Variations in Economic Analysis
MARTINDALE CENTER FOR THE STUDY OF PRIVATE ENTERPRISE CENTER STAFF J. Richard Aronson Director Robert J. Thornton Associate Director Judith A. McDonald Associate Director Todd A. Watkins Associate Director Robert Kuchta Assistant Director of Marketing Paul Brown Dean, College of Business & Economics The Martindale Center for the Study of Private Enterprise was established in 1980 with a gift from alumnus Harry Martindale and his wife Elizabeth Fairchild Martindale. Formed as an interdisciplinary resource in the College of Business and Economics, the Center contributes through scholarship to improved understanding of the American economic system.
J. Richard Aronson · Harriet L. Parmet · Robert J. Thornton Editors
Variations in Economic Analysis Essays in Honor of Eli Schwartz
13 Martindale Center for the Study of Private Enterprise College of Business and Economics Lehigh University
Editors J. Richard Aronson Department of Economics Lehigh University 621 Taylor Street Bethlehem PA 18015 USA
[email protected]
Harriet L. Parmet Department of Modern Languages and Literature Lehigh University 9 W. Packer Avenue Bethlehem PA 18015 USA
[email protected]
Robert J. Thornton Department of Economics Lehigh University 621 Taylor Street Bethlehem PA 18015 USA
[email protected]
ISBN 978-1-4419-1181-0 e-ISBN 978-1-4419-1182-7 DOI 10.1007/978-1-4419-1182-7 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2009936724 © Martindale Center for the Study of Private Enterprise, Lehigh University, 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
My Favorite Two Corporate Finance Puzzles . . . . . . . . . . . . . . . Harold Bierman
5
The Solvency of Federal Welfare Entitlement Programs: Social Security and Medicare . . . . . . . . . . . . . . . . . . . . . . . . . . . . George H. Borts
13
The Corporate Sector as a Net Exporter of Funds: Additional Evidence John B. Guerard, Jr.
31
Piscal Folicy 101—Economic Policy Meets Partisan Politics . . . . . . . John Hilley
51
A Taxonomy of Utility Functions . . . . . . . . . . . . . . . . . . . . . . Benjamin J. Gillen and Harry M. Markowitz
61
Equilibrium and Disequilibrium Growth: a Comment on a Comment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert Solow
71
Reputational Risk and Conflicts of Interest in Banking and Finance: The Evidence So Far . . . . . . . . . . . . . . Ingo Walter
75
Seeking Common Ground on Globalization . . . . . . . . . . . . . . . . Murray Weidenbaum
99
Tax Reform Then and Now . . . . . . . . . . . . . . . . . . . . . . . . . J. Richard Aronson
109
Joseph A. Schumpeter: Not Guilty of Plagiarism but of “Infelicities of Attribution” . . . . . . . . . . . . . . . . . . . . . Nicholas W. Balabkins Economics and the Tanakh—the Hebrew Bible . . . . . . . . . . . . . . Harriet L. Parmet
115 127
v
vi
Contents
Muted Signals in Academe: Letters of Recommendation and Grade Inflation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert J. Thornton
139
A Biography of Eli Schwartz . . . . . . . . . . . . . . . . . . . . . . . .
153
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
157
Introduction
It has been a pleasure to put together this festschrift in honor of Professor Eli Schwartz. The word festschrift is taken from two German words whose meanings are, respectively, “celebration (or feast)” and “writing”; and this festschrift does precisely that. It celebrates, through a collection of invited essays, the contributions that Eli has made over the years to his profession, to his colleagues and friends, and to Lehigh University. The volume title, Variations in Economic Analysis, is certainly appropriate. The 12 essays contained within span a wide range of subject matter—tax reform, corporation finance, fiscal policy, banking, economic growth, globalization, to name just a few. Their common thread is that Eli has taught or has written about virtually all of them in the span of his long academic career. The methodology used in the essays is equally varied—from economic theory, to econometrics, to case analysis, to armchair reasoning, to . . .yes. . .even humor and farce. Again nearly all of these approaches have been used by Eli in his writing and teaching. We present brief summaries of each of the essays below, largely in the words of the authors themselves. In “My Favorite Two Corporate Finance Puzzles,” Harold Bierman notes that a “finance puzzle” occurs when finance theory suggests that one course of action is preferred and managerial practice takes a different path. Of course, any finance decision will have its pluses and its minuses, and it can therefore never be argued that there are no possible alternatives to the optimum choice that is determined by theory. But we can still suggest that a preference does exist. Thus with distribution policy, a share repurchase policy beats a cash dividend given the tax laws that have been in existence for the past 100 years. Similarly, a corporation should use a reasonable amount of debt. If the investor in the firm’s stock finds the risk to be excessive, then that investor can shift to a mix of the firm’s debt and stock. In “The Solvency of Federal Welfare Entitlement Programs: Social Security and Medicare,” George Borts uses the concepts underlying generational accounting to identify the future federal tax burdens that will be created by the fulfillment of present obligations under Social Security and Medicare. He explores methods of measuring and closing the gap between projected federal welfare outlays and revenues, which under current conditions is expected to reach 8.5% of GDP over the
J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_1, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
1
2
Introduction
foreseeable future. This ratio is equivalent to 50% of projected federal revenues. He also discusses the revenue requirements and structural modifications that would be needed to control the “fiscal imbalance” implied by current proposals to extend the federal government’s role in the financing of health care to the entire population. John Guerard’s essay (“The Corporate Sector as a Net Exporter of Funds: Additional Evidence”) notes that in their 1966 article Eli Schwartz and Richard Aronson documented the role of the corporate sector in generating more funds than it can profitability use. In that work (published in the Southern Economic Journal) they pointed out that aggregate dividends far exceeded net new external financing. Guerard presents further evidence for the 1971–2006 period for US stocks with regard to debt and equity issuances, repurchases, and the relationship of dividends. His evidence substantiates the original Schwartz–Aronson hypothesis of the corporate sector as a net exporter of funds, while it also shows how net corporate exports have evolved. The next essay is on the somewhat lighter side in its approach, which emulates a Washington satirical ensemble called the “Capitol Steps.” In “Piscal Folicy 101—Economic Theory Meets Partisan Politics,” John Hilley draws lessons from three episodes of economic policy-making played out in the crucible that is political Washington. The first episode illustrates unintended consequences: the Bush attempt in 1990 to lower the capital gains rate propelled the largest deficit reduction legislation in history. Second, the failure of the 1993 Clinton stimulus bill counsels that sometimes it’s better to leave the textbooks on the shelf. Finally, the 1997 balanced budget agreement offers hope that our public servants can “get it right.” The essay closes by offering suggestions to address our current national problems as fiscal realists. The next essay is by Harry Markowitz (with Benjamin Gillen) and is titled “A Taxonomy of Utility of Functions.” In a classic 1952 article (“The Utility of Wealth,” in the Journal of Political Economy), Markowitz hypothesizes a utility function—bounded above and below—with three inflection points. The boundedness assumptions plus the assumptions of monotonicity and a continuous second derivative by themselves imply that the utility function has an odd number of inflection points. The number of inflection points and their placement relative to current wealth determine an economic agent’s actions in the face of risk. This essay presents a classification of such utility functions and notation to conveniently note the various classes. Robert Solow’s contribution, “Equilibrium and Disequilibrium Growth: A Comment on a Comment,” takes the form of an imaginary conversation between him and Eli on Chapter 10 of Eli’s 1993 book Theory and Application of the Interest Rate. In the chapter Eli discusses, in his own theory-plus-common-senseplus-figures way, the relations among the aggregate growth rate, the interest rate, and the profit rate. Solow’s essay is a brief commentary on that chapter. The main focus is on the tendency for nonfinancial corporate profits to exceed interest payments, and the sources of this tendency. At the microeconomic level, quasi-rents (or Schumpeterian profits) can be regarded as a disequilibrium phenomenon, but
Introduction
3
in a progressive economy the continuing existence of a float of quasi-rents is a characteristic of equilibrium. The next essay is Ingo Walter’s “Reputational Risk and Conflicts of Interest in Banking and Finance: The Evidence So Far.” In recent years the role of various types of financial intermediaries has evolved dramatically, as capital market deregulation and innovation have resulted in intensified competition. Consequently, market developments have periodically overtaken regulatory capabilities intended to promote financial stability and fairness as well as efficiency and innovation. It is unsurprising that these conditions would give rise to significant reputational risk exposure for banks and other financial firms involved. However, dealing with reputational risk and controlling exploitation of conflicts of interest can be expensive. On the other hand, reputation losses associated with conflict of interest exploitation can cause even more serious damage. Murray Weidenbaum’s essay, “Seeking Common Ground on Globalization,” attempts to identify some common ground on where the various parties to the continuing controversy on globalization might agree. He presents a number of suggestions for undramatic but hopefully useful policy changes: (1) reforming the World Trade Organization by opening the proceedings to the public, (2) helping the people hurt by globalization to regain productive positions in the workforce, (3) strengthening the International Labor Organization, and (4) giving the public a formal voice on global issues via the Internet. In “Tax Reform Then and Now,” Richard Aronson argues that the separate tax treatment of corporate and personal income must be changed. Taxing all shareholders at a common rate creates inequities, and taxing dividend income at both the corporate and personal level causes inefficiencies. Aronson reviews the various techniques of integrating the two taxes and concludes that the withholding method is best. In the next essay (“Joseph A. Schumpeter: Not Guilty of Plagiarism but of ‘Infelicities of Attribution’ ”) Nicholas Balabkins notes that Schumpeter has been the “darling” of the economics profession for more than a century. Terms coined by Schumpeter such as “creative destruction” and “innovation” are standard terms even today in economics textbooks. Economists have always assumed that the idea of the successful innovator popped from Schumpeter’s head, like Athena from the head of Zeus. And throughout his life, Schumpeter created the impression that he, and no one else, was the progenitor of the innovator concept. However, in 1961 Professor Knut Borchardt, at the University of Mannheim, Germany, resurrected Albert Schäffle’s volume of 1867 and demonstrated stunning parallels between Schäffle’s and Schumpeter’s types of innovations. This was despite the fact that Schumpeter had repeatedly written that Schäffle had left not a single permanent result for posterity to the economics profession. Harriet L. Parmet’s essay, “Economics and the Tanakh—the Hebrew Bible,” presents various economic scenarios and concepts found in this ancient text. Included in her discussion are social policies, the Sabbath, the jubilee years, the agrarian society, and the ethics of social justice viewed from a sociological as well as from a theological perspective. Further topics related to economics that she treats
4
Introduction
are land and money in ancient times, the vulnerability of wage earners in biblical societies, slavery in Israel, and redistribution as political economy. The Torah or Humash (also known as the five books of Moses) is referenced as is the larger corpus of biblical text, the Tanakh. In the final essay of the festschrift (“Muted Signals in Academe”), Robert Thornton discusses the twin problems of grade inflation and less-than-frank letters of recommendation in today’s academic world. Because both grades and recommendation letters have lost a good deal of their ability to distinguish good performers from bad performers, the results are higher costs to employers in seeking out reliable information about the students they hire. Students also face costly consequences with respect to effort, incentives, and the ability to distinguish themselves from others in their job search.
