployment increases. There is a pathological export of capital (Seidel 2005). Confusingly for the politicians who are not trained economists, this capital export comes along with a current account surplus, since such a surplus by definition is a capital export. Thus politicians even applaud the pathological overreaction as a sign of a high competitiveness of the domestic economy. Things are similar with the specialization effects described above. Consider first horizontal specialization. Suppose that the West, rich in human and real capital, opens up trade with the East, which is poor but has an abundance of unskilled labor. As explained above, a natural reaction of the rich countries is to give up some of their labor-intensive sectors so that the factors of production can move to the capitalintensive export sectors where the rich countries have a comparative

The Welfare State and the Forces of Globalization

423

advantage. As the capital-intensive sectors, by definition, are unable to absorb all the workers released in this process of fully absorbing the real and human capital, the sector shifts in the first instance cause unemployment. Suppose for a moment that the West has well-functioning labor markets with flexible wages. The initial unemployment will then lead to declining wages; the declining wages in turn will incur counterreactions that ultimately avoid unemployment for two reasons. For one thing, all sectors switch to more labor-intensive production techniques. For another, a brake is imposed on the extent of the sector shifts, as with lower wages a larger part of the labor-intensive sectors of the economy is able to survive. The economy reacts efficiently in the manner described in the textbooks. If, however, we take into account that wages are rigid, these two counterreactions cannot take place, and permanent unemployment will prevail. In particular, the brake on the sector shift, which an efficient economy imposes via declining wages, is not operative. Thus there is a landslide shift in the economy’s sector structure from labor-intensive import-competing to capital-intensive export sectors, causing mass unemployment together with a pathological overshooting of value added earned in exports (Sinn 2004, 2006). The process is further reinforced by excessive vertical specialization from labor-intensive upstream to capital-intensive downstream activities, namely by the bazaar effect. Again, a flexible economy, whose wages decline following job losses, would also exhibit the bazaar effect to some extent. Production depth would decline and export quantities would increase faster than value added in exports, as described above. However, the wage decline would help to fine-tune the structural change by again imposing a brake on the development. In the absence of such a decline, with rigid wages, the bazaar effect, too, becomes excessive. Too large a fraction of the upstream value-added chains are cut off, and too much capital and labor is moving to the downstream sectors. Thus value added in exports not only grows too much. What is more, the ratio of export quantities and value added generated in exports expands too quickly. For two reasons then the measured export quantities are becoming too large. This is particularly alarming news for a country like Germany, whose exports seem to perform excellently. Germany is number two in world exports of goods and services. In terms of goods exports alone, Germany even ranks first (WTO 2005). While most observers interpret

424

Hans-Werner Sinn

the excellent German export performance as a sign of competitive strength, the analysis above at least suggests that some caution is appropriate. Germany’s growth has been miserable during the last ten years, the country being the laggard of Europe next to Italy, and its standardized unemployment rate has risen from 8.0 to 9.5 percent. Given that Germany has very high social replacement incomes, which have made it a world champion in terms of the rate of unemployment among the unskilled (OECD 2002: 117, table A11.2), it seems possible that Germany would have done better without such high exports. Lower wages for the unskilled, determined by the forces of the market rather than the social preferences of government and unions, would have implied that more of the labor-intensive sectors of the economy had survived and that fewer jobs would have disappeared via outsourcing and offshoring. Value added in exports and export quantities would have been lower, but value added in upstream activities as well as other labor-intensive parts of the economy would have stayed higher, more than compensating for the more moderate export performance. Growth and employment would have been higher because overspecialization, excessive capital exports, and migration into unemployment could have been avoided. 13.6

Activating Social Assistance

What is the policy conclusion that follows from this analysis? Some would say that the welfare state needs to be curtailed in order to reduce the high wage competition it exerts in the labor market so that wages become flexible and the economy is able to react efficiently to the forces of globalization. But that would mean giving up the European dream of an equitable society avoiding crime and social unrest, since a welfare state provides useful insurance against the multiple economic risks that the market economy encounters. A better policy reaction is to improve the welfare state by making its redistributive activities compatible with wage flexibility. In principle, this is not difficult: instead of paying people for staying absent, the state could pay them for participating; the state could provide wage subsidies to workers instead of wage replacement incomes. If the state pays wage subsidies, there is no lower bound on wages, as people do not need a wage that compensates for the loss of social benefits. On the contrary, they would be willing to accept very low wages knowing that

The Welfare State and the Forces of Globalization

425

the subsidy will be paid in addition to these wages, augmenting their incomes. The state’s role would be to give people, who are not productive enough to earn a subsistence minimum with their own work, a second income that is tailored to the individual circumstances so as to ensure that the subsistence minimum is not undercut. The proposal follows the spirit of the American literature on this theme ranging from Haveman (1988) and Solow (1988) to Phelps (1994a, b, 1997, 2000). However, it is based on subsidies to individuals rather than their firms. In theory, both approaches lead to similar results when the market has reacted. Individual subsidies that are tailored more easily to individual circumstances may therefore be a more efficient tool for targeting the poor. A practical way of securing the subsistence minimum, inspired by the US Earned Income Tax Credit, is the Ifo Institute’s system of Activating Social Assistance (Sinn et al. 2002, 2006). Activating Social Assistance can best be understood as a reform of the existing welfare systems like the German Unemployment Benefit II or the Dutch Bystand. In Germany, a single individual without a job receives 670 euros from the state; in the Netherlands, he receives 600 euros. If this individual takes up a job, he can earn up to 100 euros in Germany or up to 150 euros in the Netherlands without a withdrawal of transfers. Thereafter, in Germany, the transfer is cut by 80 cents for every additional euro earned, and in the Netherlands, by a full euro until the transfer has disappeared. It is clear that this is an insurmountable obstacle for the labor market, generating excessive wage demands that private employers can hardly meet. According to Activating Social Assistance, not only 100 or 150 euros, respectively, would be free but also 500 euros. People could earn that much income by themselves without having to fear any transfer withdrawals. As a consequence they would demand very low wages, and at low wages there are jobs. Activating Social Assistance even means that the first 200 euros of earned income are subsidized at a rate of 20 percent such that the point at which the state pays the most money to the individual is not where his own effort is zero. A minimum of effort has to be shown in order to receive the maximum state support. In order for the system to be financially viable for the state, two additional measures are necessary. For one thing, beyond 500 euros the transfer has to be curtailed gradually. Here the withdrawal rate could

426

Hans-Werner Sinn

be 70 percent or perhaps only 50 percent, as the Scientific Advisory Council at the Ministry of Economics has suggested (Bundesministerium für Wirtschaft und Arbeit 2003). For another, the level of basic social assistance has to be cut. In the German case, a reduction by a third is necessary to keep the government budget balanced. This new system is likely to provide the necessary wage flexibility for the unskilled to enable the economy to return to full employment. In the German case, wages for the unskilled are expected to fall by one-third, with 3.2 million new jobs being created. At least in Germany a further provision of the system is required to ensure that no one who is needy can fall below the subsistence minimum even if he does not find a job in the private sector. This is communal jobs. In case of need, when no private job is found, everyone can demand employment with his local community providing him a wage for a full-time job equal to today’s social assistance or Bystand, for that matter. That is a wage of 600 euros in the Dutch case and 670 euros in the German case. One may wonder how the local community could meaningfully employ all the people who might be coming. But the simple answer is loan labor. The community can use the private loan labor business, which is already well developed in the Netherlands, and lease its labor force to the private sector at a fee that is determined by supply and demand. And for sure, there will be a fee above zero, at which the bulk of the people who come to the communities can be leased successfully to private employers. Augmented by communal loan labor contracts, the system offers two ways of wage subsidization: a direct way, where people themselves find employment in the private sector, and an indirect way, where they are employed in the private sector by the help of local communities and loan employment firms. The incentive structure is designed such that the first way is preferable. Despite the cut of the basic social assistance level, people can earn an overall income equal to the previous level of social assistance by working half-time in the private sector, and they can earn more than that by working more than half-time. If they rely on communal jobs, they have to work full time for the same income. Overall, Activating Social Assistance is a watertight new polder system that both respects the social aims of the European welfare state and the needs of a market economy that struggles to survive the international low-wage competition brought about by globalization.

The Welfare State and the Forces of Globalization

427

The old welfare state based on the idea of paying wage replacement incomes has come to its historical end. It represents an experiment of history that failed because it produced mass unemployment. It is becoming more and more obvious in these times how utopian that system really is. There is no way it can be maintained despite the forces of international factor price convergence. The new welfare system that provides people a state income while they work is the better alternative. Help for participating instead of for staying away is the new slogan. References Barro, R., and X. Sala-i-Martin. 1995. Economic Growth. New York: McGraw-Hill. Bundesministerium für Wirtschaft und Arbeit. 2003. Die Hartz-Reformen—ein Beitrag zur Lösung des Beschäftigungsproblems? Gutachten des Wissenschaftlichen Beirats. BMWADokumentation 518. Haveman, B. 1988. Starting Even: An Equal Opportunity Program to Combat the Nations’s New Poverty. New York: Simon and Schuster. OECD. 2002. Education at a Glance. Paris: OECD. Phelps, E. S. 1994a. Economic justice to the working poor through a wage subsidy. In D. B. Papadimitriou, ed., Aspects of Distribution of Wealth and Income. New York: St. Martin’s Press, 151–64. Phelps, E. S. 1994b. Low-wage employment subsidies versus the welfare state. American Economic Review 84: 54–58. Phelps, E. S. 1997. Rewarding Work: How to Restore Participation and Self-support to Free Enterprise. Cambridge: Harvard University Press. Phelps, E. S. 2000. The importance of inclusion and the power of job subsidies to increase it. OECD Economic Studies 31: 85–113 Seidel, T. 2005. Welfare effects of capital mobility with rigid wages. Applied Economics Quarterly, Supplement, Beihefte der Konjunkturpolitik 56: 61–75. Sinn, H.-W., Ch. Holzner, W. Meister, W. Ochel, and M. Werding. 2002. Aktivierende Sozialhilfe—Ein Weg zu mehr Beschäftigung und Wachstum, ifo Schnelldienst 54 (9, special issue). Sinn, H.-W. 2004. Die Basar-Ökonomie. Deutschland: Exportweltmeister oder Schlusslicht? Munich: Econ. Sinn, H.-W. 2005. Migration and social replacement incomes. How to protect low income workers in the industrialized countries against the forces of globalization and market integration. International Tax and Public Finance 12: 375–93. Sinn, H.-W. 2006. The pathological export boom and the bazaar effect: How to solve the German puzzle? The World Economy 29: 1157–75 (The World Economy Annual Lecture, Nottingham 2005).

428

Hans-Werner Sinn

Sinn, H.-W., Ch. Holzner, W. Meister, W. Ochel, and M. Werding. 2006. Redesigning the Welfare State. Germany’s Current Agenda for an Activating Social Assistance. Cheltenham, UK: Elgar. Statistisches Bundesamt. 2004. Volkswirtschaftliche Gesamtrechnungen, Input-Output-Rechnung. Importabhängigkeit der deutschen Exporte 1991, 1995, 2000 und 2002. Wiesbaden. Solow, R. 1988. Work and Welfare. Princeton: Princeton University Press. WTO. 2005. Press Release: World Trade Report 2005. Geneva.

14

Payroll Taxes, Wealth, and Employment in Neoclassical Theory: Neutrality or Nonneutrality? Hian Teck Hoon

14.1

Introduction

The theoretical proposition that temporarily below-normal tax rates on labor this year, when merged with the prospect of reversion to normal rates next year, will encourage households to squeeze more work into this year and to work less in future years is well-founded. This proposition was recently tested anew on Icelandic data and performed well empirically (Bianchi, Gudmundsson, and Zoega 2001). But would a permanent cut in tax rates on labor encourage more work permanently—with no diminution of effectiveness? Conversely, does a permanent increase in tax rates on labor cause a permanent decline in hours worked? Recently Prescott (2004) argued that the substantial decline in labor supply of French, Germans, and Italians in the past three decades could be fully explained by the increase in their effective marginal tax rates on labor. (Americans today work 50 percent more than their counterparts in the Big Three continental countries, although Europeans worked more than Americans in the early 1970s.) Using a neoclassical model of labor supply, he argued that the divergence in relative hours worked between Americans and Europeans over the three decades or so could be quantitatively accounted for entirely by the difference of marginal tax rates on labor without appealing to differences in preference for leisure or subjective rate of time discount. His quantitative model, however, assumed zero international capital mobility, an assumption that might be questioned given the lowering of barriers to international capital flows between Europe and America since the late 1960s (Obstfeld and Taylor 2004). A central result obtained in this chapter is that in a world of perfect international capital mobility, the country that raises its payroll tax rate does indeed contract

430

Hian Teck Hoon

employment initially. However, a reduced take-home pay rate also has negative effects on saving and thus on wealth next year and beyond. In the long run, wealth could tend to decrease in the same proportion as after-tax wages. As it is the after-tax wage relative to nonwage income from wealth ratio that determines the optimal number of hours supplied to the market (Hoon and Phelps 1996 first derived this relationship), and that ratio is pinned down by the common world interest rate, the number of hours worked is equalized across countries in the long run if preferences are identical.1 Therefore payroll taxes alone cannot explain cross-country differences in employment over the long run. We must proceed cautiously, however. If the payroll tax increase were used to finance government transfers—social assistance and social insurance, which constitute social wealth—instead of government purchases, the gradual decrease in private wealth is unable to fully offset the increase in social wealth even if the external rate of interest is exogenously given. Then the after-tax wage to income from both private and social wealth suffers a permanent decline so the reduction in labor input has a permanent component. The issue is an empirical one. Faggio and Nickell (2006) have also questioned the adequacy of an explanation about the American–European work difference based on labor tax differences by pointing out a puzzle. Although the Scandinavian countries (Denmark, Finland, and Sweden) have increased their marginal tax rates on labor every bit as much as the Big Three continental countries, their labor input has not declined by as much over the past three decades and is now only about 10 percent below that in the Anglo-Saxon countries. From the perspective of our theory, the long-term differences in hours worked among continental Europe, the Scandinavian countries, and America must be due to differences in time and leisure preferences or the relative importance of social wealth in these countries. 14.2

The Textbook Model

To understand the role played by wealth adjustment in offsetting the negative effect of labor taxes on labor supply, it is helpful to review the textbook model. In this model the representative household maximizes – – – U(C, L − L) subject to C + vh(L − L) = vhL + yw,

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

431

– where C is consumption, L is the fixed time endowment, L is the – number of hours worked (so L − L is leisure), vh is hourly after-tax wage and yw is unearned income, that is, income from private – wealth. Note that vhL is the potential wage earning if all the time endowment is devoted to paid work in the market and is called the full-wage income. In the standard exercise an increase in the payroll tax reduces vh, other things held constant including the unearned income yw. The fall in vh produces the standard income and substitution effects. If the latter effect dominates, which we take to be the case here, the increase in marginal tax rate on labor income discourages labor supply. In terms of figure 14.1, the household moves from the initial equilibrium point E1 to E2. This, however, is only the initial impact of higher payroll taxes. As the household’s take-home pay is reduced, personal savings fall and wealth decumulates. As unearned income, yw, gradually falls, the budget line (with a gentler slope reflecting the reduced hourly after-tax wage) shifts toward the origin. Could the unearned income fall so as to fully restore the number of hours worked, that is, to move the

C

E1

E2

BL1

BL2 E3

y w1

L

0 L2 L1 Figure 14.1 Textbook model

432

Hian Teck Hoon

household to point E3? To answer this question, we need a generalequilibrium model that endogenizes the process of wealth accumulation. In the next two sections we set out to do this, handling first the case of a small open economy that takes the external rate of interest as given and then the case of a large open economy whose asset accumulation has an influence on the external rate of interest. 14.3 The Small Open Economy Following Prescott (2004), our focus will be on an individual’s choice of his time spent in market work, without substantive interest in the choice between nonmarket housework and time for leisure. However, in working with a model with overlapping generations as described in Blanchard (1985), we face the possibility of some individuals who live forever having a rising consumption profile over their lifetimes even when the economy is in a steady state.2 Such individuals who live forever and become very rich in this model may want to retire from the workforce thus making it difficult to aggregate across generations. To preserve the tractability of the Blanchardian model despite endogenizing the work or leisure choice, we want to obtain an economy where every individual alive has an incentive to spend a positive (though variable) number of hours in the market sector.3 Our modeling choice is described in the next paragraph. We explicitly model the choice of time spent in three activities: the market sector, nonmarket housework, and leisure. Building on Benhabib, Rogerson, and Wright (1991), suppose that the utility function is given by – U = log Cˆ + A log [L − lm − ln] + B, if lm > 0, – = log Cˆ + A log [L − lm − ln], if lm = 0, where A, B > 0, and Cˆ ≡ [Cme + Cne ]1/e, e ≤ 1. Here lm is time spent working in the market sector, ln is time spent in nonmarket housework, Cm is consumption of the market good, and Cn is consumption of the home produced nonmarket good. We assume that the nonmarket good is produced according to Cn = snln; sn > 0. Notice that as in Benhabib, Rogerson, and Wright (1991), we suppose that working in the market sector gives positive direct utility, presumably because one enjoys certain social interactions and mental stimulus at work that are especially valued. We assume that there is a fixed positive utility value from

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

433

working in the market sector (given by B) that is independent of the actual number of hours worked. In contrast, the utility value derived from housework comes indirectly from consuming the home-produced good generated by the time input into the nonmarket sector. Making the further assumption that the direct utility value from spending a positive amount of time in the market is sufficiently large, in particular, that – – (A + 1) [log L − log (L − 0+)] < B, we ensure that every living person in the economy spends a positive amount of time working in the market.4 Benhabib, Rogerson, and Wright (1991) show that solving out for nonmarket housework, ln, we end up, in the Cobb–Douglas case with e = 0, the following reducedform utility function: – log Cm + (A + 1) log (L − lm) + B. The production structure of the economy is described as follows. There are two market goods: one internationally traded and the other a nontraded good. The internationally traded good, following Obstfeld (1989), is a Solow good that can be used either for consumption or for addition to the capital stock. The nontraded good is a pure consumption good. We choose the traded good as numéraire. Demographics are described as in Blanchard (1985). Agents face an instantaneous probability of death q that is constant throughout life. We now leave out the m subscript used earlier to index market good and market work. Then, using the reduced-form utility function (having solved out for ln and Cn), we refer only to market goods and time spent in market work. Let c(s, t) denote consumption at time t of an agent born at time s, l(s, t) the number of hours worked in the market, w(s, t) nonhuman wealth, and h(s, t) human wealth. Note that c(s, t) ≡ cT(s, t) + pcN(s, t), where cT(s, t) is consumption of the traded good, cN(s, t) is consumption of the nontraded good and p is the relative price of the nontraded good. Also let yg(s, t) be entitlement received and vh(s, t) be the after-tax real hourly wage (both measured in units of the traded good), where vh is related to the hourly labor cost to the firm, vf, by vf ≡ (1 + t)vh, t being the payroll tax rate. We make the assumption that workers of all age cohorts have the same productivity, face the same tax rate, and receive the same entitlement so vh(s, t) = vh(t) and yg(s, t) = yg(t) for all s. We let r(t) denote the real instantaneous short-term interest rate and r (> 0) the pure rate of time preference.