My Favorite Two Corporate Finance Puzzles Harold Bierman
My favorite two corporate finance puzzles are: 1. The dividend puzzle. 2. The capital structure puzzle. Long ago, Fischer Black (1976) wrote the classic paper, “The Dividend Puzzle.” But even though the paper clearly defined the puzzle, it is appropriate that more attention be paid now to this interesting issue. It is not sufficient that we conclude as Fischer Black did with “We don’t know.”
The Dividend Puzzle When Black wrote his paper, there was a wide gap between the tax rate on ordinary (dividend) income and capital gains. The puzzle was why corporations forced their investors to pay high ordinary tax rates when a lower capital gains tax rate was available. Example: A firm has 1,000 shares outstanding. Assume an investor owns 100 shares of a common stock. The stock price is $100. The investor’s tax basis is $98 per share. The corporation can pay a cash dividend of $2 per share or can do a share repurchase of $2,000 and the investor who does not sell will have a $200 unrealized capital gain (with the 100 shares). Assume the investor pays a tax rate of 60% on ordinary income and a 25% tax rate on realized capital gains. With a dividend, the investor receives $200 and nets after tax: 200 (1 − 0.6) = $80 and owns 100 shares (10% of the outstanding shares) worth $9,800 or total assets of $9,880.
J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_2, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
5
6
H. Bierman
With share repurchase, assume the investor sells two shares for $200 in total ($100 per share) and has a $4 capital gain. The tax is $1: 4 (0.25) = $1 The investor nets $199 and still owns 98 shares worth $9,800 and total assets of $9,999. The investor who does not sell has 100 shares worth $10,000. With share repurchase, the firm buys $2,000 of stock or 20 shares. There are then 980 shares outstanding. Assume the investor sells two shares, thus has 98 shares left or 10% of the firm’s outstanding shares (same as with the dividend). But with share repurchase the investor has $199 of cash whereas with the dividend the investor only has $80. With the above facts, it is obvious that share repurchase is more beneficial for the investor who sells than a cash dividend. Let us change the tax basis for the 200 shares. If the investor’s tax basis was $100 per share, the investor selling two shares would have netted $200. If the investor’s tax basis was zero, the investor would have a $200 capital gain and netted $150 cash after tax: 200 (1 − 0.25) = $150 Today (2007) the tax rates on eligible dividends and capital gains are both 0.15; but thanks to a tax basis greater than zero, or the investor choosing the alternative not to sell, share repurchase is at least as good as dividends for the investor and is likely to be preferred if the investor’s tax basis is greater than zero. Thus to benefit taxed investors, corporations in general should do share repurchase rather than cash dividends. But the more significant puzzle has to do with why managers and the board of directors insist on paying ever-increasing dividends rather than do more and larger share repurchases. We will assume the members of the firm’s board of directors own a material amount of stock and pay income taxes. We should note that there should be a set of corporations with little or no growth opportunities that should (strategically) pay dividends. These corporations will tend to appeal to investors with low tax rates who want a relatively secure set of cash flows. In this paper, we consider investors paying a high rate of taxes who are not indifferent as to whether or not the corporation pays them dividends or pays them in some other format. We now consider the distribution decision from the viewpoint of management (including the board of directors). For simplicity, we will consider a firm that is not growing; and, consistent with the previous example, the firm has $2,000 that it can use to pay a dividend ($2 per share) or to repurchase shares. Each year, the firm earns $2,000. If the firm pays a $2,000 dividend, the total stock value before the dividend is $100,000 at the end of the year, and the price per share is $100 before the dividend. After the cash dividend, neglecting the tax factor, the firm value is $98,000 in total
My Favorite Two Corporate Finance Puzzles
7
and $98 per share. These facts are valid for the end of each year (the firm size is constant each year at the same time of the year). Now consider what happens with a $2,000 share repurchase. The firm buys 20 shares at a price of $100 per share. The total firm value after the $2,000 repurchase is $98,000 and the value per share is: 98,000 = $100 980 If the firm earns $2,000 in the next year, the value at the end of the second year is again $100,000 and the value per share is: 100,000 = $102.04 980 With $2,000 of available cash at the end of the year, the firm can buy 2,000 = $19.60 shares 102.04 and after the repurchase of $2,000 of common stock (19.6 shares) the value per share is again $102.04: 98,000 98,000 = = $102.04 shares 980 − 19.60 960.40 The firm is buying back 2% of its shares: p=
19.60 = 0.02 or 2% 980
The growth rate in stock price for this zero growth firm is g=
p 0.02 = = 0.0204 1−p 0.98
Initially, the stock price is $100. After 1 year, it is P1 = 100 (1.0204) = $102.04 For verification, see the above calculations. After 10 years, the stock price will be $122.38: P10 = 100 (1.0204)10 = $122.38 If the firm had been more profitable, the growth rate would be larger (remember that this is a zero real growth firm). For example, if the firm could buy back 0.10 of its stock each year
8
H. Bierman
g=
0.10 p = = 0.111 1−p 1 − 0.10
and now at time 10: P10 = 100 (1.111)10 = $286.51 By doing a share repurchase instead of a dividend, the stock price goes up by 2.865 times over a 10-year period if the firm can buy 0.10 of its shares each year.
Other Studies of the Puzzle In 1990 Fischer Black returned to the dividend issue and in half a page summarized the issues “Why do firms pay dividends? I think investors simply like dividends.” He concludes with “I think dividends that remain taxable will gradually vanish.” They are being paid but decreasing in importance. But the puzzle remains, if not actually stated. As P.L. Bernstein (1996) first notes (p. 16), “Today with the yield on the S&P 500 down below 2.50% investors are more puzzled than ever.” Yields were about 4% when Black wrote his first piece. Bernstein then concludes (p. 16), “My analysis suggests that dividend yields have no particular significance as a stock market forecasting device.” He notes that the total returns of stock have been good (he includes share repurchases in his analysis (p. 17)). His final conclusion (p. 21) is “The harder we look at the dividend picture, the more it seems like a puzzle, with pieces that just don’t fit together.” This is perfectly consistent with Black (1976), who used the exact same words. Frankfurter (1999) offers an excellent summary of academic thinking regarding dividends. He summarizes the literature into four major groups of papers. However, he fails to find here the answers to the puzzle. Like Black he concludes (p. 80) “Investors love dividends.” He then asks, “Is the Puzzle Solvable?” His conclusion is that (p. 83) “it is either not possible, or extremely difficult, to find an economically rational solution to the dividend puzzle.” At least I think it is his conclusion.
My Conclusion To obtain a different perspective on the dividend puzzle, I will redefine the issue. Remember that a share repurchase program is basically a dividend policy with tax and other advantages for investors. The advantages of share repurchase to management that owns stock options is obvious. Rather than a flat stock price through time, management can manufacture a significant growth rate of earnings per share with share repurchase that can lead to a much larger stock price. The puzzle is “Why would management support a dividend policy that leads to a constant stock price with cash dividends rather than a share repurchase policy that
My Favorite Two Corporate Finance Puzzles
9
leads to an increasing stock price?” If management owns stock options, the dividend policy (an increasing cash dividend) is costly to management. Management is casting aside self-interest in order to benefit the stockholders. But unfortunately, this policy that does not benefit management also does not benefit the taxed stockholders. It is a puzzle why so many corporations pay dividends. In 1976 Fischer Black wrote (p. 8), “What should the corporation do about dividend policy? We don’t know.” In 1990 he wrote (p. 5) “I think dividends that remain taxable will gradually vanish.” This 1990 quote implies that by 1990 Black thought that corporations either should not pay taxable dividends or thought that they should not pay dividends.
The Capital Structure Puzzle In the 1980s, corporate raiders made money-acquiring firms with little debt, leveraging them, and reaping the benefits from the debt tax shields. When the corporate tax rates were reduced from 0.46 to 0.35, the benefits of this acquisition strategy were also reduced. Higher interest rates and higher price multiples, as well as changes in the investment banking industry, reduced merger activity by the end of the 1980s. With the new millennium came the surge in private equity deals. More and more studies showed it was difficult to beat the stock index funds by picking the winning stocks (the stocks that did better than the market). Private equity firms posted ROI records that frequently beat the returns earned by buying the overall market. We will focus here on two of the primary ways that the private equity firms were able to record a history of earning superior returns. Consider a 1-year $1,000 market investment that yields 0.15. 0 − 1,000
1 + 1,150
The investors earn 0.15 investing in the market. Now assume that funds can be borrowed at 0.05 and that $900 of debt is raised: 0 − 1,000 + 900 − 100
1 + 1,150 − 945 + 205
Investment in market Debt Equity investment
The 15% return on equity with zero debt is increased to 105% by the use of debt. If only $800 of debt is used, we have: 0 − 1,000 + 800 − 200
1 + 1,150 − 840 + 310
The return on equity is reduced from 105 to 55%. Of course, if less than $1,150 is earned at time one, the ROE will be reduced to less than 55%. The use of
10
H. Bierman
debt increases the risk to the equity investors given that the 0.15 basic return on investment is uncertain. We can generalize. If the investment can earn a higher expected return than the cost of debt, the expected ROE can be increased by the use of debt. The more debt, the higher the expected return on equity. If the investor’s focus is on the expected ROE with no (or little) attention on the amount of leverage being used, a comparison of the 55% versus the 15% earned in the market demonstrates the apparent superiority of the private equity investment (the comparison is not correct since it ignores the different amounts of leverage and the risk). Continuing the example that uses $800 of 5% debt, now assume that the $1,150 benefits of year one is an after-tax return and that debt interest is tax deductible. The corporate tax rate is 0.35, and thus the after-tax cost of the debt is 0.05 (1–0.35) = 0.0325 and we have for the investment flows: 1 0 Investment in market Debt − 1,000 + 1,150 + 800 − 826 Equity investment − 200 + 324 The ROE is increased from 15 to 62% because of the tax advantage of using debt rather than equity as well as the leverage effect of using 0.05 debt to finance an investment yielding 15%. We know that the prime strategy of a private equity firm is to substitute debt for the use of equity of the firm being acquired. A simple calculation of the value added is tB where t is the corporate tax rate and B is the amount of debt substituted for equity. While tB somewhat overstates the value added, it is a reasonable rough estimate of the value added. In addition, if the use of leverage inflates the ROE earned by the equity investor (who fails to consider that debt is being used and thus risk to the equity investment is increased), value is increased for the private equity firm that is rewarded based on both profitability of the equity and the amount of equity. The private equity firm must find public corporations that are using less debt than the firms could support. There are many such corporations. Thus, this is the first capital structure puzzle. Why do so many corporations use less debt than is feasible if debt is less costly a capital source than equity? One answer is that the CEO and CFO want to control the amount of risk of the investor. But the investor can control risk by buying a mix of the firm’s debt and equity. The CEO is likely not to be aware of this, but it is basic finance. A second answer is that more volatility of equity returns puts the CEO’s job at risk, and the CEO likes his/her job. It is up to the board of directors to solve this obstacle. There is a third answer. The CEO wants the firm to be raided by private equity. The private equity firm is likely to explain how going private can result in the CEO becoming very rich. Why should the CEO try to avoid this welcome fate? Thus we can explain (to some extent) why corporations use less debt than theory suggests is desirable. But it is easy to suspect that in many cases none of the above
My Favorite Two Corporate Finance Puzzles
11
three explanations apply; and we are left with corporations using less than optimum debt and a vibrant private equity industry willing to exploit the situation by taking firms private. Strebulaev (2007) finds that (p. 1747) “firms seem to use debt financing too conservatively, and the leverage of stable, profitable firms appears particularly low.” But his main concern is how firms move to their target level of leverage, rather than explaining why that target is too low.
Conclusion This paper has defined two corporate finance puzzles where corporate practice is not completely consistent with what one might expect a maximizing practice would define. The two puzzles explored in this paper do not exhaust the corporate finance practices that defy good explanations. For example: a. Why is leasing so extensive? b. Why do firms insist on hedging when hedging is not feasible (see the SkillingLay trial)? c. Why do firms do many things to maintain their stock price when it would be more beneficial to all for the stock price to drop? If the price drops excessively, the firm can buy shares. d. Why is ROI still used to evaluate investments by so many firms? e. Why are income bonds not more widely used? Despite the length of the above list, the capital structure and dividend puzzles remain at or near the top of the list. Obviously, those of us who teach corporate finance also have a puzzle. Why have we been so ineffective for so many years?
References Bernstein, P.L. (1996). “Dividends: The Puzzle,” Journal of Applied Corporate Finance, 9(1), 16–22. Black, F. (1976). “The Dividend Puzzle,” Journal of Portfolio Management, Winter, 2(2), 5–8. Black, F. (1990). “Why Firms Pay Dividends,” Financial Analysts Journal, 46(3), 5. Frankfurter, G.M. (1999). “What Is the Puzzle in “The Dividend Puzzle?”” The Journal of Investing, 8, 76–85. Strebulaev, I.A. (2007). “Do Tests of Capital Structure Theory Mean What They Say?” The Journal of Finance, 62(4), 1747–87.
The Solvency of Federal Welfare Entitlement Programs: Social Security and Medicare George H. Borts
Introduction When President Clinton signed the Welfare Reform Bill in 1996, he described it as the “end of welfare as we know it.” It was a premature declaration of victory, for welfare is alive and well in the American economy. It is growing at a rate that will soon ignite public debate over the cost of federal entitlement programs. But the major beneficiaries are not the poor. They are the cohort of the U.S. population over the age of 62 that will swell the number of welfare recipients in 2008 when the first Baby Boomers become eligible to collect Social Security. The burden will jump again in 2011 when they become eligible for Medicare. The 50 million Americans who will be 62 years or older in 2010 will grow in number to 83 million by 2030 and to 100 million by 2050. Their relative importance in the population will rise from 16% in 2010 to 24% in 2050. Extended life expectancy assures that the bulge will last for many decades and will exacerbate the drain on the financial reserves of the Social Security and Medicare trust funds. The trustees of the federal entitlement programs have already projected the time path of deficits, and experts on the federal budget have calculated the revenue that the federal government will need to meet its future Social Security and Medicare obligations. Federal government forecasts indicate that the proportion of GDP going to the two programs will grow from 7.6% in 2007 to 12.7% in 2030 and 17% in 2080. (See CEA Report, p. 88.) Growth in the number of retirees relative to the working population is not the only reason for the expected increase in future welfare outlays. Social Security payments are indexed to wage levels, and Medicare outlays are driven by the price of health care. Federal expenditure on health care is rising more rapidly than the rate of growth of GDP with no sign of deceleration. Early retirement has deprived the programs of tax contributions (Social Security taxes and Medicare taxes), shortening the grace period before their trust funds have spent their accumulated portfolios of government securities. Federal entitlement programs have continued to expand despite warnings of inadequate fiscal capacity to pay for promised future benefits. There is constant political pressure to increase their scope and magnitude. J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_3, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
13
14
G.H. Borts
Examples are the growth of the unfunded liabilities of the Pension Benefit Guaranty Corporation and the never ceasing political pressure on the federal government to relieve business of its exposure to costs of insuring the health care of its employees. Medicare Part D, the prescription drug benefit for the elderly enacted in 2003, could increase the present value of federal debt obligations by some $16 trillion.1 Another entitlement program likely to be expanded is the SCHIP (State Children’s Health Insurance Program) legislation recently passed by Congress. It would enlarge the health benefit eligibility of children in families with incomes far in excess of the Medicaid eligibility ceiling. Despite popular pressure to increase entitlements, whipped up by the competitive bidding that characterizes election periods, there is no shortage of proposals to rein in spending on federal welfare entitlements. There is, however, a short attention span and in some quarters a refusal to take grim forecasts seriously. The first four sections of this paper will explore methods of measuring and closing the gap between projected federal outlays and federal revenues. It is now expected to reach 8.5% of GDP for the foreseeable future, equivalent to 50% of projected federal revenues. The fifth section will investigate current proposals to expand federal welfare commitments in the face of the “fiscal imbalance” already created by existing commitments. The paper is divided into the following sections: 1. 2. 3. 4. 5.
Generational Accounting, Fiscal Imbalance, and Generational Imbalance Measuring America’s Fiscal Imbalance The Need to Achieve Fiscal Balance Making Welfare Financially Viable A Utopian Perspective on Welfare
Generational Accounting, Fiscal Imbalance, and Generational Imbalance Awareness of the severity of the future fiscal emergency can be identified with the writings of three economists: Jagadeesh Gokhale, Kent Smetters and Laurence Kotlikoff. The conceptual and statistical work by Gokhale and Smetters was published by the Federal Reserve Bank of Cleveland in 2003. They prepared the following synopsis of its methodology and conclusions: This paper describes the deficiencies of the measures used to calculate the federal budget, make revenue and spending projections, and assess the sustainability of current fiscal policies. The nature of the deficiencies hides the tremendous impact that Social Security and Medicare commitments will have on the budget in the future, given the way the programs are structured currently and the momentous demographic shift underway as the baby boom generation approaches retirement age. This paper proposes two new simple measures that will enable government officials and the public to calculate more accurately the costs of maintaining these programs into the relevant future. The measures provide a better understanding of the costs involved, when they will be incurred, and by whom. The measures also
The Solvency of Federal Welfare Entitlement Programs
15
provide a way to meaningfully compare the various solutions that have been proposed for dealing with the impending fiscal crisis that will be caused by Social Security and Medicare. (Gokhale and Smetters, cover page abstract)
Generational Accounting and Fiscal Imbalance The methodology which the authors employ is called Generational Accounting. It may be demonstrated through the use of a “fundamental definition” of fiscal imbalance, where the term FIt measures the Fiscal Imbalance as of the year t. FIt = PVEt − PVRt − At
(1)
The government’s total fiscal policy may be considered balanced if today’s publicly held debt plus the present value of projected non-interest spending is equal to the present value of projected government receipts. The spending and revenue projections are made under today’s fiscal policies. “Present values” mean that the dollars paid or received throughout the future are discounted at the government’s long-term interest rate in order to reflect their true value today. A fiscal policy that is balanced can be sustained without changing either federal outlays or federal revenues. (Gokhale and Smetters, p. 4)
Consistent with the authors’ definition of fiscal imbalance, the term PVEt is the present discounted value of all future government non-interest expenditures, PVRt is the present discounted value of all future government receipts, and the term At is the value of outstanding government assets net of outstanding government debt. Gokhale and Smetters point out that the fundamental definition of fiscal imbalance (FI) may be applied to the entire federal budget or to specific programs. In the former case At represents the cash that could be realized by the sale of all government assets net of debt repayment. If the entire government has a negative net asset position because of past borrowing from the public, then the present value of all revenues must exceed the present value of all non-interest expenditures to keep the level of fiscal imbalance equal to zero. If, however, FI is applied to a specific program, At would represent the assets (or if negative, the debts) that the program had accumulated since its inception. The fundamental definition makes it clear that FI must be zero for the federal government’s fiscal policies to be sustainable over time. The government cannot spend and owe more than it will receive as revenue in present value. In other words, while the government can spend more than it collects in taxes on some generations, other generations must eventually “pay the piper,” thereby returning the fiscal imbalance to zero. (Gokhale and Smetters, p. 4)
Many who come across the fundamental definition of fiscal imbalance for the first time wonder why the government must keep FI equal to zero. They ask, “Why pay off the debt? Why not continue to borrow?” The answer is that the measure of fiscal imbalance would not be altered if the government did continue to borrow. If the debt is not paid off, interest payments must be made in perpetuity and would have a present value equal to the debt. Indeed, many combinations of future interest
16
G.H. Borts
payments and repayments of principal have a present value equal to the debt. So what looks like an easy solution to future problems of inadequate federal revenue is a mirage.