434

Hian Teck Hoon

The agent maximizes

∫

∞

t

{log[(cT (s, κ ))γ (c N (s, κ ))1−γ ] + ( A + 1)log(L − l(s, κ )) + B}exp(θ + ρ )(κ −t ) dκ ,

subject to dw(s, t) = [r(t) + θ ]w(s, t) + v h (t)l(s, t) + y g (t) − c(s, t) dt and a transversality condition that prevents agents from going indefinitely into debt. The solution to the agent’s problem, after using twostage budgeting, is given by c(s, t) = (θ + ρ)[h(s, t) + w(s, t)], L − l(s, t) A + 1 = h , c(s, t) v (t) where human wealth is given by κ

∞ − [ r ( v ) + θ ] dv h(s, t) = ∫ [l(s, κ )v h (κ ) + y g (κ )]exp ∫t dκ . t

Aggregating across all individuals, dropping the time index t and denoting per capita aggregate variables by capital letters, we obtain C = (θ + ρ)[H + W ],

(14.1)

( A + 1)C = vh , L−L

(14.2)

 = (r + θ )H − (Lv h + y g ), H

(14.3)

 = rW + Lv h + y − C , W

(14.4)

g

where a dot over a variable denotes its time derivative. We note that although every worker faces the same hourly pay, the fact that the members of the labor force are of different ages means that their wealth levels are different, and consequently the number of hours worked will be different across the different age cohorts. The government is assumed to run a balanced-budget policy and, for simplicity, we set government debt to zero. The government budget constraint can be expressed as

τ Lv h = y g + G,

(14.5)

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

435

where G ≡ GT + pGN is the per capita level of government purchases, and tax revenue collected is from payroll taxation. Here we assume that domestic residents own all the capital stock, K, and assuming free international lending and borrowing, W ≡ K + F, where F is the holding of net foreign assets. (Residents can always borrow from abroad to achieve this portfolio allocation.) Taking the time derivative of (14.1), and using (14.3) and (14.4), we obtain C = (θ + ρ)[rW + (r + θ )H − C].

(14.6)

Substituting (14.1) into (14.6), we obtain, after re-arrangement of terms, C θ (θ + ρ)[K + F] = (r − ρ) − . C C

(14.7)

Let us now lay out the conditions satisfied by the production side of the economy. We assume that the nontraded good is the relatively labor intensive good.5 For simplicity, we suppose that producing a unit of the non-traded good requires Λ−1 N units of labor. The traded good constant-returns-to-scale production function is given by F(K, LT), where LT is employment in the traded-good sector. Profit maximization gives v f = pΛ N = f (kT ) − kT f ′(kT ),

(14.8)

r = f ′(kT ),

(14.9)

where kT ≡ K/LT, f ′(kT) > 0; f ″(kT) < 0. From (14.8), we see that kT is positively related to p and in conjunction with (14.9), p = f(r); f ′(r) < 0. For the small open economy, r = r* (exogenously given) > 0. Consequently the relative price of the nontraded good is pinned down by the external rate of interest. To determine the domestic stock of capital, we use the market-clearing condition for the nontraded good sector: (1 − γ )C + GN = Λ N p

K⎤ ⎡ ⎢⎣ L − kT ⎥⎦ .

(14.10)

14.3.1 The Pure Case of Government Purchases with No Transfers To understand the long-run effects of labor taxes, it is convenient to first focus on the steady state before turning to the dynamics. In a steady state with no government transfers, aggregate human wealth, H, equals vhL/(r + q). Nonhuman private wealth is given by K + F,

436

Hian Teck Hoon

so nonwage income (or unearned) income from private wealth is given by yw = (r + q)(K + F), which takes the form of annuity income, q(K + F) being the component called the actuarial dividend. Using these relations and (14.1) in (14.2), and rearranging, we obtain the steadystate labor-supply relation in manhours:

θ + ρ ⎤ ⎛ vh L ⎞ 1 − ( A + 1) ⎡ ⎢⎣ r + θ ⎥⎦ ⎜⎝ y w ⎟⎠ L = θ + ρ⎤ L 1 + ( A + 1) ⎡ ⎣⎢ r + θ ⎦⎥

−1

.

(14.11)

We also note that in a steady state, (14.7) can be expressed as ⎡ ⎤ ⎢ ⎥ 1 ⎥. r = ρ +θ ⎢ ⎢ ⎛ vh L ⎞ ⎥ ⎢ 1 + ⎜⎝ y w ⎟⎠ ( L / L ) ⎥ ⎣ ⎦

(14.12)

Suppose that the government now finances increased purchases (ΔG) by raising payroll taxes. The key result from (14.11) and (14.12) is that with the interest rate pinned down by the exogenously given external – – rate, r = r*, (vhL /yw) and L/L are determined independently of the payroll tax rate, t . Hence an increase in the payroll tax rate is neutral for employment in the long run. The economic logic underlying the neutrality result is as follows. Since the external interest rate pins down the demand wage offered by the firm, vf, the after-tax wage v h ≡ v f/(1 + t) is reduced one for one by the increase of labor taxes. In the long run, however, unearned income from private wealth, yw, falls by the same proportion so as to leave the wage to nonwage income ratio unchanged. Hence labor taxes raised to finance government purchases are neutral for employment in the long run. In effect government purchases end up crowding out private consumption one for one. To understand the dynamics, it is useful to note from (14.2) that labor  supply is uniquely pinned down by C ≡ C/vh. We write L = m( C);   m′( C) < 0. Using (14.7), we can write in terms of C,  C θ (θ + ρ)[K + F] (14.13) = (r − ρ) − .   h C Cv The evolution of private wealth is dictated by K + F = r[K + F] + [μ(C ) − C ]v h .

(14.14)

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

437

Note that r = r*, and that vh is pinned down by r* and the value of t, so (14.13) and (14.14) give us a pair of dynamic equations in C and K + F, with the latter being a state variable. We assume that r* > r so that, in steady state, the representative household owns a positive level of wealth. The phase diagram in figure 14.2 shows that we obtain saddle-path stability. Figure 14.3 shows the dynamic response of the economy to a sudden permanent increase in the payroll tax rate, t, which reduces vh, used to finance government purchases. Recalling that  we see that the maximum conlabor supply is inversely related to C, traction in employment occurs initially. However, as private wealth gradually declines, employment recovers until in the new steady state, employment is exactly back to where it was before the policy. A sudden permanent increase in payroll tax rates therefore produces a temporary contraction of employment with the negative effect of higher payroll tax rates diminishing to zero as wealth decumulates. 14.3.2 Increase in Payroll Taxes to Finance Government Transfers In a steady state, aggregate human wealth, H, now equals (vhL + y g)/ (r + q), which includes the component, yg/(r + q) (the present discounted value of the stream of government transfers), that can be thought of as social wealth. Nonhuman private wealth is given by ~ C ~• C=0

•

•

K+F=0

0 Figure 14.2 Phase diagram

K+F

438

Hian Teck Hoon

~ C

~• C=0

•

•

K+F=0

K+F Figure 14.3 Dynamic response

K + F, so nonwage income (or unearned) income from private wealth is given by yw = (r + q)(K + F). Using these relations and (14.1) in (14.2), and rearranging, we obtain the steady-state labor-supply relation in manhours: g h w L 1 − ( A + 1) [(θ + ρ ) (r + θ )][ v L ( y + y )] . = L 1 + ( A + 1) [(θ + ρ ) (r + θ )] −1

(14.15)

We also note that in a steady state, (14.7) is now expressed as ⎤ ⎛ yw ⎞ ⎡ 1 r = ρ +θ ⎜ w ⎥. g⎟ ⎢ g h w ⎝ y + y ⎠ ⎣ 1 + [ v L ( y + y )] (L / L ) ⎦

(14.16)

Suppose that initially yg and t are both zero. Figure 14.4 illustrates the effect of raising payroll taxes to finance a positive level of government transfers. On the one hand, given r = r*, we have from (14.15) that – – L/L is positively related to vhL /(yw + yg), so we obtain the positively sloped line in figure 14.4. On the other hand, given r = r*, (14.16) obtains a hyperbola. The increase in payroll taxes used to finance yg has the effect of shifting inward the hyperbola along an unshifted labor supply – curve. The result is that there is a permanent decline in vhL /(yw + yg).

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

439

vhL y + yg w

L L Figure 14.4 Government transfers

In this case the decline in income from private wealth in response to reduced take-home wage fails to offset the increase in income from social wealth (yg) so the reduction in employment has a permanent component. 14.4 The Large Open Economy It might be argued that if we consider a simultaneous increase in the payroll tax rate in each of the continental European economies, it might be more appropriate to think of a large open economy where dissaving is likely to affect the world interest rate. In this section we extend the analysis to a large open economy. If payroll taxes raised to finance government purchases and transfers lead to a tightening of private savings, and the increased borrowing in the international capital market leads to an increase in the world interest rate, will the after-tax wage to unearned income ratio be permanently reduced? Taking the pure case of government purchases with no transfers, we suppose that as

440

Hian Teck Hoon

the large economy increases the stock of private wealth (K + F), it acts to lower the real rate of interest, that is, r = R(K + F); R′(K + F) > 0 with an interest elasticity (defined with a positive sign) that we assume is less than one. We first determine what happens to the long-run rate of interest in response to the increase in labor taxes. .  Setting C = 0 in (14.13) gives

θ (θ + ρ)[K + F] , C = h v [R(K + F ) − ρ] . . and setting K + F = 0 in (14.14) gives (K + F )R(K + F ) C − μ(C ) = . vh

(14.17)

(14.18)

Substitution of (K + F)/vh in (14.18) using (14.17) gives us ⎡ C − μ(C ) ⎤ R(K + F )[R(K + F ) − ρ] , ⎢ ⎥= θ (θ + ρ) C ⎣ ⎦

(14.19)

where we note that the left-hand side of (14.19) is increasing in C and the right-hand side is decreasing in K + F. We plot this relationship as the downward-sloping schedule in figure 14.5.6 Note from (14.17) that we have a positively sloped schedule that we also plot in figure 14.5. The intersection gives the initial equilibrium C and K + F, and hence the initial steady-state rate of interest. At given K + F, we see from (14.17) that an increase in the payroll tax rate lowers vh and so shifts up the positively sloped schedule. Consequently an increase in the payroll tax rate raises the real rate of interest in the long run, increases  and thus decreases employment. C, To recapitulate, an increase in the payroll tax rate in a large open economy such as continental Europe leads, in the long run, to a decline in European employment so the higher marginal tax rate does have a permanent negative effect on employment. The reason is that the dissaving that is brought about by reduced take-home pay raises the world interest rate, which prevents private wealth from falling enough to restore the after-tax wage to nonwage income ratio back to its original level. This does not imply, however, that another country, say America, with unchanged (lower) payroll tax rates will have higher employment. In fact, if preferences, including the subjective rate of time preference, and mortality rate are identical and there is no hindrance to the cross-border mobility of capital so that each country faces a common

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

441

~ C

K+F Figure 14.5 Steady-state effect in large open economy

world interest rate, the fiscal shock in Europe acts also to contract American employment by the same amount in the long run. The economic mechanism producing this result is that wealth accumulation in America (which runs a current account surplus) that is stimulated by the higher world interest rate brought about by European dissaving leads to a decline in the wage to nonwage income ratio in America. In fact, with both America and Europe facing the common world interest rate, we learn from (14.11) and (14.12) that given r, there is an equalization of the employment rates in the long run despite different payroll tax rates. Studying the dynamics is now straightforward. Setting r = R(K + F) in (14.13) and (14.14), and by the factor-price frontier relationship making vf a decreasing function of r, we can study the economy’s response to an increase in payroll taxes used to finance government purchases. The adjustment path in response to a sudden permanent increase in t is shown in figure 14.6. The impact of the labor taxes on employment is greatest initially, and wealth decumlation once again leads to diminishing negative effects of higher payroll taxes, although there is now a permanent negative component as world interest rate is pushed up.

442

Hian Teck Hoon

~ C

K+F Figure 14.6 Dynamic response in large open economy

One result in the large open economy case is noteworthy. Suppose that the government purchases child-care services, which we regard as a nontraded good and thus relatively labor intensive, and pays for it with higher payroll taxes. The model suggests that as savings decline and the economy ends up borrowing from the international capital market and pushes up the world interest rate, child-care services become relatively cheaper in terms of the traded good. The reason is that the higher interest rate means that wage costs (v f) faced by the firm are lowered. As child-care services are relatively labor intensive, their relative price declines. 14.5

Concluding Remarks

Proved in this chapter was the proposition that a permanent increase in payroll taxes used to finance government purchases in a small open economy that takes the external rate of interest as given is neutral for employment in the long run. However, if the economy in question is large in the sense that an increase in its borrowing in the world capital market pushes up the world interest rate, the neutrality breaks down.

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

443

In the large open economy case, the higher world interest rate, in lowering the after-tax wage to unearned income ratio, reduces permanently the number of hours supplied. This does not mean, however, that the other economies in the integrated world economy with lower payroll tax rates have higher employment. As these economies run current account surpluses, stimulated by the higher world interest rate, their wage to nonwage income ratios fall causing the number of hours worked in the low-payroll-tax economies also to decline. If preferences are identical across countries, there is an equalization of employment across countries despite differences in payroll tax rates. Different marginal tax rates on labor cannot explain long-run differences in hours worked across countries if capital is internationally mobile. Does the model of this chapter provide a resolution to the Faggio– Nickell puzzle that despite higher labor taxes in both the group of Scandinavian countries (Denmark, Finland, and Sweden) and the Big Three continental countries, labor inputs fell far more in the latter than in the former over the past three decades? Can one argue that the Scandinavian countries can be safely regarded as small open economies with little influence on the world interest rate so payroll taxes are neutral whereas the Big Three continental countries have market power in the world capital market so that payroll taxes are non-neutral in the long run? If the Scandinavian countries and the Big Three continental countries all operate in an integrated international capital market, the resultant higher world interest rate brought about by reduced savings in the latter would have an effect on the former through the interest channel. The higher world interest rate lowers the after-tax wage to unearned income ratio and hence reduces labor supply in the Scandinavian countries through the international capital-market linkage. From the point of view of the theory developed here, two routes are open to explain the Faggio–Nickell puzzle. If a higher proportion of payroll tax revenues is used to finance government transfers in continental Europe compared to the Scandinavian countries, then total hours worked will be lower in the former than in the latter despite every country facing a common world interest rate. Another hypothesis that has to be explored is whether the relatively high employment in the Scandinavian countries despite high marginal labor taxes is due to superior “entrepreneurial” institutions and economic culture that are represented here by different values of preference parameters.

444

Hian Teck Hoon

Notes I am grateful to Ned Phelps for the lessons I have learned from our long collaboration upon which this chapter builds. I also thank Pentti Kouri for his comments on the first version of this chapter. 1. Hoon and Phelps (1996) explored the consequences of substituting payroll taxes for VAT and showed long-run neutrality in a small open economy but nonneutrality in a large open economy. They explored the fiscal consequences in two types of labor market: a neoclassical one as well as one with an endogenous natural rate of unemployment derived from labor turnover. 2. The reason we do not use an infinitely-lived representative agent model as in Prescott (2004) is that applying such a model in a world economy with perfect international capital mobility leads to national wealth being degenerate. To obtain nondegenerate wealth in the open economy, we can either use the Blanchard–Yaari model, where all individuals face a constant and identical probability of death, or a model of overlapping and unconnected infinitely-lived families as in Weil (1989) and Obstfeld (1989). Our results carry through if we adopt the Weil–Obstfeld characterization of demographics instead of the Blanchardian characterization. 3. Since Prescott (2004) focuses on the choice of market work at the intensive rather than the extensive margin, the representative agent always supplies a positive number of hours to the market sector. 4. A very wealthy individual who might have chosen to retire in a model without a positive utility value from market work spends a very small positive amount of time working in the market (lm = 0+) given the positive utility value of market work compared to housework in our model. 5. Obstfeld (1989) notes that empirical evidence gives support to the assumption that nontradables, taken as an aggregate, are relatively labor intensive compared to tradables and cites Kravis and Lipsey (1983). 6. Notice that in the case of a small open economy where the external rate of interest is taken as given, this line is horizontal.

References Benhabib, J., R. Rogerson, and R. Wright. 1991. Homework in macroeconomics: Household production and aggregate fluctuations. Journal of Political Economy 99 (6): 1166–87. Bianchi, M., B. Gudmundsson, and G. Zoega. 2001. An Icelandic natural experiment in supply-side economics. American Economic Review 91 (5): 1564–79. Blanchard, O. J. 1985. Debts, deficits and finite horizons. Journal of Political Economy 93 (April): 223–47. Faggio, G., and S. Nickell. 2006. Patterns of work across the OECD. Discussion paper 730. Centre for Economic Performance, London. Hoon, H. T., and E. S. Phelps. 1996. Payroll taxes and VAT in a labor-turnover model of the “natural rate.” International Tax and Public Finance 3: 369–83.

Payroll Taxes, Wealth, and Employment in Neoclassical Theory

445

Kravis, I. B., and R. E. Lipsey. 1983. Toward an Explanation of National Price Levels, vol. 52. Princeton: Princeton Studies in International Finance. Obstfeld, M. 1989. Fiscal deficits and relative prices in a growing world economy. Journal of Monetary Economics 23 (3): 461–84. Obstfeld, M., and A. M. Taylor. 2004. Global Capital Markets: Integration, Crisis, and Growth. Cambridge: Cambridge University Press. Prescott, E. C. 2004. Why do Americans work so much more than Europeans? Federal Reserve Bank of Minneapolis Quarterly Review 28 (1): 2–13. Weil, P. 1989. Overlapping families of infinitely-lived agents. Journal of Public Economics 38: 183–98.

15

Economic Culture and Economic Performance: What Light Is Shed on the Continent’s Problem? Edmund S. Phelps

15.1

Introduction

Enlightenment thinkers, from Smith and Hume on to Kant and de Tocqueville, all took it for granted that a society’s culture—the people’s values, attitudes, morals, and beliefs, many of them learned at their mother’s knee—mattered for the effectiveness of business life and, more broadly, for the realization of the society’s potential. The Enlightenment is often caricatured as the doctrine that a society eschewing superstition and taboos and embracing reason and individual opportunity will with time attain perfection of its possibilities. Notwithstanding various dissenters, including Marx, who took culture to be a function of the economy’s structure rather than the reverse, the Enlightenment view on the influence of a nation’s culture remained prevalent right through the “Protestant ethic” in Weber (1905) and the “entrepreneurial spirit” in Schumpeter (1911). One could imagine their running inter-country regressions of differences in economic performance on differences in the culture, particularly the economic culture. By the middle of the twentieth century moral relativism had taken over. Most anthropologists and many other social scientists were disinclined to evaluate contrasting national cultures, seemingly believing that every nation finds its way to the culture that is best for it. Hence a society’s culture might have a downside in its ill-effects on its economy, yet the cost would be compensated by benefits in other directions. So it would be valueless and perhaps politically incorrect to run those regressions. Nevertheless, a push back against such relativism soon began. Ruth Benedict wrote that some cultures may be better or worse than others. Several works reestablished culture as a causal force that makes markets work better: Banfield on trust (1958), Titmuss on gifts (1970), the Russell Sage conference on altruism (Phelps 1973),

448

Edmund S. Phelps

and Putnam on civic virtue (1993). It is plausible again to run those regressions. The debate over economic performance in continental Europe may prove to be a testing ground for the view that culture matters—some elements of it at any rate—for a society’s results. As is increasingly admitted, the performance characteristics—one might say the specifications—of the national economy in nearly every continental country are poor compared to most performance characteristics in the United States and a few other comparators. (Productivity in the Big Three—Germany, France, and Italy—stopped catching up with US productivity in the early 1990s, then lost ground in the recent slowdowns and the US speed-up; unemployment rates, which were declining, are again far higher than those in the United States, United Kingdom, Ireland, etc.) However, the crucial point is not that the Continent’s economic systems are inferior to those of some comparators but rather the nagging sense of falling short—of structural underperformance. In my view, the continental economies had started to be underperformers in the interwar period and remained so, with corrective steps here and further missteps there, from the postwar decades onward. The structural shortfall was masked during the “glorious years” when rapid growth and high employment was stimulated by the low-hanging fruit of unexploited technologies used overseas and further powered by Europeans’ efforts to claw back the wealth they had lost in the war years. Many analyses, looking beyond market forces (e.g., the rather important influence of demographic prospects), attribute the Continent’s tendency toward relatively high unemployment and low participation, if not the lower productivity, to the Continent’s social model. Yet this explanation has not had entirely clear sailing.1 One could as easily bring up the political model. The Continent’s historic struggle between left and right may create uncertainty for those investing or innovating on the Continent. The rule of law, or procedural justice, has received much attention from Adam Smith to Douglass North. But the nations on the Continent are not a bunch of banana republics. It is not clear that they are behind their comparators in constitutional protections, property rights, antitrust, law enforcement, and judiciary independence. (Some would put interventionism in economic policy, e.g., that typical of corporatist economies, under the heading of property rights violations, but this chapter regards that as part of the economic system.) My thesis for several years has been that it is the economic model that largely accounts—more, at any rate, than the other models do—for

What Light Is Shed on the Continent’s Problem?