Generational Imbalance A second use of the concept of fiscal imbalance is to measure generational imbalance. A variety of policies may not alter fiscal imbalance and yet produce a considerable swing in the distribution of income between present and future generations. To measure this redistribution, Gokhale and Smetters introduce the concept of Generational Imbalance (GI) with its fundamental definition: GIt = PVEL t − PVRL t − At
(2)
Here PVEL t represents the present value of government non-interest expenditures on behalf of the current generation, PVRL t represents the present value of government revenues collected from the current generation, and At represents the value of the government’s net assets. Therefore, GI captures the part of FI arising from all transactions with past and living generations throughout their lifetimes. The projected contribution to FI by future generations simply equals the difference, FI minus GI. (p. 5)
This is a formal description of the folk wisdom that today’s government budget deficits must be paid by our children. Gokhale and Smetters explain the significance of the two measures, FI and GI: In the future, policymakers must achieve two objectives simultaneously: First, they must reduce the Fiscal Imbalance to zero by either increased taxes or reduced spending, or a combination of both. This can be accomplished in a myriad of ways, each of which will affect the burden placed on future generations differently. For example, lowering the growth of entitlement benefits—which affects those about to retire—will be more beneficial to future generations than increasing, say, payroll taxes—which leaves today’s older generations unaffected but negatively impacts today’s workers and future generations. Hence, the second objective for policymakers is to choose a policy that delivers the best trade-off in costs imposed on different generations. The GI measure offers policymakers a parsimonious approach for analyzing this issue and choosing among different sustainable paths. (Gokhale and Smetters, p. 7)
Gokhale and Smetters do not analyze default on government debt as one of the policy options. Indeed, should a regime default on its debts, it would be unable to borrow at favorable interest rates and ultimately forced to reduce expenditures to a level consistent with immediate tax revenues. Despite the well-known history of governments that have pursued profligate financial policies after defaulting on their debt, default is not a sustainable option for a government that wishes to finance a long-run policy of expanding welfare expenditures by borrowing. Sooner or later, the bond market turns its back on borrowers who fail to make interest payments or
The Solvency of Federal Welfare Entitlement Programs
17
repay their debts. A regime that defaults on its debts is then condemned in the long run to expenditures no greater than its current tax revenues.2
Measuring America’s Fiscal Imbalance Calculations of Fiscal Imbalance created by federal entitlement programs indicate a burden on present and future GDP of some $70 trillion. (Kotlikoff, Staticide, p. 8) This is the excess of the present value of all future government expenditures including the interest on the national debt in excess of the present value of all the government’s future taxes. This fiscal gap is approximately 8.5% of the present value of GDP. As Kotlikoff says, “We need to devote that share of GDP every year for eternity to cover the shortfall.” (Staticide, p. 1) These resources must be provided annually by taxpayers over and above what will be collected from currently projected tax programs. It implies approximately a 50% increase in federal government receipts. Estimates of the future level of Generational Imbalance are equally startling. Thomas R. Saving estimates that Social Security, Medicare and Medicaid will impose additional debt obligations on future generations of $32.5 trillion. (Saving, p. 5) The growth path of these expenditures will begin to accelerate in 2011 with no plans to increase federal revenues at a rate that would permit refilling the Social Security and Medicare Trust Funds. The longer the delay in raising additional revenue for these programs, the greater will be the fiscal burden that falls on later generations. Not all economists take the fiscal dangers seriously. Witness the following statement by Paul Krugman, a well-known commentator on public economic issues: Intergenerational responsibility is a fine thing, but I can’t see why the cost of medical treatments that have not yet been invented, applied to people who have not been born, should play any role in today’s policy. (p. 7)
The answer has been made amply clear in the discussion of the $70 trillion fiscal gap. The estimate of the gap is based on the assumption that future tax revenue will be adequate to cover it. If, however, taxes are not increased or expenditures reduced, the gap will grow over time by the amount of revenue needed to cover the interest on the additional debt. The longer the delay in closing the gap, the larger its size relative to GDP, and the greater the proportion of the tax burden that falls on later generations. If Krugman’s views make their way into economic policy, the next generation of taxpayers will be in for a grim surprise.
The Need to Achieve Fiscal Balance In The Coming Generational Storm, Laurence Kotlikoff makes it clear that the coming fiscal crisis is not to be laid at the door of any particular government program. He
18
G.H. Borts
believes it is meaningless to waste political capital on reform of one program like Social Security, without curtailing deficits in all of the others as well. The Social Security deficits by themselves are small compared to the deficits of other programs like Medicare and Medicaid. Social Security (Old Age and Survivors’ Insurance) is financed through a trust fund that has been accumulating financial assets every year since the reforms of 1984. Social Security revenue over that period has exceeded its expenditures, and the cushion of federal debt held by the Social Security trust fund will last for many years. Indeed, the trust fund is expected to continue adding to its holdings of federal securities for another 10 years, after which it will start spending more than its revenue. This cushion of federal government debt is projected to last until approximately 2045, at which time the system will be forced to reduce its expenditures to equal FICA contributions. Medicare has a number of sources of funding. One is the Hospital Insurance (HI) Trust Fund that pays out for hospital charges under Part A of the Medicare Program. Under present projections, the HI Trust Fund will be exhausted in 10 years. Medicare also receives funds to cover expenditures under Parts B and D of the program. These are expenditures for physicians’ services, diagnostic tests, and pharmaceuticals. They are financed from the federal Medicare tax, 2.9% of total wages and salaries paid by workers and employers, from general government revenues, and from beneficiary premiums. The premiums are set equal approximately to 25% of total Part B and Part D spending, respectively. Note that Medicare does not cover all of the health costs of its beneficiaries. They share the costs under co-payment and deductible provisions of the program. In 2006, Part A required individuals to pay $952 of the cost of each hospitalization, and Part B generally requires them to pay 20% of their Medicare-approved health care bills in addition to a deductible. Some beneficiaries pay Medicare deductibles and coinsurance amounts from their own pockets, while others obtain private insurance to cover these costs. Some low-income Medicare beneficiaries are also eligible for Medicaid. For these dually eligible people, Medicaid covers most of these costsharing amounts required by Medicare.3 Also note that Medicaid has no trust fund and is financed out of annual federal and state budgets. The Medicare program has a greater and more immediate financial problem than that faced by Social Security. The Medicare program does not have enough projected revenue to cover projected future spending. Under current projections made by the Medicare Actuaries and presented in the 2006 Medicare Trustees Report, the Medicare HI Trust Fund is projected to be exhausted in 2018. The projected 75-year deficit for the Medicare HI Trust Fund is 3.51% of taxable payroll. That is, the Medicare HI payroll tax would have to be immediately increased from 2.90 to 6.41% to cover all projected spending over the next 75 years. Alternatively a reduction in Medicare Part A expenditures by 51% would be necessary to make the Medicare Trust Fund solvent. As a comparison, this Medicare deficit is relatively larger in magnitude than the Social Security Trust Fund deficit. An increase in the Old Age, Survivors, and Disability Insurance payroll tax from 12.4 to 14.4% or a reduction in Social Security benefits by 13% is projected to make the Social Security program solvent over 75 years.
The Solvency of Federal Welfare Entitlement Programs
19
The Medicare Supplementary Medical Insurance Program (SMI) is considered to be solvent by the Medicare Trustees only because Part B and Part D spending is required by law to be financed by general revenues. However, the consequences of increased spending on Medicare SMI may be just as dire. Without large reductions in Medicare SMI spending or increases in taxes, either Federal budget deficits will grow rapidly or dramatic reductions in spending for other Federal programs will have to be made. (CEA Report, p. 93)
Making Welfare Financially Viable A variety of changes in the structure and pricing of Social Security and Medicare will be necessary to make the present welfare system financially viable. Actions to eliminate the revenue-expenditure gap take four forms: increase program revenues; reduce program related benefits; change the federal tax system to increase the federal government’s overall tax receipts; and change the structure of retirement and health care benefits to reduce the cost of providing benefits.
Increase Program Revenues i. ii. iii. iv.
Raise Social Security tax rates. Raise the ceiling on wages and salaries subject to Social Security taxes. Raise the Medicare tax on wages and salaries. Raise Medicare beneficiary premiums. Part B and Part D premiums could be increased to cover more than the current 25% of Medicare spending. v. Devise Medicare beneficiary premiums, co-pay percentages and deductibles that increase progressively with beneficiary income. vi. Raise Medicare beneficiary co-pay percentages and deductibles. vii. Accelerate the increase in the minimum age for full Social Security eligibility. Under current law, the onset of full benefit eligibility does not reach the age of 67 until the year 2020. The main rationale for the above proposals is to charge individuals the cost that they ultimately impose on the federal budget as beneficiaries of Social Security and Medicare. It has been argued in opposition that there is no need to use Social Security and Medicare taxes to make the system financially viable, so long as the benefits are considered socially desirable. The present Social Security and Medicare taxes are regressive and discourage employment. Any revenue shortfall should be made good through direct appropriation from the federal budget. Direct budgetary support would not, however, eliminate the fiscal drain caused by the welfare programs. Rather, it would increase the pressure to raise other sources of federal income and cut other expenditure programs. There may be an argument for subsidizing poorer workers and elderly retirees with revenue from taxation imposed on higher income groups, but the majority of the beneficiaries of Social Security
20
G.H. Borts
and Medicare do not qualify for subsidy on those grounds. So it remains important to put these programs on a self-sustaining financial footing.