449

the Continent’s inability to match the economic performance of the United States and in some respects that of other comparators.2 But what is the “economic” model—in other words, what is the “economy”? At first, like others, I meant the economic system, namely the system of economic institutions in the capital, labor, and product markets. In arguing my thesis, I pointed to the strength on the Continent of institutions understood to be bad, such as employment protection legislation and bureaucratic “red tape,” and to the weakness of institutions understood to be good, such as a well-functioning stock market and ample liberal arts education. Important modeling and testing was done both before and after my work by Aghion and Howitt (1993, 2005). This line of analysis is of limited generality, though. To justify regressing performance on institutions, one might posit that institutions are sprinkled over countries as an experimental agronomist might sprinkle trial fertilizers over various plots of land. A lecture of mine (2003b) and a book by Eggertsson (2005) discuss why country might go on with one or more “inefficient” institutions—out of ignorance of how badly the system has functioned and difficulties of statistical inference about individual institutions.3 But it may be that countries have differing institutions because they have different economic cultures causing them to prefer different systems of institutions. Then a country’s economic institutions are proxies to some unknown extent for the prevailing culture. In that case the prevailing set of institutions might not be alterable as long as the culture is unchanged. The purely institutional regression is worse yet if the cultural elements have direct effects on performance—on top of their indirect effects through the institutions they foster. In that case the inhibiting values, attitudes, and the like, are analogous to institutions that deter or bar good performance and the empowering values, attitudes, and the like, are analogous to institutions that foster and enable good performance. So they are in a sense as much a part of the “economy” and the “economic system” as the institutions (in the narrow sense). It is clear that a regression that omits a huge part of the system that the economy is composed of is extremely unreliable. To credit an institution with some estimated effectiveness when the cultural variable are omitted is to risk attributing to it the direct benefits of the cultural influences traits that caused the institution to be built. Of course, any program to explain inter-country differences by appeal to differences in cultural influences would be incomprehensible from the standpoint of neoclassical or neo-neoclassical theory. The

450

Edmund S. Phelps

Arrow–Debreu equations have no cultural elements—and not any economic institutions either, other than private ownership. It follows that a rationale for cultural effects, if found, must go outside the neoclassical paradigm to recognize entrepreneurship, management, engaging jobs, learning and personal growth, and team players—thus Knightian uncertainty and creativity as well as imperfect information. Elsewhere I have stressed the importance of “dynamism” for the performance of a market economy, which I will compress here. (These thoughts suggest that in an economy where entrepreneurial activity is important, the culture of the people available for work is analogous also to their “know-how”: just as a stage and a hall will not “work” if the assembled players have not acquired the ability to act and interact, so the plant and hardware of an entrepreneurial company will not “work” if the personnel is unwilling or unhappy to be organized and to organize themselves as a team.) This chapter, then, will explore for effects of several cultural values, attitudes, and the like, on some of the main dimensions of economic performance. 15.2

Cultural Influences on Performance—A Conceptual Frame

If we are to obtain estimates of the performance effects of national cultural attributes that have any claim to reliability and interpretability we had better base our investigation on some conceptual framework, however informally formulated it may be—rather than try whatever off-the-shelf variables are at hand. This appears to require some notion of what a system of economic institutions and economic culture is in view of economic change and particularly the processes of innovation, their benefits and drawbacks, and their consequences for the main indicators, economic growth and prosperity. The neoclassical framework, with its premise of perfect knowledge and perfect coordination, is too narrow for much understanding of underperformance and the possible role that institutions and culture may play in it; so we want to go beyond neoclassical economics. Contrary to myth, what we commonly call the West is not polar with respect to the character of its economies, with the so-called AngloSaxon economies all operating on the system called capitalism, with or without an accompanying welfare state, and all the continental economies operating on the system called corporatist, social market, or Rhenish. Denmark’s economy is thought to be different in some way,

What Light Is Shed on the Continent’s Problem?

451

and Italy’s is surely more industrious than most of the Anglo-Saxon economies. The Nordic nations, from Finland to Iceland, do not fit neatly into either category. Nevertheless, there is some utility in considering two extremes—two ideal types—each of which resonates somewhat with one or more actual economies in the West. At one extreme we have a private-ownership system structured for cutting-edge innovation. It is fertile in coming up with innovative ideas with prospects of profitability; shrewd and adept in selecting among these ideas for development; finally, prepared and venturesome in evaluating and trying the new products and methods that are brought out. A semiclassical theory of innovation began with Schumpeter (1911). Saving is allocated to developing entrepreneurs’ proposed “innovations” only to the extent that there are businesspeople around with the initiative to “seize the moment” and the leadership to “get it done.”4 The modern theory of such dynamism—and the case for adopting such a system—began in the mid-1930s with Hayek (1948). First, virtually every employee down to the humblest worker has arcane “know-how,” some of it what Michael Polanyí called “personal knowledge,” and out of that know-how a new idea may come that few others, if any, would have.5 With openness to commercial ideas and acceptance of the entrepreneurs who develop them, a plethora of new ideas may be generated. Second, the pluralism of experience and knowledge that the financiers bring to bear in their decisions gives a wide range of entrepreneurial ideas a chance of an informed, insightful evaluation. And, importantly, the financier and the entrepreneur do not need the approval of the state or of social partners. Nor are they accountable later on to such a social body if the project goes badly, not even to the financier’s investors. So projects can be undertaken that would be too opaque and uncertain for the state or social partners to endorse. Third, the pluralism of knowledge and experience that managers and consumers bring to bear in deciding which innovations to try and which of those to adopt is crucial in encouraging entrepreneurs to conceive new ideas and financiers to back them. At the other extreme we have a private-ownership system that has been profoundly modified by the introduction of additional institutions. These include the massive components of the corporatist system of interwar Italy—big employer confederations, big unions, and big banks. The system operates to discourage or bar many entrepreneurial projects, particularly start-ups. For its “innovations”—most of them not world class, not “cutting edge,” but rather adaptations of products

452

Edmund S. Phelps

and methods recently introduced abroad—the system depends more on established companies in cooperation with local and national banks. For what it lacks in entrepreneurship it tries to compensate with technological sophistication and increased coordination. Where the former system allows any number of versions of a new product or method to be developed and launched, this latter system convenes experts to set a product standard before any version is launched. To what end is this system? What is the theory behind it? First, there is the solidarist aim of protecting the “social partners”—communities and regions, business owners, organized labor, and the professions—from disruptive market forces; also the consensualist aim of blocking business initiatives that lack the consent of the “stakeholders”—those with a stake besides the owners, such as employees, customers, and rival companies. Second, elevating community, society, and being over individual engagement and personal growth appeals to antimaterialist and egalitarian strains in Western culture. Third, there is the “scientism” that holds that such a system can be more dynamic than the former system—maybe not more fertile in little ideas, such as might come to petit bourgeois entrepreneurs, but certainly in big ideas. Not having to fear fluid market conditions, an entrenched firm can afford to develop expensive innovations based on current or developable technologies. And with confederations of firms and state mediation available, such firms could arrange to avoid costly duplication of their investments. The state, for its part, could promote technological advances in cooperation with industry by harnessing the society collective knowledge. The state could indicate new economic directions and favor some investments over others through its instrument, the big banks. The impetus for this chapter has been the intuition that several countries on the Continent—among them Germany, Italy, and France— had and still have a culture that led them to evolve and retain systems of institutions that (in most or all respects) are much closer to the latter extreme than are the systems that the culture of the United States, the United Kingdom, and Canada led them to evolve and retain. It may be that, with their culture attitudes, the former system was abhorrent to them. Or it may be that they thought that their culture would ill-equip them to do well with the former system. Or, conceivably, their culture might predispose them against dynamist behavior whatever system they adopted. So it is of more than minor interest that there is evidence, shown in appendix B, in support of that intuition: Germany, Italy, and

What Light Is Shed on the Continent’s Problem?

453

France appear to generate less dynamism than do the Anglo-Nordic nations: notably fewer firms able to break into the top ranks and fewer jobs offering freedom in decision-making (see table 15A.1 of the appendix). It further appears that Germany, Italy and France as a whole have worse performance indicators: not only lower productivity and labor compensation—in the latter contest, the United States handily beats France and Canada, and almost ties Germany, and the United Kingdom beats Italy—but also worse rates of labor force participation and unemployment (see table 15.A2 of the appendix). Nevertheless, dynamism is not the sole determinant of economic performance. Industrial composition is obviously important for aggregate performance indicators. A long-time thesis of mine is that attained wealth normalized by productivity has ill-effects on many indicators. Some results here suggest that how a society and its companies cope with the changes wrought by the dynamism around them matters a lot. So in this first exploration I will take the expedient course of testing for relationships between performance indicators and culture—not testing whether culture impacts on dynamism and dynamism on performance. What, in view of the above, are the presumably pertinent cultural values, attitudes, ethics, and beliefs in each economically advanced country in the OECD—in the available data set? And do the intercountry differences among them appear to play a role in causing inter-country differences in economic performance? The cultural data in this chapter are limited to those calculated from underlying data (on the individual respondents’ answers) contained in World Values Surveys, which, though providing a wealth of data, is not nearly as wide-ranging as we would like. A quick perusal (by disinterested research assistants!) of the 1990 to 1993 surveys has served to focus my thinking around the data that are available. At first, I found myself defining four dimensions of culture and looking for Survey questions that would serve to characterize each country’s culture in every one of the four dimensions. The mean of the individual responses in a country to a Survey question about values, attitudes, and the like, would be the national indicator locating the country with respect to that question in the dimension to which the question belonged; another Survey question in that dimension, if such was found, would provide another national indicator in the same dimension; and so forth. Then these indicators could be averaged to yield an index of the nation’s scores in that dimension. In the end,

454

Edmund S. Phelps

though, I decided to let the national mean response to each cultural question used stand alone rather than to average them into some index of indicators. One dimension has, as I would put it, Stimulation/Engagement/ Mastery/Development at one end and at the other, Being/Identity. (I think of this cultural dimension as the Wm. James/H. Bergson dimension.) One national indicator calculable from the underlying Survey data that belongs more clearly in this dimension than in the other dimensions gets at the centrality of jobholding in the culture of the country. This indicator, labeled Importance, measures the response to the question “Is your job the most important thing in your life?” (c046 in the Surveys codes6). Other national indicators that clearly belong to this dimension are calculated from responses to questions asking respondents what they look for in a job. One of these indicators, Involvement (c031), measures the respondents’ reported pride in their work. A second, Interestingness (c020), measures the preference for an interesting job. A third, Achievement (c018), measures the preference to “achieve something.” (This may be too vague or broad to be useful. All manner of employees may want to achieve “something”: making shoes or building bridges or winning promotions; it need not mean creating a new idea or a new firm.) The second dimension has at one end Loyalty/Dutifulness/Altruism and at the other end Practicality/Opportunism/Egoism. (This is Plato’s debate between Socrates’s responsibility and Thrasymachus’s unalloyed self-interest.) The sole indicator from the Surveys that appears to belong in this dimension is Willingness to Follow Orders (c061). I owe my awareness of this hypothesis to Angelo Airaghi, senior vice president for strategy at Finmecannica (and a close friend of mine). He once commented to me that in his observations over a long career, much of it in Japan and in the United States, if he had to place the United States on a spectrum with the company men of Japan at one extreme and the everyone-for-himself Italians at the other, the Americans would be 90 percent of the way toward the Japanese. (I was surprised but came to take it seriously.) The third dimension has at one end Individualism/Pluralism/Tolerance and at the other Solidarity/Conformity/Unanimitarianism/Envy. (I think of this as the Kantian dimension.) Here there is an indicator, Acceptance of Competition (e039), calculated from responses to a question of whether the respondent is positive or negative about competition. A Survey question on the attitude toward foreigners had a low

What Light Is Shed on the Continent’s Problem?

455

response rate and was not successful. Another indicator, EfficiencyFairness (c059), which came from a question asking whether it would be fair that a secretary who could do twice the work as another would earn twice as much, was such a failure that it was dropped. The fourth dimension has at one end Initiative/Venturesomeness/ Experimentalism and at the other end Passivity/Tradition. (In their different ways Schumpeter and Hayek conceived this dimension.) Culture indicators in the Surveys that fit here are Desire for Freedom to Make Decisions and thus possibly, freedom to lead—henceforth, Freedom in Decisions (e053), Preference for New Ideas over Old Ideas (e046), Self-confidence (e048), Acceptance of Changes (e047), and Initiative at Work (c016). This last may work poorly or perversely if a strongly positive response goes hand in hand with scientism—with a national reliance on state coordination of collective knowledge rather than on entrepreneurs and financiers. Other dimensions were conceived but indicators to implement them did not appear to be available from Surveys. One was Populism/Mass Protest versus Liberalism. Another was top–down Scientism versus Hayek’s bottom–up Organic Growth of Knowledge. Of course, various other observations and polls might also be used to obtain proxies for popular attitudes. For the present paper, though, it appears sufficient to make do with the data available in Surveys. 15.3

Cultural Influences—Some Statistical Tests

Tables 15.1 through 15.5 report the estimated effects of our selected cultural variables on five standard economic indicators: male labor force participation, the activity rate (also known as the employment rate), employment in percent of the labor force and two measures of labor productivity. In each table, model 1 focuses on the ten culture variables. To have a sort of benchmark with which to compare the significance of the culture variables, the tables next focus in model 2 on a set of “traditional” explanatory variables. Model 3 combines the set of culture and traditional variables. Models 4 and 5 are narrowed to those culture variables that were significant in model 3. (Some of the traditional variables that were far from significant were also omitted.) In models 1 to 4, the tables present both the ordinary least square (OLS) estimates and the more realistic generalized least squares (GLS) estimates with White standard errors. These regressions “pool” data from 1996, 1999, 2002, and 2005. The estimation in model 5 takes into account

−0.018 (0.055)

0.042 (0.010)***

−0.07 (0.044)

−0.018 (0.060)

0.002 (0.0002)***

0.031 (0.015)**

0.198 (0.102)*

0.062 (0.02)***

0.042 (0.014)***

−0.07 (0.067)

Achievement (c018)

Interestingness of Work (c020)

Involvement at Work (c031)

Importance of Work (c046)

Willingness to Follow (c061)

Acceptance of New Ideas (e046)

Freedom in Decisions (e053)

0.062 (0.019)***

0.198 (0.096)**

0.031 (0.013)**

0.002 (0.00002)***

0.093 (0.046)**

GLS, White SE

0.093 (0.053)**

OLS

Initiative at Work (c016)

Dependent variable: male participation in the labor force

Model 1: culture variables

OLS

GLS, White SE

Model 2: traditional variables

−0.064 (0.055)

−0.064 (0.066)

0.074 (0.027)***

−0.029 (0.036)

−0.029 (0.033) 0.074 (0.023)***

0.018 (0.103)

0.001 (0.016)

0 (0.0005)

0.018 (0.095)

0.001 (0.016)

0 (0.0004)

0.112 (0.065)*

−0.027 (0.036)

−0.027 (0.048) 0.112 (0.068)*

GLS, White SE

OLS

Model 3: culture and traditional variables

Table 15.1 Determinants of economics performance: Male participation rate in the labor force

0.071 (0.011)***

0.06 (0.079)

0.001 (0.0005)**

OLS

0.071 (0.009)**

0.06 (0.081)

0.001 (0.0006)*

GLS, White SE

Model 4: narrowed

0.059 (0.021)***

0.238 (0.146)*

0.001 (0.0006)***

RE

Model 5: narrowed, panel data

456 Edmund S. Phelps

−0.004 (0.009)

−0.004 (0.008)

0.002 (0.008)

0.001 (0.008)

Year 1996

Year 1999

Year 2005

68

Observations 0.44

0.44

72

0.811 (0.192)***

0 (0.001)

0.002 (0.003)

0.89

0.89

68

−0.037 (0.066)

−0.037 (0.101) 68

0.001 (0.000)***

0.001 (0.000)***

0.009 (0.002)***

−0.003 (0.006)

−0.003 (0.006) 0.009 (0.002)***

−0.002 (0.001)**

0.001 (0.006)

0.007 (0.006)

0.003 (0.008)

0.025 (0.017)

−0.087 (0.099)

−0.002 (0.001)***

0.001 (0.006)

0.007 (0.007)

0.003 (0.007)

0.025 (0.019)*

−0.087 (0.095)

0.86

68

0.14 (0.082)

0.001 (0.000)***

0.006 (0.002)***

−0.004 (0.004)

−0.003 (0.000)***

0.001 (0.007)

0.008 (0.007)

0.004 (0.007)

0.029 (0.011)**

0.86

68

0.14 (0.077)*

0.001 (0.000)**

0.006 (0.001)**

−0.004 (0.003)

−0.003 (0.000)**

0.001 (0.006)

0.008 (0.006)

0.004 (0.008)

0.029 (0.012)**

0.86

68

0.195 (0.166)

0 (0.001)

0.006 (0.003)**

(0.0002) (0.009)

(0.001) 0

0.001 (0.004)

0.004 (0.004)

−0.001 (0.004)

0.01 (0.022)

Notes: Standard errors are in parentheses. The Hausman test rejects the hypothesis of systematic difference between FE and RE: Prob > Chi squared = 0.08. Significance levels: * at 10 percent, ** at 5 percent, and *** at 1 percent.

0.81

68

72

0.811 (0.115)***

0.395 (0.060)**

Constant

0.84

0 (0.001)

Replacement ratio

R-squared

0.002 (0.002)

Dependency ratio in 2050

0.395 (0.048)**

0.012 (0.007)*

Employment protection

0.012 (0.005)**

−0.004 (0.001)***

−0.004 (0.001)***

Tax on labor income

0.008 (0.014)

0.004 (0.014)

0.001 (0.015)

0.008 (0.014)

0.004 (0.014)

0.001 (0.014)

0.001 (0.007)

0.002 (0.008)

0.028 (0.015)*

0.028 (0.019)*

Acceptance of Competition (e039)

0.043 (0.062)

0.043 (0.078)

Acceptance of Change (e047)

What Light Is Shed on the Continent’s Problem? 457

−0.054 (0.081)

−0.054 (0.101)

0.002 (0.000)***

0.052 (0.027)*

0.076 (0.142)

0.12 (0.025)***

0.046 (0.015)***

0.139 (0.060)**

Achievement (c018)

Interestingness of Work (c020)

Involvement at Work (c031)

Importance of Work (c046)

Willingness to Follow (c061)

Acceptance of New Ideas (e046)

Freedom in Decisions (e053)

(c061)*(e053)

−0.016 (0.072)

−0.016 (0.067)

Initiative at Work (c016)

0.139 (0.067)*

0.046 (0.019)*

0.12 (0.036)**

0.076 (0.139)

0.052 (0.021)*

0.002 (0.000)**

GLS, White SE

OLS

Dependent variable: employment in percent of active population

Model 1: culture variables

OLS

GLS, White SE

Model 2: traditional variables

0.062 (0.023)** −0.027 (0.051)

−0.027 (0.047)

−0.065 (0.030)**

−0.065 (0.029)** 0.062 (0.020)***

0.012 (0.086)

−0.035 (0.015)**

−0.035 (0.014)** 0.012 (0.082)

0.001 (0.002)

0.36 (0.059)***

−0.267 (0.039)***

GLS, White SE

0.001 (0.002)

0.36 (0.059)***

−0.267 (0.041)***

OLS

Model 3: culture and traditional variables

Table 15.2 Determinants of economics performance: Percentage employed of active population

0.666 (0.064)***

−0.901 (0.062)***

−0.494 (0.038)***

−0.028 (0.011)**

0.463 (0.041)***

−0.33 (0.037)***

OLS

0.666 (0.068)***

−0.901 (0.047)***

−0.494 (0.039)***

−0.028 (0.009)***

0.463 (0.041)***

−0.33 (0.028)***

GLS, White SE

Model 4: narrowed

0.619 (0.137)***

−0.9 (0.139)***

−0.47 (0.082)***

−0.024 (0.023)

0.513 (0.082)***

−0.341 (0.083)***

RE

Model 5: narrowed, panel data

458 Edmund S. Phelps

0 (0.001) 0.096 (0.14)

0.095 (0.078)

68

0.84

Replacement ratio

Constant

Observations

R-squared

0.52

0.52

72

0.096 (0.217)

0 (0.001)

0.013 (0.003)***

0.96

0.96

68

−0.74 (0.075)***

−0.74 (0.087)*** 68

0.002 (0.000)***

0.022 (0.002)***

0.014 (0.005)**

−0.002 (0.001)***

0.006 (0.005)

−0.001 (0.005)

−0.013 (0.007)

0.009 (0.016)

0.026 (0.091)

0.002 (0.000)***

0.022 (0.002)***

0.014 (0.005)**

−0.002 (0.001)***

0.006 (0.005)

−0.001 (0.006)

−0.013 (0.006)*

0.009 (0.016)

0.026 (0.081)

0.95

72

−0.16 (0.089)*

0.001 (0.000)***

0.025 (0.002)***

0.032 (0.003)***

−0.002 (0.000)***

0.005 (0.006)

−0.002 (0.006)

−0.014 (0.006)**

0.288 (0.051)***

0.95

72

−0.16 (0.082)*

0.001 (0.000)***

0.025 (0.001)***

0.032 (0.003)***

−0.002 (0.000)***

0.005 (0.005)

−0.002 (0.005)

−0.014 (0.006)**

0.288 (0.044)***

0.94

72

−0.262 (0.179)*

0.001 (0.001)***

0.025 (0.003)***

0.033 (0.007)***

−0.001 (0.000)***

0.006 (0.004)*

−0.003 (0.004)

−0.016 (0.004)***

0.304 (0.113)***

Notes: Standard errors are in parentheses. The Hausman test rejects the hypothesis of systematic difference between FE and RE: Prob > Chi squared = 0.69. Significance levels: * at 10 percent, ** at 5 percent, and *** at 1 percent.

0.84

68

72

0.013 (0.002)***

Dependency ratio in 2050

0.095 (0.082)

0.018 (0.009)**

Employment protection

0.018 (0.007)**

−0.003 (0.001)***

−0.003 (0.001)***

0.001 (0.017)

Tax on labor income

0.006 (0.01)

Year 2005

0.001 (0.017)

−0.011 (0.017)

0.005 (0.018)

−0.006 (0.011)

−0.006 (0.01)

Year 1999

−0.011 (0.017)

0.005 (0.017)

−0.019 (0.011)

−0.019 (0.012)

Year 1996

0.006 (0.011)

0.061 (0.026)*

0.061 (0.029)**

Acceptance of Competition (e039)

0.061 (0.105)

0.061 (0.094)

Acceptance of Change (e047)

What Light Is Shed on the Continent’s Problem? 459

7.604 (1.529)***

−1.334 (1.309)

39.214 (8.044)***

7.604 (2.074)***

−1.334 (1.087)

0.66 (3.875)

Importance of Work (c046)

Willingness to Follow (c061)

Acceptance of New Ideas (e046)

Freedom in Decisions (e053)

0.66 (3.532)

39.214 (9.144)***

0.411 (1.279)

0.411 (1.208)

Involvement at Work (c031)

0.129 (0.020)***

0.129 (0.020)***

−11.762 (5.016)**

−11.762 (4.716)**

Achievement (c018)

Interestingness of Work (c020)

18.874 (3.649)***

GLS, White SE

18.874 (4.183)***

OLS

Initiative at Work (c016)

Dependent variable: employment in percent of labor force

Model 1: culture variables

OLS

GLS, White SE

Model 2: traditional variables

Table 15.3 Determinants of economics performance: Percentage employed of labor force

2.591 (1.655) −0.536 (3.833)

−0.536 (4.701)

−1.338 (2.763)

−1.338 (2.861) 2.591 (1.979)

33.077 (8.642)***

−2.429 (1.793)

−2.429 (1.4)* 33.077 (8.118)***

0.064 (0.024)***

−3.067 (6.163)

11.258 (3.548)***

GLS, White SE

0.064 (0.027)**

−3.067 (5.837)

11.258 (4.122)***

OLS

Model 3: culture and traditional variables

39.084 (6.936)***

−3.425 (0.894)***

0.073 (0.021)***

10.367 (2.539)***

OLS

39.084 (7.161)***

−3.425 (0.961)***

0.073 (0.018)***

10.367 (2.607)***

GLS, White SE

Model 4: narrowed

38.466 (13.688)***

−3.47 (1.808)*

0.069 (0.038)*

10.367 (5.192)**

RE

Model 5: narrowed, panel data

460 Edmund S. Phelps

72 0.46

77.583 (8.554)***

0.031 (0.029)

0.358 (0.141)**

0.22 (0.307)

–0.168 (0.023)***

−0.364 (0.611)

−0.108 (0.703)

−2.021 (0.909)**

68 0.81

50.17 (8.608)***

0.072 (0.029)**

0.725 (0.178)***

–0.633 (0.526)

–0.13 (0.060)**

−0.405 (0.542)

−0.173 (0.557)

−2.265 (0.570)***

−2.084 (1.597)

−12.371 (8.094)

68 0.81

50.17 (9.306)***

0.072 (0.036)**

0.725 (0.207)***

–0.633 (0.459)

–0.13 (0.046)***

−0.405 (0.526)

−0.173 (0.472)

−2.265 (0.574)***

−2.084 (1.688)

−12.371 (6.926)

68 0.78

60.374 (6.659)***

0.05 (0.026)*

0.688 (0.122)***

–0.069 (0.296)

–0.12 (0.033)***

−0.405 (0.548)

−0.196 (0.552)

−2.297 (0.556)***

−4.277 (0.947)***

68 0.78

60.374 (6.271)***

0.05 (0.026)*

0.688 (0.106)***

–0.069 (0.306)

–0.12 (0.029)***

−0.405 (0.519)

−0.196 (0.481)

−2.297 (0.602)***

−4.277 (0.826)***

68 0.77

60.261 (13.586)***

0.053 (0.051)

0.694 (0.247)***

–0.073 (0.605)

–0.128 (0.043)***

−0.405 (0.397)

−0.177 (0.408)

−2.271 (0.417)***

−4.224 (1.910)**

Notes: Standard errors are in parentheses. The Hausman test rejects the hypothesis of systematic difference between FE and RE: Prob > Chi squared = 0.99. Significance levels: * at 10 percent, ** at 5 percent, and *** at 1 percent.

72 0.46

68 0.72

Observations

68 0.72

77.583 (6.560)***

83.792 (4.746)***

Constant

R-squared

0.031 (0.029)

Replacement ratio

83.792 (4.254)***

0.358 (0.110)***

−0.364 (0.795)

Dependency ratio in 2050

−0.412 (0.546)

−0.412 (0.631)

Year 2005

−0.108 (0.796)

0.22 (0.405)

−0.453 (0.527)

−0.453 (0.631)

Year 1999

−2.021 (0.798)**

Employment protection

−2.653 (0.720)***

−2.653 (0.631)***

Year 1996

–0.168 (0.032)***

−1.854 (1.556)

−1.854 (1.499)

Acceptance of Competition (e039)

Tax on labor income

1.408 (6.398)

1.408 (6.102)

Acceptance of Change (e047)

What Light Is Shed on the Continent’s Problem? 461

−0.522 (0.339)

−0.001 (0.001)

0.542 (0.098)***

−0.317 (0.048)***

−0.202 (0.189)

−0.522 (0.272)

−0.001 (0.001)

0.012 (0.07)*

0.28 (0.465)

0.542 (0.120)***

−0.317 (0.063)***

−0.202 (0.224)

1.075 (0.352)***

Achievement (c018)

Interestingness of Work (c020)

Involvement at Work (c031)

Importance of Work (c046)

Willingness to Follow (c061)

Acceptance of New Ideas (e046)

Freedom in Decisions (e053)

Acceptance of Change (e047)

1.075 (0.375)***

0.28 (0.478)

0.012 (0.088)**

0.349 (0.186)**

GLS, White

0.349 (0.242)

OLS

Initiative at Work (c016)

Dependent variable: ln(Productivity)

Model 1: culture variables

OLS

GLS, White

Model 2: traditional variables

Table 15.4 Determinants of economics performance: Labor productivity

0.437 (0.194)** −0.169 (0.34)

−0.169 (0.29)

0.005 (0.087)

0.643 (0.116)***

0.344 (0.316)

0.212 (0.048)***

0.002 (0.001)**

−1.918 (0.258)***

1.049 (0.160)***

GLS, White

0.437 (0.168)**

0.005 (0.071)

0.643 (0.103)***

0.344 (0.299)

0.212 (0.050)***

0.002 (0.001)**

−1.918 (0.211)***

1.049 (0.148)***

OLS

Model 3: culture and traditional variables

0.229 (0.083)***

0.525 (0.050)***

0.176 (0.038)***

−1.523 (0.139)***

0.836 (0.135)***

OLS

0.229 (0.081)***

0.525 (0.052)***

0.176 (0.038)***

−1.523 (0.142)***

0.836 (0.116)***

GLS, White

Model 4: narrowed

0.285 (0.164)*

0.538 (0.105)***

0.172 (0.081)**

−1.577 (0.289)***

0.853 (0.286)***

RE

Model 5: narrowed, panel data

462 Edmund S. Phelps

−0.136 (0.037)***

−0.041 (0.034)

−0.136 (0.036)***

−0.041 (0.036)

0.056 (0.036)*

Year 1996

Year 1999

Year 2005

68

Observations 0.6

0.6

72

10.209 (0.294)***

0.002 (0.002)

0.011 (0.007)

0.92

68

0.92

68

12.507 (0.508)**

−0.002 (0.001)

−0.002 (0.001) 12.507 (0.408)***

0.001 (0.009)

−0.048 (0.009)**

−0.048 (0.008)*** 0.001 (0.008)

−0.15 (0.022)**

0.002 (0.003)

0.056 (0.020)***

−0.047 (0.018)**

−0.145 (0.021)***

0.331 (0.058)***

−0.15 (0.019)***

0.002 (0.002)

0.056 (0.020)***

−0.047 (0.021)**

−0.145 (0.024)***

0.331 (0.057)***

0.9

72

12.087 (0.247)***

0 (0.005)

−0.037 (0.006)***

−0.149 (0.011)***

0.057 (0.020)***

−0.041 (0.020)**

−0.131 (0.021)***

0.319 (0.037)***

0.9

72

12.087 (0.279)***

0 (0.005)

−0.037 (0.007)***

−0.149 (0.010)***

0.057 (0.021)***

−0.041 (0.017)**

−0.131 (0.021)***

0.319 (0.036)***

0.9

72

12.298 (0.458)***

−0.008 (0.005)

−0.041 (0.012)***

−0.154 (0.022)***

0.059 (0.013)***

−0.035 (0.013)***

−0.118 (0.015)***

0.306 (0.077)***

Notes: Standard errors are in parentheses. The Hausman test rejects the hypothesis of systematic difference between FE and RE: Prob > Chi squared = 0.88. Legend: at * 10 percent, ** at 5 percent, and *** at 1 percent significance level.

0.67

72

10.209 (0.355)***

11.482 (0.274)***

Constant

68

0.002 (0.001)

Replacement ratio

0.67

0.011 (0.009)

Unemployment

R-squared

0.013 (0.006)**

Dependency ratio in 2050

11.482 (0.293)***

−0.114 (0.019)***

−0.114 (0.019)***

Employment protection 0.013 (0.005)***

0.001 (0.002)

0.054 (0.041)

−0.051 (0.038)

−0.151 (0.040)***

0.001 (0.002)

0.054 (0.038)

−0.051 (0.039)

−0.151 (0.040)***

Tax on labor income

0.056 (0.037)*

0.085 (0.12)

0.085 (0.087)

Acceptance of Competition (e039)

What Light Is Shed on the Continent’s Problem? 463

−0.405 (0.257)**

0.407 (0.071)***

−0.246 (0.036)***

−0.106 (0.138)

−0.405 (0.201)**

0 (0.001)

0.024 (0.051)

0.309 (0.342)

0.407 (0.088)***

−0.246 (0.046)***

−0.106 −0.165

0.691 (0.260)**

Achievement (c018)

Interestingness of Work (c020)

Involvement at Work (c031)

Importance of Work (c046)

Willingness to Follow (c061)

Acceptance of New Ideas (e046)

Freedom in Decisions (e053)

Acceptance of Change (e047)

0.691 (0.277)**

0.309 (0.345)

0.024 (0.066)

0 (0.001)

0.309 (0.143)***

GLS, White SE

0.309 (0.178)*

OLS

Initiative at Work (c016)

Dependent variable: labor productivity ratio

Model 1: culture variables

OLS

GLS, White SE

Model 2: traditional variables

Table 15.5 Determinants of economics performance: Labor productivity ratio

0.354 (0.160)** −0.135 (0.287)

−0.135 (0.243)

−0.034 (0.074)

0.506 (0.100)***

0.371 (0.255)

0.17 (0.038)***

0.002 (0.001)

−1.394 (0.210)***

0.81 (0.139)***

GLS, White SE

0.354 (0.141)**

−0.034 (0.059)

0.506 (0.086)***

0.371 (0.25)

0.17 (0.042)***

0.002 (0.001)**

−1.394 (0.176)***

0.81 (0.124)***

OLS

Model 3: culture and traditional variables

0.251 (0.091)***

0.415 (0.041)***

0.133 (0.033)***

−1.197 (0.125)***

0.645 (0.111)***

OLS

0.251 (0.104)**

0.415 (0.044)***

0.133 (0.024)***

−1.197 (0.154)***

0.645 (0.098)***

GLS, White SE

Model 4: narrowed

0.321 (0.184)*

0.424 (0.101)***

0.123 (0.078)*

−1.255 (0.292)***

0.659 (0.274)**

RE

Model 5: narrowed, panel data

464 Edmund S. Phelps

0.017 (0.026)

−0.067 (0.026)**

0.017 (0.027)

−0.067 (0.027)**

Year 1999

Year 2005

0.001 (0.001) 0.341 (0.269)

1.288 (0.202)***

68

0.63

Replacement ratio

Constant

Observations

R-squared

0.53

0.53

72

0.341 (0.214)

0.001 (0.001)

0.008 (0.005)

0.88

68

0.88

68

2.016 (0.420)***

−0.002 (0.001)

−0.002 (0.001) 2.016 (0.341)***

0.003 (0.007)

−0.034 (0.008)***

−0.034 (0.006)*** 0.003 (0.006)

−0.103 (0.018)***

0.002 (0.002)

−0.068 (0.016)***

0.011 (0.016)

0 (0.018)

0.207 (0.044)***

−0.103 (0.016)***

0.002 (0.002)

−0.068 (0.016)***

0.011 (0.017)

0 (0.02)

0.207 (0.048)***

0.87

72

2.03 (0.273)***

0.87

72

2.03 (0.368)***

−0.001 (0.001)

−0.002 (0.004)

−0.002 (0.004) −0.001 (0.001)

−0.033 (0.007)***

−0.102 (0.010)***

0 (0.001)

−0.066 (0.016)***

0.019 (0.015)

0.017 (0.017)

0.232 (0.031)***

−0.033 (0.006)***

−0.102 (0.009)***

0 (0.001)

−0.066 (0.016)***

0.019 (0.016)

0.017 (0.018)

0.232 (0.031)***

0.87

72

2.235 (0.566)***

−0.002 (0.002)

−0.006 (0.004)*

−0.036 (0.012)***

−0.104 (0.023)***

−0.001 (0.001)

−0.065 (0.010)***

0.023 (0.010)**

0.026 (0.012)**

0.223 (0.075)***

Notes: Standard errors are in parentheses. The Hausman test rejects the hypothesis of systematic difference between FE and RE: Prob > Chi squared = 0.99. Legend: at * 10 percent, ** at 5 percent, and *** at 1 percent significance level.