Reduce Program-related Benefits i. Accelerate the increase in the minimum age for full Social Security and Medicare eligibility. ii. Index Social Security benefits to the price level rather than the wage level. This would slow the growth of benefits. The proposed benefit reductions are motivated by the need to make the programs actuarially sound, adjusting benefits to reflect increased life expectancy and inducing workers to extend the period they remain active in the labor force. Life expectancy at birth has increased by 16 years since the time that Social Security was adopted. Part of the increasing burden of welfare expenditures under the present program arises from the fact that individuals now spend more years in retirement, financed in varying degrees by benefits from Social Security and Medicare. They are induced by the structure of benefits to retire earlier than they would under voluntary savings and insurance arrangements.
Change the Federal Tax System to Increase the Federal Government’s Revenue This goal may be accomplished by directly raising tax rates or by enlarging the definition of taxable income. One candidate for inclusion in the income tax base is the imputed rental value of owner-occupied housing. A similar suggestion was made by Laurence Kotlikoff in New New Deal, to be discussed in a later section. An approximate measure of the resulting increase in the income tax base may be seen from the following statistics that represent averages for the U.S. economy.4 Some 75 million owner-occupied dwellings had a median sales value of approximately $200,000 in the year 2006. This implies at least $15 trillion of wealth tied up in owner-occupied housing. Assuming for purposes of illustration a real capital recovery factor of 10%, the annual capital recovery component of imputed gross rent would be approximately $1.5 trillion. This amount (minus mortgage interest and property taxes) would then be taxed along with other traditional sources of personal income. Taxation of the rental value of owner-occupied dwellings has been attempted in a number of countries as well as jurisdictions in the United States. It is justified on the ground that such a tax would eliminate preferential treatment favoring those who own their dwellings compared to those who rent. The bias is eliminated when the home owner pays an income tax on the net rental value of the property, usually measured as the imputed gross rent minus mortgage interest and property taxes.
The Solvency of Federal Welfare Entitlement Programs
21
Perversely, the U.S. federal income tax at present contains provisions that increase the bias by excluding imputed net rent from taxable income yet allowing taxpayers to deduct mortgage interest and property taxes from the income tax base. Because of compliance and valuation issues, it is likely that a tax on the net rental value of owner-occupied dwellings is too expensive to administer. But phaseout of the mortgage interest and property tax exemptions could provide a source of additional federal revenue and at the same time reduce the preferential tax treatment in favor of ownership.
Change the Benefits Structure to Reduce the Cost of Providing Benefits i. Recast Social Security so that it conforms to the original goal of insuring the retirees against poverty in old age. With the income ceiling on the FICA tax base now raised to $102,000 for 2008, Social Security continues to aim at a wider range of income groups than those at or just above the poverty threshold. The financial goal of Social Security should be narrowed to the task of insuring the elderly against poverty during retirement. This would enhance the future viability of the system. It would reduce prospective outlays and permit use of the savings to meet future obligations imposed by the retiring Baby Boomers. It would also help to focus attention on reform as workers become aware of the diminished rate of return on their FICA contributions.5 Indeed, there is likely to be less interest in privatization of Social Security if the majority of workers see it as insurance against poverty in old age, rather than a major component of wealth needed to finance retirement at the workers’ current standard of living. ii. Change the reimbursement structure of Medicare to allow the introduction of competition and budget constraints.6 Projected Medicare expenditures are expected to grow more rapidly than projected program revenues. The first step in bringing Medicare expenditures under control is to change the method that is used to reimburse health care providers. Rather than fixing the price that Medicare will pay for every medical procedure under the sun, these rates should be set by market demand and supply, relying in the first instance on negotiations between buyers and providers of medical service. The current fee-for-service structure should be eliminated and replaced by a market similar in structure to the market for automobile insurance, with the proviso that basic premiums are paid by Medicare. Under auto insurance, an adjuster verifies the claimed cost of repairing property damage, and attempts to protect the insurance company (and ultimately its customers) against over-billing, use of improper parts, excessive hourly charges, excessive number of hours, etc. Not all repair shops charge the same hourly rate. Some will charge less to get business during slow periods or business that does
22
G.H. Borts
not require immediate attention. Concessionary rates are likely to be offered to new customers or to customers with large and steady business volume. The cost of health care should be determined by negotiation between insurance companies and service providers. An insurance company that offers a lot of business should be able to receive lower rates than a company that does not. The system will work best if there is competition among insurance companies and among health care providers. This means allowing free entry by insurance companies, by providers, and by health care personnel. For example, it would require cross licensing that allows physicians licensed in one state to practice in neighboring states. It would require elimination of anti-competitive state regulation, so that providers are free to establish surgical and other services outside of hospitals and insurers are free to design health insurance contracts unconstrained by state imposed mandates to cover high risk or high cost experimental procedures. Individuals should be free to obtain health insurance from any insurer in the country, so long as it meets standards of prudent behavior and financial practice. Using insurance to cover health costs will, however, require supervision of the structure of premiums to make sure that Medicare subscribers receive adequate insurance protection without the creation of a pool of high risk “orphans” who face higher rates or informal blacklisting because of adverse health conditions. It will therefore be necessary to enforce the rule that insurance companies cannot refuse coverage to potential customers. However, older beneficiaries will need greater insurance coverage than younger, based on existing Medicare experience. Under the proposal, Medicare will purchase an insurance policy for every beneficiary, providing coverage for a ceiling level of health care expenditure for 1 year. Every year the beneficiary’s coverage level rises to match the average claims experience of his/her age cohort. Medicare pays for the policy.7 In addition, any beneficiary will be free to purchase additional insurance coverage out of his/her own pocket. The additional coverage will pay health care costs in excess of the standard Medicare purchased policy.8 Of course, from year to year the insurance company will face uncertain profits because the gain from paying the bills for healthy members of an age cohort may not always cover the loss from providing care for the sick members. The company’s willingness to bear this risk is compensated by the premium it receives from Medicare as well as the premium paid by the beneficiary for additional coverage. The individual’s total insurance coverage is thus the sum of what Medicare grants his/her age cohort and the additional insurance that he/she purchases. Once that decision is made, the beneficiary is responsible for paying any health care costs in excess of the combined coverage level. The given insurance level thus acts as a budget constraint, influencing the beneficiary to find the most effective health care protection out of his/her coverage, and in the process avoid so far as possible extra outlays beyond the level allowed by the combined coverage. This should lead to more efficient and higher quality choices of health care service. The introduction of competition, private insurance, a government expenditure ceiling, and the elimination of fee-for-service reimbursement will slow the growth of Medicare expenditure. But there is no way to be certain that the cost containment
The Solvency of Federal Welfare Entitlement Programs
23
is sufficient to make the system financially viable over the long run. In response to revenue shortfalls, it will be necessary to adjust the revenue structure upward to increase the share of costs borne by beneficiaries. There are a variety of ways to accomplish this: a. Raise the level of co-pays and deductibles paid by beneficiaries. b. Introduce a progressive structure of fees on beneficiaries to help cover Medicare insurance premiums. c. Reduce Medicare’s share of purchased insurance, creating an incentive for beneficiaries to increase their own expenditure on insurance. d. Inform the insurance companies and the providers that Medicare is no longer able to increase expenditures on insurance at the historical rate of growth in health care costs. Armed with this information, insurers and providers will have a better idea of the prices that will be consistent with the planned aggregate budget levels. The providers will recognize the likelihood that they will lose business if they do not stabilize their prices. One important effect of this last adjustment will be on the cost of inputs into the health care system. It is customary for individual purchasers to treat such costs as beyond their influence, because no single buyer, or indeed no single insurance company, is likely to have much effect on costs such as the wages of health care personnel or the cost of equipment.9 But any reduction in the rate of growth of federal outlays for Medicare services is likely to change that picture. Service providers are more likely to offer lower prices to obtain business if they believe that business is static or shrinking, and that their competitors will be willing to quote lower prices to take their customers.
A Utopian Perspective on Welfare The previous sections of this paper were devoted to analysis of policies to eliminate the financial imbalance in the federal budget created by Social Security and Medicare. This earlier perspective is more limited in scope than two recent proposals, one to privatize Social Security and the other to create a national program of universal health care. Such projects were put forward by Laurence Kotlikoff in New New Deal. The earlier analysis may be applied to evaluating their effectiveness and financial viability.
Privatizing Social Security Kotlikoff would not only privatize Social Security, but he would reconstruct the federal tax system so that its revenue would rely almost totally on the proceeds from a new federal sales tax. This would replace all other sources of federal tax
24
G.H. Borts
revenue, chief among them being the personal income tax, the corporate income tax, the Social Security tax and the Medicare tax.10 The federal sales tax must provide sufficient revenue to cover future retirement funding needs and all other federal spending obligations. This includes the cost of his proposal for a system of universal health care as well as the transitional funding cost of moving from a pay-as-you-go Social Security system to the proposed privatized system. Kotlikoff would, however, put in place the requirement that workers contribute a fraction of their salaries to funding their future pensions. Responsibility for the retirement contribution formerly made by employers would be taken over by the federal budget. The existing Social Security program would be stopped. No additional Social Security taxes would be collected, no additional benefits accrued, and no additional workers added to the Social Security roster. Existing Social Security benefits would continue to be paid out to current retirees, and in the future to those retirees who have accumulated benefits under the old system prior to its close. In place of the Social Security system, Kotlikoff would create the Personal Security System (PSS), which all individuals of working age would be required to join. Those who are employed will contribute to their own future pensions. The federal government would make equivalent contributions on behalf of workers who are unemployed or earn substandard wages. The government will also replace the former employer contributions with funds derived from its own tax revenues. The new pension system, like the old, would be designed to replace approximately 40% of the current worker wage level. Under PSS, accumulated retirement balances would be invested in a marketweighted global index fund of stocks, bonds, and real estate securities. Everyone would have the same portfolio and receive the same rate of return. The government would guarantee that at retirement the accumulated account balances could not erode in real value due to inflation.11 The main benefit of PSS is to create a financial asset that is the property of each worker and reflects the accumulated pension fund contributions during his/her working life. In contrast, under the present Social Security system retiree pensions are paid from the proceeds of tax revenues contributed by presently employed workers and their employers. This reform will make the future retiree’s pension independent of variations in worker tax contributions. Under PSS, workers would be able to invest their accumulated contributions in assets that have a higher rate of return than available under the present pay-as-yougo system. The rate of return on the market basket of assets underlying the entire American economy is approximately a real 6%. The rate of growth under a payas-you-go system is approximately the overall growth of tax receipts per retiree, and this is the product of growth in contributions per worker and the growth in the number of workers per retiree. In the next 50 years, there is expected to be a continual drop in the rate of return to Social Security contributions because of the decline in the number of workers per retiree. In the future, the pay-as-you-go system will be a bad investment. PSS would provide a number of additional advantages:
The Solvency of Federal Welfare Entitlement Programs
25
a. Married or legally partnered individuals would share contributions and benefits. b. To the extent that the market basket of securities rises with the price level, pension account balances would be protected against inflation. c. Workers who die before retirement age would bequeath their account balances to their spouses/partners or children. d. Elimination of the personal income tax would do away with the present double taxation of worker’s contributions to Social Security, once when earned and once again when paid out as pensions. PSS also has major weaknesses. First there is the need to find financing to carry out the transition from the present pay-as-you-go system. The financing gap occurs because privatization requires diversion of current Social Security tax contributions from their original role of paying pensions to their new role of accumulating wealth in the accounts of presently employed workers. The money to support current retirees will have to come from new tax revenues or from the sale of new federal debt. Note that there is one special condition under which it would be sufficient to finance the transition from the sale of new debt to the PSS Trust Fund. There would then be no need to raise taxes or increase the outstanding debt of the federal government. If the annual accumulation of government debt in PSS retirement portfolios equals annual Social Security pension payments to existing retirees, pensions for current retirees would be completely funded from the PSS retirement accumulations of current workers. The debt accumulated in PSS pension accounts would not be sold to the public until PSS beneficiaries retired and began to draw their pensions. There are two reasons, however, why the condition is unlikely to be met. First, under PSS, workers will be unable to put all of their retirement assets into federal debt even if they so desire. They must invest only in the worldwide security bundle. Second, over time the number of retirees will increase relative to the number of workers, so that workers in total could not save enough under PSS or any other privatized system to fund all of the pension obligations accrued under the old Social Security system. The conclusion is that new tax revenue or sale of new debt to the public will be needed to make the transition work. Only a small fraction of the cost can be funded from retirement tax contributions made by the current generation of workers. A second weakness of PSS is that it falls short of the need to create assets that retirees will be able to pass on to their heirs after death. Under PSS, the retiring worker’s accumulated assets are placed in a fund that pays benefits to retirees in the same age cohort. The undisbursed component of these assets remains in the fund at the time of the retiree’s death and is not transferred to the retiree’s estate. It appears that the elimination of the Social Security system is not enough to get rid of the vestiges of paternalism. If the compulsory annuity feature of the retirement pension were eliminated, so that assets could be inherited by the retirees’ heirs, PSS would be a desirable alternative to the present Social Security system. Yet the transition costs may be too great
26
G.H. Borts
to justify the ultimate benefits. The appropriate time to transit from pay-as-you-go to a privatized system is a period when the worker-retiree ratio is large enough and stable enough so that worker contributions exclusively invested in government debt are sufficient to finance retiree benefits. We are not in such a period and are not likely to be in the foreseeable future. Note, finally, that there are easier ways to reduce the budgetary drain of the present Social Security system. An increase in the FICA tax from 12.4 to 14.4% would be sufficient to put Social Security in the black for the next 75 years. (CEA Report, p. 93)
Kotlikoff’s Plan for a Uniform Comprehensive System of Health Insurance, the Medical Security System (MSS) Kotlikoff’s plan to create a comprehensive system of federal health insurance (MSS) is broader and more ambitious than his proposed privatization of Social Security12 . There is no federal program of universal health insurance presently in place, and it would have to be created from the relevant parts of the existing Medicare and Medicaid (M/M) programs. In making the transition from M/M to a universal system of health insurance, it is necessary to consider that the projected future budget deficits from operation of the existing M/M programs are a far greater proportion of existing and projected program expenditures than is true of Social Security. The latter has a trust fund that has been accumulating financial assets since the last reform in the 1980s. The trust fund will be able to meet its obligations fully for another 40 years. Before that happens it will sell off its accumulated financial reserve holdings of federal government securities. In contrast, Medicare has only a small trust fund to cover hospital expenditures, and that will be exhausted in 10 years. Other federal Medicare and Medicaid expenditures are funded directly from current federal revenue receipts. If for no other reason, the enlarged scope of a universal system guarantees that it will be more expensive to operate than M/M. It is estimated that approximately 15% of the population, some 45 million people, are not covered by health insurance, and they will be covered under Kotlikoff’s MSS system. Health care costs are bound to increase in the future with the growing share of elderly in the population. There is enormous political pressure to use federal money to comply with the social imperative to prolong human life. Indeed, health care costs are already increasing faster than the growth of real wages, so that M/M expenditures are rising over time as a fraction of GDP and are expected to continue that pace unless there is a change in the structure of the system.13 The new MSS would replace Medicare, Medicaid, and all private health insurance. All legal residents of the U.S. would be eligible to join, with the government purchasing a health insurance policy for each individual. The policies will be issued by private insurance companies, at prices based on public information about the
The Solvency of Federal Welfare Entitlement Programs
27
health condition of each insured person. The insurance rates will be based on competition among insurers. Profits from the insurance policies will be limited by competition among insurers. As discussed earlier, the cost of health care procedures will be set by competitive bargaining between insurance companies and groups of health care providers, operating in an environment where they have advance warning from the federal government that total federal expenditures will not increase more rapidly than the level of real wages. The fee-for-service system of reimbursement will be eliminated. The main advantage of the proposed MSS system is that it will bring health insurance to the 45 million people who are either too poor to afford it, willing to take their chances that they will not need it, or rich enough to pay their own bills without insurance. The main weakness is that MSS will be far more expensive than the present health care system, which is itself financially unsustainable in the long run. There are many reasons: First, there are 300 million people to cover under MSS, compared with the relatively smaller number of elderly and poor covered under M/M. Second, because of its coverage of the entire U.S. population, not just the elderly and the poor, there is some likelihood that the bargaining process between insurance companies and health care providers will not be sufficient to keep costs down. We have already witnessed past failures of HMO attempts to keep costs down. They buckled in the face of public outcry against limits on reimbursement for medical procedures offering low likelihood of effectiveness in preserving the health of beneficiaries. If the public demands that expensive and experimental procedures be insured, the political system will put pressure on health care providers and insurers to include coverage in the policies bought by the federal government. It may be difficult to prevent the MSS from exceeding budgetary limits in the face of public pressure. We have already seen that much of the high cost of modern health care is a response to public pressure from both beneficiaries and providers to use high-tech medicine and the latest in pharmaceutical products, and to use it in environments where the beneficiary is subject neither to pre-screening for eligibility nor long waiting lines. Third, the incentive structures in MSS may be too weak to induce consumers of health care to compare quality, effectiveness, risk, and cost in making a decision whether it is worthwhile to undergo medical treatment. All of the efficiencyincreasing incentives in MSS focus on health care providers and insurers, but not the consumer. Nevertheless, Kotlikoff’s proposal is a step in the right direction, because it contains incentives for producers and insurers to keep costs down. It eliminates some of the incentives to provide excessive health care, such as tax deductibility of employer provided health insurance. To be fully efficient, however, it will require the stimulation of competition among health care providers and among insurers, as well as the elimination of state barriers to competition. But even an efficient MSS will be far costlier for government to finance than the present Medicare and Medicaid system.
28
G.H. Borts
Notes 1. See the testimony by Thomas R. Saving, the Public Trustee of Medicare Trust Funds, p. 5. 2. One may ask why the above conclusion appears to apply only loosely to Argentina, which defaulted on its federal government debt in 2002, and 5 years later again appears to be spending more than current tax revenue. The reason is that every country has another source of revenue, namely involuntary taxes paid by its citizenry through the inflation of the money supply. As it has in previous decades, Argentina is again using inflationary issuance of money as a source of tax revenue. (See Mary Anastasia O’Grady.) Argentina also relies on gifts from other regimes, such as Venezuela, that wish to gain influence on its foreign policy. Moreover, the Argentine Treasury has been lucky since 2002 because of worldwide increases in the prices of its primary product exports. 3. See President’s Council of Economic Advisors (CEA), p. 92. 4. Data sources are the National Association of Realtors, “Existing Home Sales,” 2006, and U.S. Department of Housing and Urban Development, “American Housing Survey,” 2006. 5. In a pay-as-you-go pension system, the rate of return to a worker’s lifetime contributions is calculated on the difference between benefits paid out during retirement and contributions paid in during one’s working life. The decline of the labor force relative to the number of contemporaneous beneficiaries will lower the rate of return on contributions and could actually push it into negative territory. 6. The ideas in this section are a modification of Laurence Kotlikoff’s proposal (see The Coming Generational Storm and New New Deal) to create a national system of health care. The proposal here is limited to making Medicare and Medicaid more efficient and financially viable. In the last section, I will explain the problems inherent in expanding coverage to the entire population. 7. It may also be necessary for Medicare to pay a higher premium on a company’s cohortspecific policies if the company finds that its customer pool is weighted toward individuals with higher risks than their cohort averages. 8. A version of this arrangement is proposed in Kotlikoff’s New New Deal. The magnitude of the beneficiary’s standard health insurance coverage depends on information contained in his/her medical history, not on the mean level of cohort health costs. In principle this is an improvement over the proposal in the above text, but I am concerned about its incentive effects. Risk-averse beneficiaries may find it in their interest to increase use of the health care system to establish a dossier of treatments that qualify for a higher level of Medicare-paid insurance. Also worth recording is the horror that my students express at this intrusion of Big Brother into the physician’s examining room. 9. Pharmaceuticals are likely to be an exception. 10. To be sure that revenue from the proposed tax system is adequate, Kotlikoff proposes expanding the scope of taxable consumer expenditures to include imputed expenditures on owner-occupied housing. This component of the sales tax base would correspond more closely to imputed gross rent than net rent. See the discussion in the section entitled “Change the Federal Tax System. . . .” 11. The guarantee implies that the nominal rate of return on the worker’s account balance could not be less than the cumulated rate of inflation. Considering that some workers nearing retirement will spend a very short time under PSS, variation in security prices might make the guarantee very expensive. For example, the Standard and Poor’s 500 Index showed a small loss between its average levels of 109.2 in 1972 and 103.01 in 1979. Between those two periods, however, the real value of stock portfolios eroded by 46% because of the increase in the Consumer Price Index from 41.8 to 72.6. 12. According to Kotlikoff, the plan has been proposed by a number of economists, including Peter Ferrara, John Goodman, Victor Fuchs, and Ezekiel Emanuel. 13. Gokhale and Smetters indicate that between 1980 and 2001, health care expenditures grew by 2.3% points faster than GDP.