0.63

72

0.008 (0.007)

Unemployment

68

0.009 (0.005)

Dependency ratio in 2050

1.288 (0.216)***

−0.079 (0.013)***

−0.079 (0.015)***

Employment protection 0.009 (0.003)***

0 (0.002)

−0.068 (0.031)**

0.01 (0.03)

−0.001 (0.031)

0 −0.001

−0.068 (0.029)**

0.01 (0.029)

−0.001 (0.031)

Tax on labor income

0.01 (0.028)

0.01 (0.027)

Year 1996

0.034 (0.085)

0.034 (0.064)

Acceptance of Competition (e039)

What Light Is Shed on the Continent’s Problem? 465

466

Edmund S. Phelps

both the variation over time in each country and the variation across countries at each time.7 The following paragraphs discuss the empirical results for each measure of economic performance. Male participation rate Table 15.1 shows under model 1 the pooled regressions on the cultural variables using OLS and GLS. In the GLS regression, Acceptance of New Ideas (e046), Acceptance of Competition (e039), and Interestingness of Work (c020) are all significant. Achievement-mindedness (c018) also proved significant. So, unsurprisingly, is the Importance of Work (c046). Initiative at Work (c016) and Involvement in Work (c031) look promising but do not survive in later columns. (Freedom in Decisions is discussed below.) It is striking that that the degree of explanation is so high—much higher than the explanation provided by the “traditional” explanatory variables in model 2. Furthermore the coefficients on the tax rate and the replacement rate are quite small, although that is also a characteristic of some of the culture variables. Model 3 is a regression that combines the cultural variables with the traditional variables. It cannot be called a fruitful marriage. But consider models 4 and 5, estimated with OLS, GLS, and a random-effects method, which have been narrowed beforehand to exclude the apparently failed cultural variables. There we find four survivors: Acceptance of New Ideas (e046), Acceptance of Competition (e039), Importance of Work (c046), and Interestingness of Work (c020). At the preliminary stages, the study included two additional measures of economic performance: labor force participation rate in percent of population between the age of 15 and 64 and female participation rate in the labor force. The former yielded results very similar to the ones reported here and so provide no additional insight. The female participation rate was not studied, since it reflects not only economic forces and culture but also attitudes toward gender differences that we do not know how to control for. Economic activity rate Table 15.2 on the economic activity rate regressions confirms the main findings of the previous one. Here again the cultural variables in model 1 outperform the traditional variables in model 2. In models 3, 4, and 5, Acceptance of Change (e047) again performs well. Achievement (c018) is again significant while Initiative at Work (c16) and Involvement in Work (c031) are wrong-signed. What is noteworthy here is the interaction of Willingness to Follow Instructions (c061) and Freedom in Decisions (e053) at work. The coefficients

What Light Is Shed on the Continent’s Problem?

467

of both these variables taken separately are negative and this is unambiguously the wrong sign. But these two variables might well be complementary. To allow for that possibility, the interaction term is introduced. That term turns out to be highly significant and has so large a positive coefficient that the total effect of each of the two variables is positive (at their mean values). Thus both Willingness to Follow and Freedom in Decisions are positive influences when their interaction is taken into account, Leadership without “followership” and, likewise, followership without leadership would contract employment but the two together operate to increase employment. Employment relative to labor force Table 15.3 records the results from regressions of employment in percent of the labor force on the cultural variables. The results here are mixed. In models 4 and 5, both Initiative at Work (c016) and Importance of Work (c046) are highly significant with large coefficients. Unfortunately, Involvement at Work (c031) and Acceptance of Competition (e039) are wrong-signed. These mixed results suggest that some other powerful forces are important drivers of inter-country differences in the unemployment rate. Labor productivity The performance of the cultural variables is generally excellent in Table 15.4. A highly interesting result of the analysis there is that neither the cultural group nor the traditional group performs well by itself. Yet when married in models 3, 4, and 5, the performance of both groups improves—especially that of the cultural variables. Initiative at Work (c016), Willingness to Follow (c061), Freedom in Decisions (e053), Involvement in Work (c031), and Acceptance of Competition (e039) are all highly significant with large, positive coefficients. Achievement (c018) is wrong-signed. Among the traditional variables, the employment protection variable and the old-age dependency ratio in 2050 are significant and have the expected signs. This specification includes the unemployment rate on the right-hand side with the aim of “adjusting” the measured productivity differences for the differing extent to which countries keep out of employment persons of low productivity and send them to the unemployment pool. Contrary to all past experience, that trick did not work here. Possibly, as the percentage of employed members of the labor force increases, a rising proportion of employed persons will be part of the variable workforce, manning the assembly lines and producing output, thus a decreasing proportion will be tied up in overhead tasks (Okun’s law).

468

Edmund S. Phelps

Productivity level as a ratio to the US level A potential drawback of the preceding productivity regressions is that that they do not take into account a possible catch-up process going on in which economies are tending to close or narrow the gap between themselves and the productive leader (also in the sample). To allow for this possible effect and to directly evaluate the potential forces leading to the gap, we ran regressions in which the dependent variable is the ratio of a country’s productivity level to the level of the leading economy, here that of the United States. (The ratio is 1 minus the percentage “gap.”) The results are presented in table 15.5. In a comparison of model 1 and model 2, the culture variables appear to perform at least as well as the traditional explanatory variables. Model 3 leads to an improvement in the performance of some of the culture variables and some of the traditional ones. In particular, Initiative at Work (c016), Willingness to Follow Orders (c061), Freedom in Decisions (e053), and Acceptance of Competition (e039) become highly significant and have the theoretically predicted positive sign. (Achievement (c018) remains negative and significant even after the marriage of the culture and mode traditional models.) As in the preceding tables, models 4 and 5 confirm the major results of model 3. In particular, a comparison between the narrowed model 4 and model 5 suggests that adopting the panel method mentioned above does not lead to considerable change in the estimated results. Among the traditional variables, employment protection and the expected dependency ratio in 2050 appear to have a significant negative effect on the productivity ratio, thus widening the gap. It is worth noting that the tax rate has lost its significance in both tables 15.4 and 15.5. I should confess that there exist some traditional variables, all going back to my own work, that are so powerful as to blow most of the other variables out of the water. One of these blockbusters is private wealth, as estimated from private saving rates. Another is a “social wealth” variable taken from social insurance outlay data. These last-minute findings lead me to believe that another variable, stock market capitalization, would also wreak havoc with all the coefficients. I suspect the same is true of the famous bureaucratic red tape index. Why is this? As remarked earlier, all the traditional variables here are to some extent a function of economic culture—their effects are the culture’s “indirect” effects—so introducing the controls awards them culture’s indirect effects, thus leaving the cultural variables with only their direct effects,

What Light Is Shed on the Continent’s Problem?

469

if they have any. Yet the results are pretty good because the traditional variables in the regressions are not sufficiently good proxies for culture to do much damage to the coefficients of the culture variables in the tables. But some traditional variables, such as the blockbuster variables, express so effectively a cultural mindset that is are highly correlated with the culture variables in the regressions that there is nothing much left for the individual culture variables to explain. 15.4

What Implications for the Continental Nations?

This section will have a preliminary look at what the data suggest, pending more systematic calculations and significance tests. We may reasonably infer from the detailed empirical results here that some particular cultural attributes, namely those with significantly positive regression coefficients, really do matter for economic performance in one or more respects. They are key attributes a deficiency of which in a country would operate to pull down its economic performance in the affected dimensions. If the nations on the Continent are deficient in some or all of the key (and not super-endowed where they are not deficient), that would help explain the widespread perception that the continental economies as structured now are “underperformers.” Is the Continent predominantly deficient in these key cultural attributes? The brief examination that follows is confined to comparing the cultural scores of the Big Three on the Continent with the usual comparators, the United States, the United Kingdom, and Canada. And the comparison is limited to a few cultural variables. Two culture variables have scored pretty well. Importance of Work (c046), which is so important for participation and unemployment, and Involvement (or Pride) in One’s Work (c031), which is important for productivity. (A close substitute, Interestingness of Work (c020), drove out involvement in the participation results of table 15.1.) In these respects many of us think of the Europeans as painstaking craftspeople, the Americans as more practical, so we would not be surprised if the Continent’s average scores on these two variables were comparable or better than those of the comparators. In fact, according to our survey data, the nationals on the Continent are deficient on these two scores. The data set shows that with respect to Importance of Work (c046), the Americans’ score of 0.17 tops Germany’s 0.11, Canada’s 0.11 tops Italy’s 0.08, and Britain’s 0.07 tops France’s 0.04. With respect to Involvement

470

Edmund S. Phelps

(c031), America’s 2.87 tops Italy’s 2.03, Britain’s 2.80 tops Germany’s 1.79, and Canada’s 2.70 tops France’s 1.74. This echoes de Tocqueville’s contrast in 1835 between the “tumultuous and boisterous gaiety” in aristocratic societies such as French society and the democratic Americans, who “prefer those more serious and silent amusements which are like business” (de Tocqueville 1835). In recent correspondence Richard Robb, whose financial business has taken him for many years to Japan and now to Europe, drew a similar contrast in the present day: continental Europeans are not nearly as immersed in their work and career as are the Japanese and the Americans.8 Thus Airaghi’s perception is borne out. Also powerful when their interaction was taken into account was another pair of cultural variables, Willingness to Follow Instructions (c061) and Freedom in Decisions (e053). The former (c061) delivered spectacularly: it raises productivity and even lowers the unemployment rate. On this score, the continentals score decisively below their comparators: America scores 1.47, Canada 1.34, and Britain 1.32; France scores 1.19, Germany 1.13, and Italy 1.04. With respect to the latter (e053), the United States scores 0.61, Canada 0.65, and United Kingdom 0.43. Germany has 0.57, France 0.57, and Italy 0.54. The aggregates are about equal, though the Continent loses the competition 2 matches to 1. Acceptance of competition (e039) appears to have a powerful effects on productivity, as hypothesized, and even on participation and thus, given the unemployment rate, employment. Here the United States scores 1.11, Canada 1.01, and the United Kingdom 0.57. Germany scores 1.21, thus topping the United States, while France has 0.68 and Italy 0.49. The preference for jobs offering Initiative at Work (c016) was also a significant cultural attribute in the productivity tables and, fitfully, in the participation table. On this culture attribute too the Continent’s Big Three is not dominated by the three comparators. Germany scores 0.59, beating Canada’s 0.55. Yet America’s 0.52 beats Italy’s 0.47, and Britain’s 0.45 beats France’s 0.38. Also the Big Three’s aggregate score is lower than that of its comparators. Conclusions The basic point to carry away, obviously, is that the empirical results lend support to the Enlightenment theme that a nation’s culture ulti-

What Light Is Shed on the Continent’s Problem?

471

mately makes a difference for the nation’s economic performance in all its aspects—activity as well as productivity. (Testing for the effects of elements of the culture directly on dynamism and also on reported job satisfaction are possible future steps in this line of research.) Thus a country’s initiation of a program to reform the institutional machinery with the aim of achieving a major improvement of economic performance—though a much-needed step—would, if undertaken alone, very likely succeed only to a degree and thus cause considerable disappointment. A transformation of the economy to one of dynamism, with the teamwork to implement it and to adapt well to it, can be obtained only if the economic culture and possibly other “background conditions” are conducive, not just the institutional machinery. This lesson, I note, does not hinge on any estimated “complementarity” between institutions and culture; none such was estimated here. It follows simply from an estimated linear equation and a data set having the feature that, on the whole, cultural variables get some of the credit (along with some representative institutional variables and other “controls”) for the better performances among the countries in the data set. (The added problem for institutional reform that the “interaction terms” of complementarity imply, where they are positive and significant, is that the effect of a new or improved institution may be small or nil unless that effect is potentiated by the emergence of a culture that can better exploit the institutional change.) An aspect of the results that are of particular interest to me is that every one of the cultural “dimensions” had at least one cultural variable representing it that performed significantly in at least one of the regressions. In the first dimension, Stimulation/Engagement/Development, the (proportionate) number reporting that their job is most important in their lives (c046) is significant both in raising male participation and (to a lesser extent) raising employment. (Yet it has no effect of its own on productivity, given all the other cultural attributes in the regression.) In the same dimension the pride taken in one’s work (c031) is more mildly labor force raising and more powerfully unemployment lowering. This Pride/Involvement in Work is seen as raising productivity as well. In the second dimension, Loyalty/Dutifulness/Altruism, the willingness to take a job that requires following instructions (c061) was the sole variable entering the regressions. It delivered spectacularly in combination with the Freedom variable (e053).

472

Edmund S. Phelps

In the third dimension, Individualism/Pluralism/Tolerance, it appears that Acceptance of Competition (e039) had powerful effects on productivity, as hypothesized, and even on participation, possibly through circuitous channels. Here the continental Big Three makes it a contest but as a group still loses badly to the comparators as a group. On e039, the United States scores 1.11, Canada 1.01, and the United Kingdom 0.57. Germany scores 1.21, thus topping the United States, while France has 0.68 and Italy 0.49. In the last dimension, Initiative/Venturesomeness/Experimentalism, two cultural attributes had considerable explanatory power. The preference for Initiative at Work (c016) was extremely significant in the productivity equations. It was significant also for the unemployment rate, boosting employment without boosting participation. The desire for Freedom in Decisions (e053), also dubbed here the Willingness to Assume Responsibility, perhaps to lead, was highly significant in the productivity equations. I would comment that in my previous work I had organized my thinking around the intellectual currents of reaction on the Continent to the Enlightenment and to capitalism in the nineteenth century: the solidarism, consensus, anticommercialism, and equalitarianism. It would be understandable if such a climate had a dispiriting effect on potential entrepreneurs. But, to be candid, I had not imagined that Continental Man might feel less entrepreneurial. It did not occur to me that Continental Man lacked an “entrepreneurial spirit,” or intellectual curiosity, or creativity. After all, this is a region that I treasured for the creativity of its Beethoven, Wagner, Picasso, and Keynes. In the early twentieth century Schumpeter was writing about the entrepreneurial spirit of Austrians and Weber that of the Germans! Apparently the Europeans’ creativity, once unmatched and perhaps so still, does not translate to business. Do the data then reflect “two cultures,” as argued by Bourguinon (2006) Or are the inter-country differences here purely random disturbances around the same all-West means? In fact variances are so low, owing to the large sample sizes, that the differences in scores between the Big Three and their comparators are statistically significant at stringent confidence levels. Such comparisons could easily be misunderstood, however. What is the meaning of the higher score in Germany? Perhaps it only means that the Germans, far more than the Americans, are deprived of opportunities for initiative. They have a craving for

What Light Is Shed on the Continent’s Problem?

473

additional initiative as a result—far more than the Americans do; thus initiative is on the mind of the Germans. If so, the Germans’ greater interest in those rewards of work does not imply that at the same level of opportunity they would value more initiative than the Americans. In short, the “value” expressed by the Surveys respondents are apt to be biased by their current conditions: in countries where there is deprivation of supply relative to the mean in the sample, the value attached to more is thereby increased, and as a possible result, respondents place more weight on that value; symmetrically, where is abundance relative to the mean, there is downward bias. (If respondents were asked whether in choosing the city to live in they put weight on water supply, they would say they don’t. It is a confusion between Smith’s “value in use,” i.e., total utility, and “value in exchange,” i.e., marginal utility.) That suggests that the true inter-country differences in reported values, in so far as what is being reported is the value of more, are apt to be much greater than the measured differences. In conversation on such comparisons my friend Jean-Paul Fitoussi made a related point. He thinks that the French economy is extremely hierarchical. At the micro level, the typical French company has many ranks, thus many layers of command and little latitude left to any one employee. At the macro level, there is the stratification by education, which predetermines the level to which a participant in the economy can rise. Both institutions, in limiting the range of responsibilities that a participant can have in the present and, generally speaking, over his or her career, are apt to cause French respondents in the Surveys to say that they want a position offering a lot of initiative and freedom—more so than respondents in other countries who, thanks the organization of their national economy, are already enjoying a lot of those rewards. Thus the relatively low (average) score that the French give to freedom in decision-making and to initiative is surprising and alarming. They ought to be rattling the bars of their cages.9 It may be, then, that the French, having long since despaired of having more freedom and more initiative, have learned not to care much about those values. Similarly, it may be that Americans, having assimilated large doses of freedom and initiative for generations, take initiative and freedom for granted. That appears to be what de Toqueville thought. (I have distilled his paragraph into one sentence.) The greater involvement of Americans in governing themselves, their relatively broad education and their wider equality of opportunity all encourage the emergence of the “man of action” with the “skill” to

474

Edmund S. Phelps

“grasp the chance of the moment” (p. 461). These thoughts suggest that even if deprivation can sensitize respondents to the value of initiative (or freedom or whatever reward is of interest), a long history of it may finally inure them to their relative deprivation. And a history of abundant reward in some respect may cause the respondents unthinkingly to underestimate its importance, even though they may act on its being there. So perhaps the responses in the Surveys can be taken at face value. To sum up this exploration of culture effects on the Continent: There is a loose correspondence between the continental countries’ relative endowment of some cultural attributes and the relative performance of their national economies in some if not all respects—though it is not yet clear how much of such effects are indirect through the culture’s impact on the nation’s selection of economic institutions and how much of such effects are direct. Yet, not all of the cultural attributes hypothesized to be important were found to matter for performance. And not all continental countries were under-endowed (some were well-endowed) in some of the cultural attributes that matter a lot. Two caveats: That continental countries tend to differ from comparators with regard to some cultural attributes—the Continent is “different”—does not compel us to agree with the opinion that the continental Europeans have chosen economic institutions that are different yet “optimal” for them, given those values.10 The values expressed by the continental Europeans do not contrast with those in comparator economies so radically as to suggest that the Continent would reject institutional changes demonstrated to deliver greater innovation and, as a result, higher productivity and a more rewarding workplace— notwithstanding some decrease in job security. The theme that big, even radical, innovations must come from the entry of start-ups (e.g., Schumpeter 1911; Arrrow 1960; Bhide 2000) and also, I think, the theme that the Continent’s corporatist institutions are inimical to dynamism in all companies, both new and established (Phelps and Zoega 2004) continue to be plausible guides to needed institutional reform on the Continent. We need not agree either that the continental Europeans have adopted the right values—right for them. It would be appropriate and possibly therapeutic if citizens in nations with unsatisfactory economic performance would compare their attitudes with those in other nations and ask whether they would not benefit from changing some of those values. That may be a long road. To embark on modifications of the

What Light Is Shed on the Continent’s Problem?

475

economic culture and the economic institutions to implement them would be a voyage of discovery—one having parallels with the “discovery procedure” that is the essence of capitalism. Appendix: Methodology for the Empirical Tests The objective of the empirical part of this study is to investigate whether variables reflecting economic culture, belonging to the four cultural currents, have causal effect on economic performance. The study faces the following two major objectives: first, to address concerns about the possible endogeneity of economic performance and culture; and second to cover a set of economically similar countries that is large enough to allow for meaningful econometric estimation. The empirical study includes the following OECD countries: Austria, Belgium, Canada, Denmark, Finland, France, Germany, Iceland, Ireland, Italy, Japan, the Netherlands, Norway, Poland, Portugal, Spain, Sweden, the United Kingdom, and the United States. The database comprises cross-sectional data for 1993, 1996, 1999, 2002, and 2004. Economic performance is measured by labor productivity, the ratio of a country’s productivity to the highest productivity level in each cross section, male participation rate in the labor force, employment in percent of population between the age of 15 and 64, and employment in percentage of labor force. In addition, the empirical results depend on one crucial assumption and one key property of the culture variables in the World Values Surveys (WVS). First, the assumption is that past beliefs do not affect present economic performance (a variation of Markov equilibrium). It is conceivable that present beliefs and expectations about future beliefs may affect present performance. However, by this assumption, past beliefs do not enter directly the model determining economic performance at present. Second, beliefs are relatively stable over time. This is the prevailing view among scholars in the field. Moreover it is also empirically supported by the high persistence in the values of the variables in the Surveys that have longer time series. Combining these two points, one can use culture variables from the 1990 to 1993 survey as proxies for the culture variables at time 1996, 1999, 2002, and 2004; the culture variables from 1990 to 1993 do not belong directly to the model, but are correlated with the nonobservable culture variables for the period 1996 to 2004. This setting addresses possible reverse causality concerns: the future (productivity at time

476

Edmund S. Phelps

1996 to 2004) cannot cause the past (culture in 1990 to 1993). The assumption that the culture variables from 1990 to 1993 do not belong directly to the models determining performance in 1996, 1999, 2002, and 2004 also implies that the culture variables from 1990 to 1993 are not correlated with the error term in models of economic performance in subsequent periods. On the basis of these methodological assumptions, the empirical study proceeds in the following stages. First, the departing point of the study is a set of simple OLS regressions of the measures of economic performance on culture variables. Second, control variables for the structure of the labor market, government intervention, market forces are introduced into the model. Third, time dummies are included in the OLS models to account for short-run fluctuations and time-specific shocks. The structure of the data set, however, leaves no doubt that the OLS regressions can be only suggestive. There are two ways to address the implications arising from the panel structure of the data set. One approach is to use an estimation procedure that is consistent in the presence of heteroskedasticity. The White estimation procedure provides consistent and asymptotically efficient estimates. This procedure is very attractive because it does not impose any particular assumptions on the type of the heteroskedasticity. However, in small samples the White estimation procedure is less efficient than WLS. The WLS relies on some restrictive assumptions on the structure and form of the heteroskedasticity. Most important, the econometricians must find the proper weights to eliminate the heteroskedasticity. Unfortunately, in the present study, finding these weights proved to be an insurmountable task, so the regression tables report only the White estimation results. Another approach is to use panel data analysis. Its major advantage is that it directly accounts for the panel structure of the data set and imposes fewer (heroic) assumptions about the structure of the data than OLS or heteroskedasticity-adjusted LS. Unfortunately, it also has drawbacks in the context of this study. Most important, since 1990 to 1993 culture variables are used as proxies for the 1996 to 2004 cross sections, fixed effect estimation is not useful as it is based only on within group variation over time. Thus the significance of the culture variables can be tested only in a random-effects setting. The random-effects estimation, however, relies on heavy assumptions about the structure of the error terms, whose appropriateness should be verified by a Hausman test. Moreover, even if there is no

What Light Is Shed on the Continent’s Problem?