The Solvency of Federal Welfare Entitlement Programs
29
References Gokhale, J. & Smetters, K. (2003). Fiscal and Generational Imbalances: New Budget Measures for New Budget Priorities, Federal Reserve Bank of Cleveland, Washington, DC: The AEI Press. Kotlikoff, L. (2006). “Averting America’s Bankruptcy with a New New Deal,” (New New Deal) The Economist’s Voice. February 2006, rev. March 2007. Kotlikoff, L. (2007). Staticide, America’s Suicidal Healthcare Status Quo, Boston University Working Papers. Also http://people.bu.edu/kotlikof/Staticide%20March%205,%202007.pdf. Kotlikoff, L. & Burns, S. (2004). The Coming Generational Storm, Cambridge: The MIT Press. Krugman, P. (2004). “Confusions About Social Security,” The Economists’ Voice, 2(1), Berkeley Electronic Press. National Association of Realtors (2006). Existing Home Sales, Chicago, IL. O’Grady, M.A. (2007). “Back to the Future in Argentina,” Wall Street Journal, October 29, 2007. Saving, T.R. (2005). “Medicare Now and in the Future.” Testimony before the U.S. Senate Budget Committee, February 7, 2005. U.S. Council of Economic Advisors (CEA) (2007). Economic Report of the President, Washington, DC: United States Government Printing Office. U.S. Department of Housing and Urban Development (2006). American Housing Survey.
The Corporate Sector as a Net Exporter of Funds: Additional Evidence John B. Guerard, Jr.
Using aggregate data, Schwartz and Aronson (1966) documented the role of the corporate sector in generating more funds than it can profitability use over 40 years from 1924–1964. One of the primary issues behind the corporate sector exporting funds was the controversy of dividends. Schwartz and Aronson noted that aggregate dividends far exceeded net new external financing. In this study, we present evidence for the 1971–2006 period for all stocks covered by the Compustat tapes, some 200,000 firms (approximately 2,300–6,000 firms per year). We substantiate the original Schwartz and Aronson hypothesis of the corporate sector as a net exporter of funds, and offer additional evidence as to how the components of the net exporter sector calculations have evolved over time. We address additional questions with regard to debt and equity issuances, repurchases, the relationship of dividends and stock prices, and whether significant sector effects are present. Schwartz and Aronson (1966) held that a “balanced economic growth model” carried the assumption that the amount of income from capital equals the total of aggregate savings and real investment and that perforce the rate of return on capital and the growth rate of the economy are equal. However, the cash flow generated by the operating assets exceeds the amount of possible net real investment; and it necessarily follows that at the end of the fiscal period, the overall corporate sector will have more funds than it can desirably re-invest in the business. In short, the data shows that if in any given year the growth rate is 5.0% and the aggregate after tax-earnings on the real value of corporate equity is 10.0%, then the corporate sector one way or another will distribute about 50% of its equity earnings or profits to the rest of the economy. The outflow of cash in the form of dividends, interest paid on debt, buybacks, and repayment of debt has substantially exceeded new funds raised on the capital markets, primary raised by issuing debt. In 1971, the corporate sector exported over $19 billion of funds, and by 2003 the corporate sector funds exported grew to over $400 billion.1 The surplus of funds over any possible reasonable capital investment policy is the rationale behind the cash buyback of shares and the payment of dividends. In short, the corporate sector continues to be a net exporter of funds to the rest of the economy, and the amount has risen almost consistently throughout the 1971–2006 period, except when net equity repurchases fell and new debt issues
J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_4, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
31
32
J.B. Guerard
rose in the mid- to late-1990s. If we define a variable to designate the net export of funds, ES, of the corporate sector, then ES = Dividends Paid + Interest Paid + Net Equity Repurchased − Net Debt Issued (1) The reader is referred to Table 1 for annual statistics of the components of ES taken from the Compustat database. A second way to examine the corporate sector net export of funds is to examine the dividends paid less equity repurchased, which also is shown in Table 1. Dividends paid exceeded equity repurchased of the Compustat firms from 1971–2004, although equity repurchases have risen relative to dividends since 1982. An inspection of the ES components, shown in Chart 1, reveals that interest paid has risen faster than dividends paid during the 1971–2006 time period. Net debt issues have risen at an undiminished rate, with the notable exception of 2001–2005. Net equity repurchases, positive in 1988, increased substantially in the 2002–2006 time period.2 In this study we, as a result of several conversations with Professor Schwartz, specifically examine the role of dividends and equity repurchases in corporate exports. As noted previously, dividends increased during most of the 1971–2006 time period (with the exception of 2006). Dividend policies show a high correlation of dividends with earnings, although as earnings rise, there is usually some lag before former payout levels are resumed (Fama & Babiak, 1968). Buybacks are more difficult to predict. In the past, before the prevalence of buybacks, if a company had a “normal” dividend payout policy, the prospective investor would not be far wrong if he concentrated on the trend of earnings in evaluating the worth of a share of stock. Equity repurchases increased substantially during the 1971–2006 time period. Buybacks exceeded dividends in 2005 and 2006 in our database. Why? Equity repurchase decisions, whether privately negotiated or via a tender offer, specify the number of shares, the repurchase price, and the expiration date by which the firm chooses to repurchase its equity (Dann, 1981). Dann examined 143 cash tender offers to repurchase equity made by 122 different firms during the 1962–1976 period and reported a 22.46% tender offer premium, relative to the previous day of the announcement (20.85% relative to the one-month period before the announcement). Firms repurchase equity to enhance stockholder wealth! There were marginal debt effects, and approximately 95% of the enhanced value accrued to stockholders. Lakonishok and Vermaelen (1990) reported excess returns to repurchases continuing through 1986, but at a (slightly) diminished rate.3 The level of current and projected earnings impacts dividend policy and potential buybacks (Bierman, 2001). A forecast of dividend levels can be obtained by studying past payout rates and attempting to predict future earnings levels. We specifically examine the period after that studied by Schwartz and Aronson (1966) to address several issues: (1) whether dividends exceeded net new equity issues, which they did; (2) and whether dividends exceeded net new capital issues, which they did not do because of net new debt issues.
Funds exported
19403.1 25579.6 26907.0 26782.5 22386.8 39436.4 47481.1 48638.7 50089.3 52304.5 57820.1 88695.5 111452.8 123552.7 128281.3 110454.1 192420.6 227834.7 219173.3 257582.5 217577.9 241470.8 197573.5 251487.8 234389.8 251203.4 226000.1
Year
1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
21353.4 24334.3 26252.7 29010.6 30476.1 34861.6 40964.9 45794.7 53414.5 58801.3 64502.2 68109.3 72620.3 77052.6 81745.4 92347.6 103140.3 122067.9 122976.8 129935.7 132470.6 137089.6 144714.0 158750.6 186295.4 206642.8 211645.6
Dividends
18089.1 20366.4 27245.0 36889.9 38493.1 39546.2 43718.9 50960.9 62522.8 80535.4 99845.3 108184.5 105122.5 117791.4 126787.6 135504.9 150275.8 184544.2 230065.4 233759.6 233711.4 219722.5 213924.1 232164.4 278433.5 297219.0 333989.3
Interest paid
Net debt issued 13747.6 13461.8 19986.3 33189.7 34933.2 21771.9 24302.4 35106.2 48335.2 61029.5 73388.9 57537.1 12565.5 68249.7 75749.7 100325.2 55587.5 100271.1 133193.1 112516.1 100829.2 59505.0 74710.6 90625.2 209583.1 187909.6 314330.9
Net equity repurchased −6290.9 −5642.3 −6604.0 −5928.3 −11649.3 −13199.5 −12900.3 −13010.8 −17512.7 −26002.5 −33138.6 −30061.2 −53724.6 −3041.4 −4501.9 −17073.3 −5408.0 21493.6 −675.7 6403.3 −47775.0 −55836.4 −86354.0 −48802.0 −20756.1 −64749.0 −5303.3 1382.6 1976.4 3818.5 1987.1 1287.5 2309.7 4520.9 4911.0 6001.9 7201.0 7071.9 11441.3 10951.8 33008.8 47275.0 49245.7 61949.9 57415.9 59013.3 49706.5 31509.8 39625.7 46109.5 54060.9 86492.7 112268.5 162431.7
Equity repurchased 19970.8 22357.9 22434.2 27023.5 29188.6 32551.9 36444.0 40883.7 47412.6 51600.3 57430.3 56668.0 61668.5 44043.8 34470.4 43101.9 41190.4 64652.0 63963.5 80229.2 100960.8 97463.9 98604.5 104689.7 99802.7 94374.3 49213.9
Dividends less equity repurchased
Table 1 Funds exported, 1971–2006 ($MM)
15062.6 18692.0 19648.2 23082.3 18826.8 21662.1 28064.6 32784.0 35901.8 32798.6 31363.6 38048.1 18895.8 74011.2 77243.4 75274.3 97732.3 143561.5 122301.1 136339.0 84695.7 81253.3 58360.0 109948.6 165539.3 141893.8 206342.3