477

systematic difference in the coefficient estimates under fixed and random-effects estimations, a random-effects regression incorporates only a fraction of the cross-sectional variation, which is where the potential evidence for the significance of culture resides. Thus, given the small sample size, it will not be surprising if the panel regressions find fewer culture variables to be significant than the pooled regressions do. To summarize, contingent on favorable Hausman test results, random-effects estimation provides a conservative estimation of the effect of culture on economic performance. Data: Sources and Definitions Labor productivity is based on the Penn World Tables and the productivity growth reported by the OECD for 2002 and 2004. All other variables come from the OECD or the UN. The reason to choose the Penn World Table productivity estimates is that unlike most other sources, these are based on careful cross-country PPP adjustments. As a result the Penn estimates are less contaminated by methodologically induced noise. Here is the way the variables are defined in the regressions, along with a short reference to the sources: Productivity Output per worked hours over one calendar year, from the Penn World Tables. Labor income tax Average labor income tax for a two-parent family with children, from the OECD. Dependency ratio in 2050 Proportion of potentially active population between the age of 15 and 64, based on UN estimates. Employment protection index (EPL) OECD index that is averaged for the period 1987 to 2004, from the OECD. The averaging is a standard procedure applied in research based on the index, since only several countries have experienced marginal changes in their level of employment protection over the last twenty years. Culture variables Values retrieved from the World Values Survey 1981 to 2004.

478

Edmund S. Phelps

Table 15.A1 Measures of dynamism

Country

Decisionmaking freedom at work

Turnover of listed firms

Patents granted per working age person

R&D intensity adj. for industry structure

United States

7.4

118%

3.7

2.9

Canada

7.2

106%

1.3

1.8

United Kingdom

7.0

65%

0.8

1.9

France

6.4

79%

0.9

2.2

Italy

6.7

63%

0.4

1.0

Germany

6.1

42%

1.5

2.2

Notes: Decision-making freedom at work is measured on a scale from 1 to 10, 10 highest, averaged for 1990 to 1993 (Human Beliefs and Values: Inglehart et al.); turnover of listed firms represents the number of exits from and entries into each country’s MSCI National Stock Index from 2001 to 2006 as a % of the number of firms in 2001; patenting data is averaged for 1990–2003 (World Intellectual Property Organization); R&D intensity adjusted for industry structure is the average % of business sector value added for 1999–2002 using the G7 industry structure (OECD).

2.9

2.7

2.8

1.7

2.0

1.8

Country

United States

Canada

United Kingdom

France

Italy

Germany

7.0

7.3

6.8

7.4

7.9

7.8

Job satisfaction, 1990–1993

79%

76%

76%

85%

85%

85%

Male labor force, % of workingage men, 2003

91%

91%

90%

95%

92%

94%

Employment, % of the labor force, 2003

$23,946

$21,822

$24,192

$22,008

$23,751

$31,994

Labor compensation per worker, 1996

92

—

92

73

—

100

Market output per hour, 1992

Notes: Pride derived from one’s job is measured on a scale from 1 to 3, 3 being the highest, and job satisfaction on a scale from 1 to 10, 10 highest; both are averaged for 1990 to 1993 (Human Beliefs and Values; Inglehart et al.); men in the labor force in percentage of working-age men and employment in percent of the labor force are computed for 2003 (OECD); labor compensation per worker is computed as the ratio of total compensation to the labor force using 1996 data (Extended Penn World Tables); market output per hour worked is for 1992 (Solow and Baily).

Pride derived from the job, 1990–1993

Table 15.A2 Benefits of dynamism

What Light Is Shed on the Continent’s Problem? 479

480

Edmund S. Phelps

Notes McVickar Professor of Political Economy and Director, Center on Capitalism and Society, Columbia University. I dedicate this paper to Viviana Montdor Phelps, who has often commented to me on cultural differences among countries and had the idea that this paper should study the importance of such differences. I am indebted too to Jean-Paul Fitoussi and David Jestaz for many conversations on the subject. The paper has benefited enormously from the talents and efforts of my research assistants Raicho Bojilov and Luminita Stevens. Lastly, I acknowledge with gratitude a grant for this research from the Kauffman Foundation. 1. Yet, theoretically, welfare entitlements could be largely neutral for activity. When paychecks are driven down by the social charges, saving and ultimately the stock of private wealth is driven down too, and these two forces work in opposing directions, the one reducing work and the other increasing it (see Hoon 2006). Employment in the United Kingdom and Ireland has in fact been strong despite the large welfare outlays. 2. Research in the late 1990s compared the breadth of institutions in boomers with that in non-boomers (Phelps 2000). Most of the early ideas were ultimately published in book form in my Enterprise and Inclusion in the Italian Economy (2002). Subsequently economic institutions were used to explain inter-country differences in the levels of structural performance characteristics, such as productivity and male participation rate, in Phelps and Zoega (2001), Phelps (2003a), and Phelps and Zoega (2004). 3. Phelps (2003b) and Eggertsson (2005). 4. In Schumpeter’s model, the entrepreneurial ideas are always “in the air,” so it was more a matter of the right entrepreneur being in the right place and the right time than it is a matter of his/her creativity. 5. A column in The Wall Street Journal told of a deliveryman who was asked whether he found it best to work from the top floor down or the reverse. “It depends on the time of day,” he replied. A perfect Hayekian moment! 6. The code in parentheses reflects the encoding system used in the consolidated database of World Value Surveys 1981–2004. 7. The Hausman test rejects the null hypothesis of systematic differences between the fixed effects and random effects estimates. So the random effects estimates are consistent and efficient. Accordingly the tables report the random effects specification. Appendix A contains more details on the estimation methodology as well as the data sources. 8. “Somehow, European employees seem to have great difficulty identifying personally with their firm. They see their jobs as contracts for services and do not care beyond the terms of the contract. (On a flight) it is always interesting to see the American business people . . . with their communications devices and their self-help books. I’m sure the content of these books is nonsense, but to me it illustrates a point: they care in a personal way about their work. This is why they chose to read. I have the sense the Europeans generally do not. I think this kind of personal engagement is necessary for entrepreneurship . . .” (email, July 15, 2006). 9. Unless, of course, Jean-Paul has got the conditions reversed: it is the American employees who operate with a tight leash. My friend Roman Frydman said to me that he was awed at the authority of Air France clerks to take actions far beyond the authority given to their counterpart at Delta Airlines.

What Light Is Shed on the Continent’s Problem?

481

10. A colleague of mine once made the cruel remark that “a country gets the economic system it deserves.” As the book by Eggertsson and my paper (2003b) argued, some and perhaps most of the institutions a country adopts are apt to be a non-optimal fit with its values. The problem is that it is impossible to infer all the maladies and the remedies.

References Aghion, P., and P. Howitt. 1998. Endogenous Growth Theory. Cambridge: MIT Press. Aghion, P., and P. Howitt. 2005. Appropriate growth policy: A unifying framework. Schumpeter Lecture. Nice, September; revised, December. Arrow, K. J. 1962. Economic welfare and the allocation of resources for invention. In R. R. Nelson, ed., The Rate and Direction of Innovative Activity. Princeton: Princeton University Press. Banfield, E. 1958. The Moral Basis of a Backward Society. New York: Free Press. Bhidé, A. 2000. The Origin and Evolution of New Businesses. Oxford: Oxford University Press. Bourguinon, P. 2006. Deux èducations, deux cultures. Le Cercle des economistes, L’Europe et les Etats-Unis. Paris: Descarte et Cie. Eggertsson, T. 2005. Inefficient Institutions. Ann Arbor: University of Michigan Press. Hayek, F. 1948. Individualism and Economic Order. Chicago: University of Chicago Press. (Edition contains “Socialist Calculation I, II,” of 1935, “Economics and Knowledge” of 1937, and “The Use of Knowledge in Society” of 1945.) Inglehart, R. 2006. World Values Surveys 1981–2004. Ann Arbor: University of Michigan. Phelps, E. S., ed. 1973. Altruism, Morality and Economic Theory. New York: Basic Books/ Russell Sage Foundation. Phelps, E. S. 2000. Europe’s stony ground for the seeds of economic growth. Financial Times, August 1. Phelps, E. S. 2002. Enterprise and Inclusion in the Italian Economy. Dordrecht: Kluwer. Phelps, E. S. 2003a. Commentary II. In P. Aghion, R. Frydman, J. Stiglitz, and M. Woodford, eds., Information and Knowledge in Modern Macroeconomics. Princeton, Princeton University Press. Phelps, E. S. 2003b. Globalization and development. Public lecture in Bangkok, Beijing, Tianjin, Shanghai, Singapore, and Havana, January–February. Phelps, E. S., and G. Zoega. 2001. Structural booms. Economic Policy 25 (April). Phelps, E. S., and G. Zoega. 2004. Searching for routes to better economic performance in continental Europe. Forum 5, no. 2. CESifo, Munich. Putnam, R. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, Princeton University Press. Schumpeter, J. A. [1911] 1932. Theorie der wirtschaftlichen Entwicklung (Theory of economic develpment), Vienna. Reprint R. Fels, trans., Theory of Economic Development. Cambridge: Harvard University Press, 1932.

482

Edmund S. Phelps

Schumpeter, J. A. [1912] 1926. Theorie der wirtschaftlichen Entwicklung (Theory of economic develpment), 2nd abridged edn. (ch. 2 expanded and ch. 7 omitted). Leipzig: Humboldt. Titmuss, R. 1970. The Gift Relationship: From Human Blood to Social Policy. London: Allen and Unwin. De Tocqueville, A. [1835] 1981. De la Democratie, 2 vols. (trans. 1840). Paris. Reprint T. Bender, ed., Democracy in America. New York: Random House. Weber, M. [1905] 1930. Die Protestantische Ethik und der “Geist” der Kapitalismus. Sozialwissenschaften und der Sozialpolitik, 20 and 21. Reprint T. Parsons, trans. (foreword by R. H. Tawney), The Protestant Ethic and the Spirit of Capitalism. London: Allen and Unwin; New York: Scribner.

Index

AABM (Aghion, Angeletos, Banerjee and Manova), 224, 227 Abilities cognitive, 289, 290 and achievement vs. ability, 313n.13 noncognitive, 289, 290 ABRR (Aghion–Bacchetta–Ranciere– Rogoff), 227 Absorptive capacity, 171 Accountability and corporate entrepreneurship, 54–56 effects of, 40–46 Activating Social Assistance, 424–27 Active labor market (ALM) programs or policies, 254, 257, 266, 270, 291, 308 in three groups of countries, 393 in Nordic countries, 396 and poverty rate, 395, 396 Administrative burdens, as business environment area, 103, 129 Agency problems, in venture–capital relationships, 141 Airaghi, Angelo, 454, 470 Allen, Paul, 215n.9 “American Competitiveness Initiative,” 174–75 American Research and Development, 141 Angel capital, 17, 127 as business environment area, 127 Animal spirits, and entrepreneurship, 135, 147 Anticommercialism, 23 AR(1) MCMC method, 237, 238, 244, 246–48 AR(1) process, 233, 234

Arrow–Debreu equations, 449–50 Australia, 388. See also English–speaking economies Austria, 388. See also European Union, core countries of as “the better Germany,” 156 Austrian school, 20 Bank for International Settlements, 87 Bankruptcy laws as business environment area, 103, 129–30 and innovation deficit in Europe, 88 Banks entrepreneurial projects in, 327, 328, 329 and family businesses, 163 and German under–capitalization, 160 guaranties for entrepreneurial loans from, 127 Mediobanca (Italy), 162 mergers of, 197 as risk averse, 53 in solidarist–influenced system, 452 as source of start–up capital, 42 venture capital firms contrasted with, 141 Baverez, Nicolas, 156 “Bazaar effect,” 21, 418, 423 Behavioral research, 185 Belgium, 388. See also European Union, core countries of Bell, Alexander Graham, 216n.10 Benedetti, Carlo de, 75 Benedict, Ruth, 447

484

Benefit entitlements or schemes need to reduce, 309 and unemployment, 273–75, 276 Bergson, Henri–Louis, 454 Berlusconi, Silvio, 156–57 Berner–Lee, Sir Timothy, 190–91 Bettio, Francesca, 161 Big–box retailers, 197, 204, 374, 375 Big Pharma, 207 Bismarck, Otto von, 153, 158–59 Black (underground) economy, 377 Bloomberg LP, 87 Boeing aircraft company, 88 Boycko, Maxim, 138 Bretton Woods system, 27 Britain. See United Kingdom Brown, Gordon, 175 Budget deficits. See Cyclicality and countercyclicality of budget deficit among OECD nations Bureau Van Dijk (BvD), 97–98, 107 Bush, Vannevar, 146 Bush administration (second), and technological leadership, 174 Business angels, 17, 127 as business environment area, 127 Business education, traditional, as business environment area, 103, 126 Buyouts, as business environment area, 103, 128 Buyouts, leveraged, 60n.9 Canada, 388. See also English–speaking economies and family firms, 163 Capital access to, 40–46 from business angels, 127 vs. venture capitalists, 17 as entrepreneurship factor, 101 and factor price convergence, 415 and family businesses, 163 Capital exports, as pathological overreaction, 422 Capitalism continental vs. “Anglo–Saxon” model of, 161 family, 160 “Nordic model” of, 154 in Russia, 157

Index

Schumpeter on creative destruction in, 178–79 and thrift, 203–204 Weber on, 208 Capitalism, Socialism and Democracy (Schumpeter), 181 Capital markets and entrepreneurships, 47–48, 49 international, 87 in model of transmission of ideas, 331, 338 and retreat from joint–stock company, 164 Capital taxes, as business environment area, 103, 128 Carmel Ventures, 86 Case, Karl, 143 Case–by–case approach, 206 CERN lab, Switzerland, 190–91 Chandler, Alfred, 160, 162, 163–64 Chile, 154 China family firms in, 162 as globalization winner, 153–54 Chirac, Jacques, 156 Cisco as “Chinese company,” 174 ERP at, 214 nonprofits in start of, 215n.9 CME Group, 143 Cognitive abilities, 289, 290 and achievement vs. ability, 313n.13 Columbia Business School, 186, 217n.16 Columbia Capital, 86 Communal jobs, 426. See also Employer of last resort Communication about heroes, as business environment area, 103, 130 Communist economic system, collapse of, 413 Comparative advantage, loss of, 176 Comparison of standard of living in Europe and United States. See Standard of living in Europe compared with United States Compensation, executive, 56–59 Compensation of labor. See also Wages in Anglo–Nordic vs. major continental countries, 453 on Continent, 9–10

Index

Competition and debt–ridden environment of English–speaking countries, 405 and innovation, 169 in scientific research, 145 Composite indicators, handbook on, 102 Compton, Karl, 141 Confucian values, and Asian economic progress, 151 Conspicuous consumption, 203 Consumption, Euler equation for, 296 Consumption, venturesome, 19, 170, 171, 182–88, 192–93, 202–205, 208 of IT users, 199–201, 204 Consumption patterns, for IT (United States compared), 193–95, 196 Control, and corporate entrepreneurship, 54–56 Convergence theories, 177, 192 Corporate environment, entrepreneurship in, 48, 50 accountability and control in, 54–56 and executive compensation, 56–59 and ownership structure, 50–53 Council of the European Union, Research Council formed by, 147 Creative destruction and capitalism (Schumpeter), 178–79 in Nordic economies, 388 Creative Technology, Ltd. 192 Credit multiplier, 224 Cultural factors. See also Economic culture and cross–country differences in attitudes toward risk, 139–40 and cross–country differences in high–growth firms, 99 and cross–country differences in innovation and entrepreneurship, 136–39 and economic performance, 151 and entrepreneurship, 37, 135, 323 and innovation deficit in Europe, 88 in social prosperity, 28 Customer engagement, in product development, 183 Customers, venturesome and resourceful, 170. See also Consumption, venturesome

485

Cyclicality and countercyclicality of budget deficit among OECD nations, 223, 225, 227, 234–37, 245–46 data on, 227–28, 229 determinants of countercyclicality, 225, 226–27, 237–40, 245–46 and growth, 223, 224, 224–25, 225, 228, 230–31, 240–43 computation of countercyclicality for, 231–34 robustness tests for, 243–45 Daimler, Gottlieb, 216n.10 Danish entrepreneurship policy, 102 David, Paul, 171, 172, 197, 204, 206 Debt finance, 40 Dell, Michael, 215n.9 Demand for labor as shifted toward skilled workers, 307 skill–based, 264–65 Democracies, procyclical fiscal policy for, 225–26 Demographics as adversely affecting economic performance, 1 and unemployment, 13 Denmark as Nordic country, 387 (see also Nordic countries) policy challenges for, 113 Derivatives, 87 index–based, 142–44, 148 Diamond, Jared, 11 Disability spending, 270–71 Doerr, John, 73 Dolan, John, 143–44 Doriot, Georges F., 141 Duryea brothers, 216n.10 Dynamism, economic. See Economic dynamism Early childhood interventions, 290–91 Early–retirement schemes, need to correct, 309 Earned Income Tax Credit, US, 425 EC. See European Union, core countries of Econometric methods, for computing countercyclicality, 231–34 Economic activity rate, in statistical tests of economic culture, 466–67

486

Economic culture. See also Cultural factors dimensions of, 453–55 and economic performance in continental Europe, 22–23, 447–50, 470–75 conceptual framework for study of, 450–55 data sources and definitions in, 477 and economic dynamism, 452–53, 478–79 statistical methodology in study of, 475–77 statistical tests on, 455–70 and “two cultures” view (Bourguinon), 472–73 and institutions, 449 Economic development, and technology adoption, 171 Economic dynamism, 451, 452–53. See also Innovation and accountability structures, 56 and effect of culture on performance in continental Europe, 452–53, 478–79 and European slowdown, 15–16, 450 as type of economy, 29 Economic freedoms, and economic performance, 12 Economic growth from adoption of entrepreneurial innovation, 330, 338 and human capital, 332 and macroeconomic policy, 223–24 and taxation, 387, 388, 407 supply–side arguments on, 387, 402–406 Economic growth in Europe, 27 and budget deficit cyclicality (OECD), 223, 224, 224–25, 225, 228, 230–31, 240–43 computation of countercyclicality for, 231–34 robustness tests for, 243–45 vs. United States, 28–31 and entrepreneurship, 31, 33 (see also Entrepreneurship) Economic history(ies) and European development, 11–2 periods of change vs. periods of consolidation in, 29

Index

Economic institutions and economic performance on Continent and dynamism, 16 as suboptimal, 3 Economic performance in continental Europe, 323 and belief in merit vs. connections, 17 comparative evidence on, 3–13, 25n.11 disappointing decade for, 1 and economic culture, 22–23, 447–50, 470–75 conceptual framework for study of, 450–55 data sources and definitions in, 477 and economic dynamism, 452–53, 478–79 statistical methodology in study of, 475–77 statistical tests on, 455–70 and “two cultures” view (Bourguinon), 472 “glorious years” of (1955–1975), 10–11, 14, 448 and lack of economic dynamism, 15–16, 323 subjective indicators of, 24 and welfare state, 12, 20–22 Economic systems and economic performance, 10 ever–changing complexity of, 210–11 and interventionism in economic policy, 448 neoclassical economic theory on, 1–2 spectrum of, 450–52 Edison, Thomas, 216n.10 Education funding of, 262–64 greater priority for preschool and primary, 309 and importance of early years, 289–90 increase in attainment of, 259–62 and innovation deficit in Europe, 88 liberal, 449 in model of transmission of ideas, 331 and adoption of foreign technologies, 325 and productivity growth, 330 rising returns to, 258–259 and technological diffusion, 201–202 and unemployment, 270 upstream innovation served by, 209

Index

Education, business, as business environment area, 103, 126 Education, higher doctorates granted in US and rival countries, 173–74 and growth (model of transmission of ideas), 336, 338 subsidies for, 303–304 Education in entrepreneurship, as business environment area, 103, 126 Efficiency, vs. equity, 308 at later vs. earlier stages of life, 254 Eggertsson, Thrainn, 3 Elbion AG, funding of, 76 Emerging market economies, and countercyclical fiscal policy, 226 Employee engagement, in various countries, 24. See also Job satisfaction Employer of last resort, state as, 396, 398. See also Communal jobs Employment protection and growth (model of transmission of ideas), 336 in statistical tests on economic culture, 457, 459, 461, 463, 465 in US vs. Europe, 38–40 Employment relative to labor force, in statistical tests of economic culture, 467 Energy, and real GDP in Unied Stataes, 372–76 English–speaking economies (ES), 388. See also Taxation–growth relation in three groups of countries and fiscal distress, 405 as free societies, 403–404 government outlays in, 389, 391, 392 household saving rates of, 405 income inequality in, 402 per capita income of, 401 unemployment in, 398, 399 Enlightenment, 447, 470–71, 472 Enterprise Resource Planning (ERP) software, 186, 209–11, 211–14 Entitlements. See Welfare systems and states Entrepreneurial business environment and creation of high–growth firms, 104–105, 107–10 defining and quantifying, 100–104, 124–30

487

and macroeconomic conditions, 101–102 quality of data for, 124 in EU vs. US, 105–6 Entrepreneurial firms, as innovation users, 170 Entrepreneurial motivation as business environment area, 103, 130 as EU policy challenge, 112 Entrepreneurship, 31, 91, 92–93, 135, 147 as choice access to capital and effects of accountability in, 40–46 and capital markets, 47–48 and labor market, 37–40 and risk–taking differences (US vs. America), 33–36 and continental creativity, 472 as corporate risk taking, 48, 50 accountability and control in, 54–56 and executive compensation, 56–59 and ownership structure, 50–53 and cultural factors, 37, 135, 323 cross–country differences in, 136–39 and economic culture, 23 and economic performance, 15–16 factors affecting, 100–102, 323 and favorable circumstances, 451, 480n.4 intuitive decisions in (“tacit knowledge”), 32, 41 in model of transmission of ideas, 327–29 and number of start–ups (EU and US), 94–97 and personal engagement, 480 and risk management, 135, 141–44, 147–48 and scientific enterprise, 136, 144–47, 148 in US vs. continental Europe, 91, 92, 323 and US incomes lead, 192 venture capital role in, 65 (see also Venture capital firms) Entrepreneurship education as business environment area, 103, 126 as EU policy challenge, 111, 112, 113–14 Entrepreneurship infrastructure, as business environment area, 103, 127

488

Entry barriers/deregulation, as business environment area, 103, 125 Equality, vs. incentives, 256–58 Equity vs. efficiency, 308 at later vs. earlier stages of life, 254 Equity–based financing, 40–41 ES. See English–speaking economies Ethnic homogeneity, and social welfare state, 406 Euro.nm (stock market consortium), 48 Europe as globalization loser, 153 and governance structure of firms, 99–100 human capital in, 27, 254 insufficient utilization of, 253–54 policy on, 309 and skill use or maintenance, 255–56 as large open economy (payroll tax– hours worked relationship), 440–42 science funding in, 146, 147 unemployment in, 257, 275 use of advanced products in, 201 venture capital finance in, 66, 74–75, 88, 142 as EU policy challenge, 111, 112, 113–14 and European Commission, 65–66, 86, 142 and France–Luxembourg connection, 78–79, 80–81 and tax structure, 75–76, 77 and United Kingdom, 76–78 Europe, economic performance in. See Economic performance in continental Europe. European Central Bank, 223–24 European Council, and entrepreneurship, 91 European Institute of Technology, 175 European Investment Fund (EIF), 142 European Monetary Union (EMU) budget pressures for governments in, 262 and cyclicality of fiscal policy, 226 of budget deficit, 223, 225, 227, 236, 237, 239–40 and growth, 246 European Research Council (ERC), 147 European Stability and Growth Pact, 223

Index

European standard of living compared with United States. See Standard of living in Europe compared with United States European Union (Commission) enlargement of, 414 entrepreneurship promoted by, 91, 113 and business environment, 104–105, 107–10 and business environment compared with US, 105–106 and definition of entrepreneurship, 93 and high–growth firms, 99–100 and number of start–ups, 94–97 policy challenges in, 111–13, 114 questions on, 91–92 and relative importance of policy areas, 106–10 and relative importance of policy areas (data description), 114–23 fiscal discipline in, 157 IT productivity gains of, 198–99 and Lisbon agenda, 153 number of high–growth firms generated in, 97–99, 100 research spending in, 175 and venture capital, 65–66, 86, 142 European Union, core countries of (EC), 388, 408. See also Taxation–growth relation in three groups of countries and fiscal distress, 405 as free societies, 403–404 GDP of, 401 government outlays in, 389, 391, 392 household saving rates of, 405 income inequality in, 402 unemployment in, 398, 399 Euro–pessimists, 387, 388, 408 EUROSTAT Business Demography Project, 94 EUROSTAT database, 92, 96 Executive compensation, and corporate entrepreneurship, 56–59 Exit markets, as EU policy challenge, 111, 112, 113–14 Factor price convergence, 413, 414–15, 427 forces of, 415–18 Faggio–Nickell puzzle, 430, 443 Falck, Giorgio Enrico, 164 Family, 152

Index

Family firms, 160–66 of Europe, 18 high growth unlikely in, 99–100 as risk averse, 61n.13 risk diversification lacking in, 50 Federal Express, 215n.5 Female labor force participation rates, 284, 308, 312–13n.10 in Europe, 351, 364–65 and home production practices in Europe vs. United States, 369, 381 and participation of older workers, 287 and part–time jobs, 273 in United States, 349, 351 Financial development, and cyclicality of structural investments, 224, 241 for budget deficits, 238, 239, 245 and relation to growth, 228, 240, 243 Finland, 387. See also Nordic countries Fiorina, Carla, 164 Fiscal discipline, 157–58 Fiscal distress, as argument against welfare state, 403, 404–405 Fiscal incentives, as business environment area, 103, 128–29 Fiscal policy. See Cyclicality and countercyclicality of budget deficit among OECD nations Fiserve, Inc., 143 Fitoussi, Jean–Paul, 473 Foreign ideas, adoption of, 326–27 Foreign markets, access to, as business environment area, 103, 125 FoxMeyer, 214 France, 388. See also European Union, core countries of family firms in, 162 malaise and decline in, 154–55, 156 Franklin, Benjamin, 137 Freedom, as argument against welfare state, 403–404 Freeman, Richard, 173, 174, 175, 176, 177, 188, 190 French Revolution, and innovative culture, 136–37 Friedman, Milton, 403 Gaenslen, Fritz, 138 Gains from trade after collapse of Communism, 413 reality of, 418–19

489

Gates, Bill, 215n.9 “Gazelles,” 93, 97 GDP, per person of working age, 7, 8 GED recipients, and noncognitive abilities, 290 Geographic matching, of VCs and entrepreneurs, 79, 81–86 German model, of corporate governance, 55 German university model, 137 Germany, 388. See also European Union, core countries of export performance of, 423–24 low level of growth in, 418–19, 424 malaise in, 154–56 and outsourcing, 417–18 as scientific–research leader, 145 Giersch, Herbert, 12 Global Entrepreneurship Monitor (GEM) Project, 96–97 Globalization and Activating Social Assistance, 426 Europe as loser in, 153 for large vs. small states, 153–54, 158 and late–19th–century state, 158–59 Nordic system jeopardized by, 407 pathological overreactions to, 421–24 reactions to in welfare state, 420–21 Goldman Sachs, 143 Google, 215n.9 Gordon, Robert, 13, 197, 199 Gore, W. L., 214 Governance arrangements, in US vs. Europe, 54–55, 99–100 German model, 55 Government funding, for scientific research, 145 Government purchases with no transfers, in payroll–taxes model, 435–37 Government regulation and innovation deficit in Europe, 88 and work hours per capita, 360 Government transfers, in payroll–taxes model, 437–39 Great Britain. See United Kingdom Group–specific initiatives, as business environment area, 103, 130 Happiness research, 9 Hausman test, 476–77, 480n.7

490

Hayek, Friedrich von, 12, 388, 403, 409n.15, 451, 455 Health care. See Medical care in US Heckscher–Ohlin mechanism, 21 Hershey Foods, 214 Hewlett, William, 204 High–growth firms and business environment, 104–105, 107–10 as EU entrepreneurship key, 99, 113–14 generation of (entrepreneurial business environment), 104–105, 107–10 key policy areas for stimulating of, 99–100 measuring generation of, 97–99 Hofstede, Geert, 138 homepricefutures.com. 143–44 Household production, and decline in working hours per capita, 369, 381 Housing, and real GDP in United States, 372 Human capital and comprehensive theory of skills, 292 model for, 293–300 policy impacts of, 300–307 and economic growth, 325, 332 in Europe, 27, 254 insufficient utilization of, 253–54 policy on, 309 and skill use or maintenance, 255–56 formation of in schools vs. families, 313n.19 and taxation, 278 and innovations, 19 investment in and children from disadvantaged families, 291–92 complementarities from, 254, 255, 265, 285, 292, 296, 297, 299, 299–300, 306, 308, 309 and early vs. late stages of life, 254, 288–89, 291, 308–309 educational, 297 equality–incentives tradeoff in, 257–58 implicit taxes on returns of, 308 insufficient motivation for, 307 and low–skilled workers, 265 and productivity growth, 268 and retirement, 306

Index

as stagnant, 262 and taxation, 267–68 and technology of skill formation, 287–92 maintenance of, 308 policy on and early childhood intervention, 309 economic context of important, 254–55 importance of for Europe, 307 reinvention of required, 253, 254 skills contributing to, 256 creation of, 259–69 economic environment of, 256–59 formation of as cumulative, 255 mainland Europe’s problems with, 255–56 need to educate for, 254 utilization rate of, 307–308 and benefit entitlements, 273–75, 276 and decrease in working hours, 271–73 and distortions in labor markets, 309 and labor force participation, 269–71, 272 and labor market regulations, 273, 275, 277–78 and pensions or early retirement schemes, 285–87, 288 and retirement age, 281–85, 286, 308 welfare–state policies as lowering, 255 Humboldt, Alexander von, 137 Hume, David, 447 Hurd, Mark, 164 Ifo Institute, 425 Ignition Partners, 86 Imagine Communications, 86 Immigrants. See also Migration Nordic model stressed by, 407 and real GDP in United States, 377 and social problems, 253, 307 Immigration policies, upstream innovation served by, 209 Implicit taxation. See Taxation, implicit Incentives and entrepreneurship, 101

Index

vs. equality, 257 for training, 266 among low–skilled workers, 265–66 Income per capita of rich and poor (three groups of countries), 401–402 and social spending (three groups of countries), 400–401 Income redistribution, perverse, 309 Income taxes, personal, 280 as business environment area, 103, 128 Index–based derivatives, risk management through, 142–44, 148 India barriers to technology adoption in, 202 family firms in, 162 as globalization winner, 153–54 as market participant, 413 Industrial policy, in large states, 154 Inefficiency, and real GDP in United States, 376–77 Inequality of income, 256–57, 258–59 in three groups of countries, 401–402 Inflation targeting, and countercyclical budgetary policy, 225, 227, 238, 241, 244, 245–46 Information technology (IT) customers’ risks in, 185 and European Union, 170 and US productivity, 195, 197–99 US spending on, 193–95, 197 venturesome users of, 199–201, 204 Initial public offering (IPO), 47–48, 49, 65 Innovation. See also Economic dynamism absorptive capacity for, 171 acquisition and use of (“downstream”), 169, 170–71 elusive underpinnings of, 201–205, 206 by venturesome consumers, 182–88 as argument against welfare state, 403, 405–406 and cultural differences, 136–39 and economic growth, 338 and education or immigration policies, 209 in established firms vs. VC backed start– ups, 87–88

491

in Europe (short supply of), 65 as evolutionary, 180–82 and family firms in India and China, 162 free market system of, 188 and growth (Schumpeter), 178 and destruction, 178–79 and human capital, 19 (see also Human capital) and increased wealth, 21 in international trade, 188–92 managers’ role in diffusion of, 18 multiple factors in success of, 19 multiple players in, 180 private–ownership system structured for, 451 promoting diffusion of, 205 and risk, 31 in solidarist–influenced private– ownership system, 451–52 specialization in, 181–82, 215–16n.10 as system, 171–72 venture capital role in, 65 “Innovation market,” 328 Insecurity, and real GDP in United States, 376–77 Interest politics, 159 and large vs. small states, 159 Interest rate, in relation of payroll tax to employment, 440–42, 443 Intergenerational equity, 152 International governance, and small vs. large states, 158 International Swaps Dealers Association, 87 Internet bubble, and VC risk taking, 69, 72 Invention, and innovation, 180 Investment, and influx of new workers, 13 Investment Property Databank (IPD), 144 iPod, 192 Ireland, 154, 388 Iron Curtain, fall of, 413 Italy, 388. See also European Union, core countries of corporatist system of, 451 family firms in, 162 lack of VC funding in, 75 malaise in, 154–55, 156

492

James, William, 211, 454 Japan car companies in, 192 deficit increase in, 157–58 R&D investment of, 175 reconstruction and export–led boom of, 177 as rival, 172 superiority of alleged, 30 Jetelová, Magdelena, 155 Job involvement, on Continent, 5–6 Job satisfaction. See also Employee engagement on Continent vs. in comparator countries, 4–6 in various countries, 24 Job security, in Europe vs. America, 38–40 Job training programs, 291 Jospin, Lionel, 156 JTPA program, US, 291 Kalman filter, 234, 246–48 Kant, Immanuel, 447 Kelly, Pat, 205 Kindleberger, Charles, 160 Koch, Christopher, 212, 213 Kohl, Helmut, 155 Korea, R&D investment of, 175 Korobov, Vladimir, 138 Labor force, participation rates in, 269 by age, 282 and pensions or early retirement schemes, 285–87, 288 and skill formation, 292 Labor market regulation as business environment area, 103, 129 as distortion–producing, 309 as EU policy challenge, 111, 112, 113–14 and labor market performance, 275, 277–78 and unemployment, 273 and utilization rates of human capital, 253–54 Labor markets demand as skill–based in, 264–65 and entrepreneurship, 37–40 equity–incentive tradeoff in, 256–57 in Europe and 1995 hours–per–capita and productivity turnaround, 370, 380, 382

Index

rigidity of, 419–20 in model of transmission of ideas, 338 1995 hours–per–capita and productivity turnaround in, 344, 345–51, 355, 357, 363 Nordic countries’ employment strategy, 393 unskilled workers in, 253 Labor productivity, in statistical tests of economic culture, 467 Labor supply, and payroll tax, 429 for large open economy, 439–43 for small open economy, 432–39, 442 textbook model of, 430–32 Labor tax. See Payroll tax; Taxation, labor Lagardère, Jean–Luc, and Lagardère dynasty, 165 Laissez–faire tradition, arguments of, 402–406 Lala.com, 86 Landes, David, 11, Lazonick, William, 161 Leadership. See also Managers technological, 172–78 and upstream innovation, 169 Leisure demand for, 296 and level of human capital, 303 social vs. market coordination of, 400 valuation of, 356, 357, 367, 367–70, 379, 380, 401 diminishing marginal value of, 366 Life–cycle profile of labor earnings, 305 Life satisfaction, and productivity level, 9 Lifetime utility, maximizing of, 309–11 LignUp.com, 86 Lisbon agenda, 153, 175 Lisbon Summit (March 2000), 1 Literacy, and educational attainment, 260 Loan labor, 426 Loans, as business environment area, 103, 127 Loebs, Terry, 14 London Futures and Options Exchange, 143 Louis XIV, on making money, 23 Lowenstein, George, 185, 186 Luxembourg holding company (“Luxco”), 78–79

Index

Maastricht Treaty, 240 Macroeconomic policy, and economic growth, 223–24. See also Cyclicality and countercyclicality of budget deficit among OECD nations MacroMarkets LLC, 143, 144 Male participation rate, in statistical tests of economic culture, 466 Managerial jobs, growth in, 209 Managers in diffusion of innovation, 18 question of time spent examining new ideas, 324, 326, 338 Market(s), and state, 152, 160 Market forces, and growth slowdowns, 15 Marketing, in diffusion of innovations, 207–208 Markov chain Monte Carlo (MCMC) methods, 234, 246–48. See also AR(1) MCMC method Marx, Karl, 447 Massachusetts Institute of Technology, and practicality, 137 Masucci, Sam, 143, 144 McClelland, David, 138 McKinsey & Co., 170 Medical care in US private–sector inefficiency in, 394 wastefulness of, 345, 376–77 Medical entitlements, and growth slowdowns, 15 Merkel, Angela, 155 Messier, Jean–Marie, 165 Metropolis–Hastings (MH) sampler, 247–48 Metropolitan dispersion, and real GDP in United States, 372–76 Microcomputers, evolution of, 181 Middelhoff, Thomas, 165 Migration. See also Immigrants and factor price convergence, 415–17 from France and Italy, 22 and unemployment, 421–22 Mincer equation, 297 Minimum–maximum method, in normalization of indicators, 104 Minimum wages, 253, 256–57 Model of skill formation, use and maintenance, 292–300 Mohn family, 165 Monsanto, 214

493

Moore, Gordon, 202 Moral hazard, 254 Moral relativism, 447 Moscatelli, Vittorio, 75 Motivation, and entrepreneurship, 101 Multinational corporation effect of on host countries, 201 faults of, 165 global supply chains of, 191–92 “National family perspective,” 358 National Science Foundation, 146, 147 National Venture Capital Association, 67 Nelson, Richard, 18, 171–72, 176–77 Neoclassical economic theory, 449–50, 450 Chicago school, 20 on “economic system,” 1–2 and evaluation of innovation, 185 and labor supply, 429 and labor taxation, 400 on social welfare systems, 2 supply–side criticisms of Continent’s welfare systems, 12 Neo–neoclassical theory, 449–50 Netherlands, 388, 408n.1. See also European Union, core countries of New Zealand, 154, 388. See also English– speaking economies Nomad Jukebox, 192 Noncognitive abilities, 289, 290 Nondestructive creation, 178–80 Nordic (Scandinavian) countries, 387–88, 388–89, 451. See also Taxation– growth relation in three groups of countries Faggio–Nickell puzzle over, 430, 443 fiscal distress absent in, 404–405 as free societies, 403–404 government outlays in, 389, 391, 392 household saving rate of, 405 income inequality in, 402 as small open economies (payroll tax– hours worked relationship), 443 social welfare policies of, 393 and GDP, 400–401 labor market outcomes of, 396–400 poverty educed by, 394–96 question of solidity of, 406–407 technological excellence of, 405–406 unemployment in, 398, 399 North, Douglass, 448

494

North–South model and destructive creation, 179–80 “make where you innovate” assumption of, 192 and overseas innovation, 188, 189 and technological leadership, 173, 175–76, 178 Norway, 387. See also Nordic countries OECD, budget deficits of. See Cyclicality and countercyclicality of budget deficit among OECD nations OECD range of social spending and taxation in, 387 Offshoring, 417 On–the–job training, 292, 297 Openness, and countercyclicality, 238, 241 Operating systems, sale of (worldwide comparisons), 193, 194–95 Opportunities, as entrepreneurship factor, 101 Outsourcing, 417 Ownership structure, and corporate entrepreneurship, 50–53 Packard, David, 204 Palmisano, Samuel, 165 Participation rate, male, in statistical tests of economic culture, 466 Pasetti, Giuseppe, 75 Patents, objections to filing for, 183–84 Payroll tax and labor supply (hours worked), 429–30 for large open economy, 439–43 for small open economy, 432–39, 442 textbook model of, 430–32 and unemployment, 278 Penn World Table, 477 Pensions and growth slowdowns, 15 and labor force participation of older workers, 285–87, 287 need to correct, 309 Perspectives on Technology (Rosenberg), 180 Peter I the Great (tsar of Russia), 189 Peugeot, Armand, 216n.10 Phelps, Edmund S., 18 Physicians Sales and Services company, 205

Index

Pioneer New Media Technologies, 214 Polanyí, Michael, 451 Political factors, and institutional inefficiency, 3 Political model, 448 Poverty, as reduced by social outlays (three groups of countries), 394–96 Poverty traps, 264, 307 Pre–entry of entrepreneurs, 114 Prestowitz, C., 172, 175, 190, 193 Private demand conditions, as business environment area, 103, 125 Private ownership, individualistic vs. solidarist versions of, 451–52 Private wealth, in statistical tests of economic cultures, 468 Problem solving, resourceful, 187–88 Procurement regulation, as business environment area, 103, 126 Productivity in Anglo–Nordic vs. major continental countries, 453 on Continent vs. in comparator countries, 6–9 and entry of foreign workers, 13 in Europe vs. United States, 198, 343, 348, 448 (see also Standard of living in Europe compared with United States) questions on criteria of, 9–10 as satisfaction of new wants, 179 of service sector, 191 and skills, 268 slowdowns in, 14 US gains in, 192–93, 195, 197–99 Productivity, labor, in statistical tests of economic culture, 467 Productivity growth and education, 330 in model of transmission of ideas, 329–30, 331, 332–33, 338 “Productivity paradox,” 197, 198 “Protestant ethic,” 447 Public good, scientific research as, 145 Public policy, and technology diffusion, 205–10 Public transit, and real GDP in United States vs. Europe, 373–74 Purchasing power parities, 7 Quarterly Census of Employment and Wages, 94

Index

Questionnnaires, attitudinal and situational factors in, 139 Racial antagonisms, and redistribution policies, 406–407 Rational Institutions theory, 2–3 “Rat race” syndrome, 400 R&D, share of as related to cyclicality, 224 Real estate risks, derivative products for, 143–44 Real GDP pr capita, in Europe vs. United States, 343. See also Standard of living in Europe compared with United States) Regionalism, 159 Regulation. See Government regulation Religious orientation, and cross–country differences in high–growth firms, 99 Rent seeking by big European firms, 88 and corrupt state, 152 Replacement incomes, 13, 20–21, 273, 419–20, 427 in Germany, 424 Research on cyclicality and monetary policy, 246 danger of excess in, 171 on effect of economic culture, 471 Resourceful problem solving, 187–88 Restart possibilities as business environment area, 103, 126 as EU policy challenge, 111, 112, 113–14 Retailers, big–box, 197, 204, 374, 375 Retirement decreasing age of, 308 early, 306–307, 309, 351, 361, 367–70 implicit tax on, 306 Retirement ages, and skill depreciation, 281–85, 286, 308 Ricardo, David, 211 “Rise and Fall of American Technological Leadership, The” (Nelson and Wright), 176–77 Risk cross–country differences in attitudes toward, 139–40 in entrepreneurship, 31, 135 limitation of for venture capitalists, 73–74 unmeasurable and unquantifiable, 184–86

495

Risk management, 135 and entrepreneurship, 135, 141–44, 147–48 through index–based derivatives, 142–44, 148 through venture capital, 141–42, 147–48 Risk taking and corporate entrepreneurship, 48, 50 accountability and control in, 54–56 and executive compensation, 56–59 and ownership structure, 50–53 in Europe and America, 33–36 Road to Serfdom, The (von Hayek), 409n.15 Robb, Richard, 470 Roberts, Eric, 211–12 Rosenberg, 180 Russell Sage conference on altruism, 447 Russia, discontent in, 157 Sapir, André, 154 Savings, household, as argument against welfare state, 403, 405 Savings and investment, stimulation of, 208 Scandinavian countries. See Nordic countries Schooling and on–the–job training, 292 in Protestant regions, 25–26n.12 Schröder, Gerhard, 155 Schumpeter, Joseph and “creative destruction,” 178 on “entrepreneurial spirit,” 447 on entrepreneurial spirit of Austrians, 472 and entrepreneurship, 11, 93, 181 and Initiative/Passivity dimension of culture, 455 on innovation, 87, 178, 180, 182, 451 Science, The Endless Frontier (Bush), 146 Scientific research encouragement of, 136, 144–47 and factor price convergence, 415 US decline in, 173–75 Scientism, 23, 452, 455 Seillière, Ernest–Antoine, 165 Selection Principle (Sinn), 3 Self–employment rate, and economic performance, 93

496

“Self–made man,” in US vs. Europe, 37, 112–13 Service sector, productivity of, 191 787 Dreamliner, 88 Signaling theory, on geographic affinity of VCs and entrepreneurs, 84, 85–86 Singapaore, 154 Sinn, Hans–Werner, 3, 155 Skill(s). See also Human capital as entrepreneurship factor, 101 loss of during unemployment, 359 Skill creation, 259–69 and skill–based demand for labor, 264–65 Skill formation, 19 technology of, 287–92 Skill formation, use and maintenance, theory of, 292–300 policy impacts of, 300–307 Skill maintenance, 281–87 Skill utilization, 269–81, 307–308 Small Business Administration (SBA), 94 Smith, Adam, 201, 211, 447, 448 Social expenditures (outlays) and economic health, 387 in three groups of countries, 389–96 Socialism, Hayek’s changed view of, 409n.15 Social model, 20, 448 Social partners, protection of, 23, 452 Social problems, among unskilled persons, 253 Social prosperity, multiple dimensions of, 28 Social security discrimination, as business environment area, 103, 129 Social wealth as private wealth vs. payroll–tax beneficiary, 430 relative importance of, 430 in statistical tests of economic culture, 468 Social welfare systems. See Welfare systems and states Solow, Robert, 195 South Korea, 154 Soviet Union business–oriented attitudes in, 138 as technological also–ran, 202 Specialization and factor price convergence, 417 reality of, 418–19

Index

horizontal, 417, 422–23 unfavorable outcomes of, 422–24 vertical, 417, 423–24 Staged financing, and venture–capital firms, 141 Stakeholders, 23, 452 Standard of living in Europe compared with United States, 343, 377–83 and explanations for declining hours per capita, 351, 355, 356, 363–65 from age distribution of unemployment and labor force participation, 351–55 and evaluation of leisure, 356–57 labor taxes, 357–58, 361–63 unionization and regulation, 360–61, 363, 382 welfare state, 358–60, 381 and mis–measurement of US real GDP, 372 in energy and metropolitan dispersion, 372–76 in housing, 372 from immigration and underground economy, 377 from insecurity and inefficiency, 376–77 and 1995 turnaround in hours–per– capita and productivity, 344, 345–51, 355, 357, 363, 370, 380, 382 and “time effects,” 364, 382–83 and “unmeasured leisure” hypothesis, 343–44, 348 and wasted production in US, 344–45 and welfare implications of decline in hours per capita, 365 cost of higher unemployment, 366–67, 380–81 early retirement and valuation of leisure, 367–70, 380 idle youth, 370 and welfare evaluation of work vs. leisure, 370–72 Standard & Poor’s, 143 Standout countries on Continent, 7 Start–up firms number of (EU and US), 94–97 question of definition of, 94 State, and markets, 152, 160 State ownership in Europe, 51 and risk taking, 50

Index

Statistics Denmark, 92, 96 Stock market, weakness of, 449 Stock market capitalization, and growth (model of transmission of ideas), 336 Stock markets and buyouts, as business environment area, 103, 128 Subsidies for education, 267–68 higher education, 303 wage, 202, 292, 424–27 Supply–side viewpoint, 387 arguments of, 402–406 Sutton, Willie, 191 Swaps, 142 Sweden, 387. See also Nordic countries Switzerland, 156 Syndication methods, and venture– capital firms, 141 Systems competition, market failure in, 3 Taiwan, 154 Taxation and characteristics of social outlays, 393–94 as EU policy challenge, 112, 114 and growth, 387, 388, 407 supply–side arguments on, 387 on human capital formation, 307 and labor market performance, 278–81 in Nordic countries, 407 vs. core European countries, 405, 408 and poverty reduction through social outlays, 394–96, 397 and reduction in working hours, 271 and skill formation, 266–68 and utilization rates of human capital, 253–54 Taxation, business, as business environment area, 103, 128–29 Taxation, capital, as business environment area, 103, 128 Taxation, implicit, 254, 301, 308, 314n.29 cross–country plotting of, 288 and early retirement schemes, 286–87, 298 on retirement, 298, 302, 306 and skill formation, 255, 307 Taxation, income, 280 as business environment area, 103, 128

497

Taxation, labor. See also Payroll tax and decline in European working hours, 349, 356, 356–57, 381 impact of, 301, 303 and OJT, 306 in statistical tests on economic culture, 457, 459, 461, 463, 465 and working hours, 400 Taxation, personal, as business environment area, 103, 128 Taxation, wealth and bequest, as business environment area, 103, 127 Taxation–growth relation in three groups of countries, 387–88, 407 and arguments against welfare state, 402–406 and labor–market outcomes of Nordic social welfare policies, 396–400 and levels of social outlays, 389–93 and per capita income of rich and poor, 401–402 and social spending effect on GDP, 400–401 Tax credits, for capital outlays, 208–209 Tax multiplier effect, 344, 381 Tax rate and innovation deficit in Europe, 88 on payrolls, 20, 21 Tax structures, and European venture capital, 75–76, 77 and France–Luxembourg connection, 78–79, 80–81 and United Kingdom, 76–78 “Techno–fetishism,” 19, 169 Technological knowledge, and factor price convergence, 415 Technological leadership, 172–78 Technological progress, belief in, 202 Technology, of skill formation, 287–92 Technology adoption barriers to, 201–202 and economic development, 171 Technology diffusion, and education, 201–202 Technology transfer as business environment area, 103, 125 “Techno–nationalism,” 169 Tenure in jobs, on Continent vs. comparator countries, 25n.6 Thatcher, Margaret, 151 Theory of Capitalist Development, The (Schumpeter), 181–82

498

Theory of the Leisure Class (Veblen), 203 Theory of skill formation, use and maintenance, 292 model for, 293–300 policy impacts of, 300–307 Thrift belief in, 203–204 and labor force participation rates, 14 Time effects, in changes of work hours per capita, 364, 382–83 Tocqueville, Alexis de 447, 470, 473 Tolentino, Cristina, 75 Total Entrepreneurial Activity (TEA), 96–97 Total factor productivity, 6. See also Productivity Total factor productivity growth, 338–40 Training for low–skilled workers, weak incentives for, 265 Transmission of entrepreneurial ideas between and within countries, 323, 338 model of, 323–24 and adoption of foreign ideas, 326–27 empirical evidence on, 331–38 entrepreneurship in, 327–29 and local manager’s role, 324–26 and productivity growth, 329–30 Tull, Bob, 144 Twain, Mark, and death of social welfare state, 408 Uncertainty and entrepreneurship, 184 and family firms, 161, 162 on institutional innovations, 3 Unemployment and benefit entitlements, 273 comparative data on, 4 and countercyclical policies, 225 and education level, 269–70 and entrepreneurship, 101 in Europe, 275, 382 among low–income workers, 257 welfare state of, 359–60 from increases in working–age population, 13 and job tenure, 39 and labor market regulations, 273 and migration, 421–22

Index

real vs. official rates of, 270 and replacement incomes, 13, 20–21, 420, 427 and social model, 20 from specialization, 423 state–financed, 418–19 in statistical tests on economic culture, 463, 465 in three groups of countries, 398 and wage rigidity, 7–9 and wealth/wages ratio, 13–14 welfare cost of increase in, 366–67 Unions and unionization and reduction in working hours, 271, 356, 360, 363, 382 and rise in tax rates, 278 and wage rigidity, 419 United Kingdom, 388, See also English– speaking economies and countercyclicality of budget deficit, 227, 236, 237, 238, 246 and family firms, 162, 162–63 IT employees in, 200 and scientific research, 175 United States, 388. See also English– speaking economies attitudes toward entrepreneurship in, 112–13 business competition in, 34–35 concerns over entrepreneurial culture of, 136 and countercyclicality of budget deficit, 227, 235, 236, 237, 238 deficits in, 157–158 despondent period in, 27–28 vs. Europe in access to capital, 43–46 in economic growth, 28–31 in entrepreneurship, 31, 33, 91, 92, 323 in environmental business environment, 105–106 in executive compensation 57–59 in governance arrangements, 54–55, 99–100 in labor markets, 37–40 in ownership structure, 51–53 in productivity gains, 198, 343, 348, 448 in returns to education, 264 in risk taking, 33–36

Index

in standard of living, 343 (see also Standard of living in Europe compared with United States) in venture capital , 65 and European postwar performance, 10–11 and family firms, 162, 163 as incomes leader, 192–93 and IT consumption patterns, 193–95, 196 and productivity gains, 195, 197–99 inequality in, 256–57 IT spending of, 193, 197 mis–measurement of real GDP in, 372–77 number of high–growth firms generated in, 97–99, 100 number of start–ups in, 94–97 poverty rate in, 395 practicality admired in, 137 “productivity slowdown,” 172 scientific research in, 146, 146–47 self–employment preference stronger in, 16–17 social expenditures in, 394 and technological leadership, 172–75, 190 venture capital finance in, 66, 67–74, 88, 141, 142 and age of investee companies, 68 and early exit, 72–73 and Internet bubble, 69, 72 risk limitation for, 73–74 “Unmeasured leisure” hypothesis, 343–44, 348 Unskilled workers, 253 and welfare–state dependency, 253 Utility function, 432, 433 Values, in workplace, 22 Veblen, Thorstein, 203 Venture capital firms (VCs), 17, 18, 42, 60n.9, 65–66 as business environment area, 103, 128 vs. established firms as promoting innovation, 87–88 in Europe, 66, 74–75, 88, 142 as EU policy challenge, 111, 112, 113–14 and European Commission, 65–66, 86, 142

499

and France–Luxembourg connection, 78–79, 80–81 and tax structure, 75–76, 77 and United Kingdom, 76–78 exit option for, 47 geographic matching of entrepreneurs with, 79, 81–86 governance structures of (US vs. Europe), 44 and R&D, 215 risk management through, 141–42, 147–48 sectoral distribution of investments by (US vs. Europe), 45, 46 sources of funds for (US vs. Europe), 43 in United States, 66, 67–74, 88, 141, 142 and age of investee companies, 68 and early exit, 72–73 and Internet bubble, 69, 72 risk limitation for, 73–74 Venturesome consumption, 170, 171, 182–88, 192–93, 202–205, 208 of IT users, 199–201, 204 Vesting programs, and venture–capital firms, 141 Virgil, quoted, 1 Volkswagen, 214 Von Hippel, Eric, 183 Wage cost, for West vs. ex–communist countries, 414 Wage flexibility, under wage–subsidies system, 426 Wage rigidity and unemployment, 7–9 and unions, 419–20 Wages. See also Compensation of labor on Continent, 9–10 minimum, 253, 256–57 Wage subsidies, 202, 292, 424–27 Wal–Mart, 197, 204, 208, 375 Wealth, private, in statistical tests of economic culture, 468 Wealth, social. See Social wealth Wealth adjustment, and relation of payroll taxes to labor supply, 430 Wealth/wages ratio, 13–14, 21, 430, 440, 443 Weber, Max, 11, 25n.12, 203, 208, 211, 447, 472

500

Weiss, Allan, 143, 144 Welfare systems and states, 307 arguments against, 402–406 Bismarck’s invention of, 158–59 and calculation of real per–capita income, 345 dependency in, 253 as disincentive, 307 and economic performance, 1, 2, 12, 20–22 and comparative evidence, 10 and ethnic homogeneity, 406 and globalization, 419–21 and hours per capita in Europe, 358–60, 362, 381 and human capital investments (implicit tax burdens on), 254–55 improvement of (through wage subsidies), 424–27 replacement incomes in, 13 and unemployment, 8 Whirlpool, 214 Wiener, Martin, 151 Women in workforce. See Female labor force participation rates Worker–owned firms, as risk averse, 50 Work ethic, erosion of, 275 Work hours per capita (Europe) in comparison of three groups of countries, 398, 400 decrease in (explanations), 351, 355, 356, 363–65 from age distribution and labor force distribution, 351–55 and evaluation of leisure, 356–57 from labor taxes, 357–58, 361–63, 381, 429–30 from unionization and regulation, 360–61, 363, 382 from welfare state, 358–60, 381 and tax–rate level, 344 welfare implications of decline in, 365 cost of higher unemployment, 366–67, 380–81 early retirement and valuation of leisure, 367–70, 380 idle youth, 370 and welfare evaluation of work vs. leisure, 370–72

Index

World Values Surveys (WVS), 475 Wright, Gavin, 176–77 Zegna, Ermenigildo, 163–64 Zhirinovsky, Vladimir, 157