Dividends less net equity 1314.0 5230.2 −338.1 −10107.4 −16106.4 −109.7 3762.2 −2322.3 −12433.4 −28230.9 −42025.2 −19489.1 6330.3 5761.5 1493.7 −25050.9 42144.8 43290.5 −10892.1 23822.9 −16133.5 21748.3 −16350.6 19323.4 −44043.8 −46015.8 −107988.6
Dividends less net new capital 2300 3329 3911 4391 4404 4438 4435 4394 4364 4449 4708 4758 5192 5298 5355 5625 5888 5779 5662 5620 5782 6008 7124 7553 7695 8213 8255
N of firms
The Corporate Sector as a Net Exporter of Funds 33
Funds exported
163142.4 57423.2 31602.4 200507.4 410245.3 487086.9 700771.1 982594.9 686557.3
Year
1998 1999 2000 2001 2002 2003 2004 2005 2006
254009.5 243240.2 253861.1 269390.2 283916.1 318087.9 400185.9 484068.2 468005.4
Dividends
395905.7 431477.8 553350.4 588656.3 506826.6 494466.0 501615.3 613358.9 716989.4
Interest paid 23305.0 −53927.8 −131271.3 −39501.1 41666.3 36205.2 93343.2 254315.2 410001.4
Net equity repurchased 510077.9 563367.0 644337.8 618039.0 422163.8 361672.4 294373.3 369147.4 908438.9
Net debt issued 217806.3 222684.6 223850.6 206198.6 201382.9 217445.8 350611.5 509698.9 680467.6
Equity repurchased
Table 1 (continued)
36203.2 20555.6 30010.5 63191.6 82533.2 100642.1 49574.4 −25630.7 −212462.2
Dividends less equity repurchased 277314.5 189312.4 122589.8 229889.1 325582.5 354293.1 493529.1 738383.4 878006.8
Dividends less net equity
N of firms 7894 7652 7526 6935 6496 6179 6114 6000 5243
Dividends less net new capital −232763.3 −374054.6 −521748.1 −388148.9 −96581.3 −7379.2 199155.8 369236.0 −30432.1
34 J.B. Guerard
The Corporate Sector as a Net Exporter of Funds
35
Chart 1 The corporation as a net exporter of funds, 1971–2006
The amount left after common dividends are paid represents retained earnings or earnings reinvested in the firm. Retained earnings add to the equity of the common shareholders, and the amount of correlated funds can be used to finance additions to the operating assets of the corporation or to retire debt. Additional assets, properly employed, should add to the future earnings of the corporation. An increase in assets financed by ownership capital as opposed to debt improves the credit standing of the firm and enables it to acquire debt funds at a relatively lower rate. New debt issues exceed new equity issues by a multiple exceeding eight; a result consistent with Dhrymes and Kurz (1967), and Guerard, Bean, and Andrews (1987). Funds represented in earnings should increase the future profits of the shareholders and eventually result in buybacks or higher dividends. It is not the increment in the book value of the shares, but a hoped-for sequence of increased earnings that makes retained earnings of value to the shareholder.4 To illustrate the corporate fund generation process, we show the ten largest and smallest corporate exporter firms in 1983 (see Table 2) and 2006 (see Table 3). AT&T, IBM, and several of the large oil companies dominated positive corporate exports in 1983 as they paid large dividends and interest and generally repurchased more debt than was issued (which made a great deal of sense given the level of interest rates in 1983). A similar process occurred in 2006 as Microsoft, Pfizer, and the oil companies dominated the largest corporate exporting firm (IBM fell to only the 24th largest exporter in 2006). IBM is a very interesting individual case, as it is generally a large net corporate exporter of funds, except in 1993, its near-bankruptcy year. IBM pays large dividends, repurchases huge amounts of equity, is a net issuer of debt, and issued no new equity after 1995. Indeed, IBM’s repurchase of equity far exceeded its dividends paid during the
36
J.B. Guerard Table 2 1983 Corporate sector exports ($MM) Dividends paid
Interest paid
Net equity repurchased
Net debt issued
CoName
ES
SLM CORP MAXUE ENERGY XEROX MCI 7-ELEVEN SIGNAL CERIDIAN DELTA AIR STONE CONTAINER COMDISCO ROYAL DUTCH PETROLEUM ALTRIA IBM OCCIDENTIAL PETROLEUM DUPONT FANNIE MAE GM BP EXXON AT&T
−1742.2 −1166.5
3.0 134.8
711.3 143.3
−356.5 −1547.4
2100.0 −93.6
−955.8 −826.2 −733.4 −625.0 −599.4 −597.8 −515.2
285.1 0.0 35.3 88.0 22.8 40.0 6.5
190.4 185.8 78.8 78.0 471.5 92.9 25.1
−820.0 −197.1 −276.4 −948.0 −26.7 0.0 −137.2
611.3 814.9 571.2 −157.0 1067.0 730.5 49.6
−500.0
4.0
53.7
−53.4
504.4
1973.0 2004.1 2069.0
683.0 365.8 2251.0
904.0 362.7 429.0
0.0 89.2 −788.0
−386.0 −1186.4 −177.0
2412.6 2416.0 3248.7 3383.9 3696.0 4342.7 5098.8
240.7 595.0 10.5 879.3 65.0 2673.8 5495.9
634.5 792.0 7988.7 1401.8 913.0 1019.7 4307.2
133.7 −106.0 −7.6 212.0 −30.0 762.5 −3902.1
−1403.7 −1135.0 4742.9 −1314.8 −2178.0 113.3 802.2
1995–2006 time period (see Table 4). IBM’s dividends exceeded its net new equity issues and net new capital issues in all years except 1993. Guerard and Schwartz (2007) used IBM as one of three firms to illustrate corporate financial policy.
Why Do Firms Issue Debt? In the previous section, we discussed the Schwartz and Aronson (1966) hypothesis that the corporate sector was a net exporter of funds. In Chart 1, the reader can see the vast increase of corporate debt, leading to the question of why firms issue debt. The relations among the investment, dividend, and external finance behavior of firms have been studied systematically, but not for a very large sample of firms over a very long period of time. Quite clearly, given the institutional milieu of the modern corporation, there exists at least a presumption that these three aspects of the firm’s decision-making process exhibit some interaction. Yet, in the current literature the view is frequently advanced that investment decisions are made on solely “real” (nonfinancial) considerations, that dividend policies are characterized by a considerable degree of inertia, and that the financing of investment by internal or external funds is a mere detail. Corporations rely on internal funds to finance capital
The Corporate Sector as a Net Exporter of Funds
37
Table 3 2006 Corporate sector exports ($MM) CoName
ES
Dividends paid
Interest paid
AIG ING ANADARKO PETROLEUM TELEFONICA CAMPHIA VALE FORD BAYER AMERICAN EXPRESS ARCELOR MITTA SLM CORP IBM ALTRIA PFIZER ROYAL DUTCH PETROLEUM VERIZON FEDERAL HOME LOAN MICROSOFT BP GM EXXON–MOBIL UBS
−25992 −22488 −15169
1690.0 3538.1 167.0
7009.0 0.0 730.0
−143.0 1377.8 55.0
2100.0 −93.6 611.3
−14685 −14510 −12654 −11386 −9662
3466.9 13.6 468.0 915.9 692.0
5116.5 5379.9 8783.0 1838.3 120.0
3096.0 308.8 −248.0 0.0 4093.0
814.9 571.2 −157.0 1067.0 730.5
−9452 −8584 11019 13458 13769 15871
522.0 398.4 1683.0 6954.0 7268.0 8142.0
1124.0 5122.9 981.0 1331.0 517.0 1713.0
−8.0 290.3 6399.0 768.0 6979.0 8047.0
49.6 14395.4 −1956.0 −4405.0 995.0 2031.0
16342 20516
4781.0 1310.0
2785.0 36883.0
1526.0 480.0
−7250.0 18157.0
20700 23857 30232 36912 40995
3594.0 7686.0 563.0 7628.0 2635.5
0.0 1196.0 16945.0 1184.0 66322.3
17106.0 15151.0 0.0 28385.0 2970.9
0.0 176.0 −12724.0 285.0 30933.0
Net equity
Net debt
investment, and this signifies a strong aversion to the use of the capital market. Thus, it would seem quite reasonable to suppose that the three decisions—to invest, to pay dividends, and to resort to external funds—are mutually determined. Hence, it is desirable to examine this problem in the context of a simultaneous-equation model. If our conjecture about the modus operandi of the system is correct, then we should expect that the coefficients of the jointly determined variables—investment, dividends paid, and external finance—would be significant, at least in several instances, where they serve as explanatory variables. Dhrymes and Kurz (1967) hypothesized interdependencies among these three decisions and econometrically estimated a simultaneous equation system for the dividend, investment, and new debt decisions for a sample of 181 firms over the 1947–1960 period. It is the purpose of this paper to study the question of establishing such a link and to elucidate the extent of the interdependence of these decisions for a much larger sample of firms over the 1952–2002 period. The approach employed makes use of a series of cross sections. In this study we use Compustat balance sheet and income statement data for the 1950–2002 period for firms with assets exceeding $200 million in 2002. Our main findings are the following: (1) a strong interdependence is evident between the investment and dividend decisions;
38
J.B. Guerard Table 4 IBM corporate exports ($MM)
Year
Funds exported
Interest Dividends paid
Net equity Net debt repurchased issued
Dividends Equity equity repurchased repurchased
1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
350.4 337.4 547.1 924.8 787.7 1129.5 2909.9 2260.9 380.3 1885.0 1510.0 1772.0 2069.0 2295.0 2416.0 4553.0 4874.0 2881.0 1955.0 3718.0 2874.0 5006.0 −2042.0 5634.0 9670.0 3760.0 3162.0 6213.0 9631.0 5676.0 8496.0 3358.0 9238.0 9263.0 7676.0 11019.0
598.2 626.2 654.3 819.7 987.0 120.8 1910.7 1763.1 1506.0 2008.0 2023.0 2053.0 2251.0 2507.0 2703.0 2698.0 2654.0 2609.0 2752.0 2774.0 2771.0 2765.0 905.0 585.0 572.0 686.0 763.0 814.0 859.0 909.0 956.0 1005.0 1085.0 1174.0 1250.0 1683.0
−214.6 −270.4 −325.0 −280.4 −284.5 −139.2 939.5 472.3 37.7 62.0 −423.0 −613.0 −788.0 −73.0 −133.0 1465.0 1290.0 1003.0 1788.0 491.0 129.0 90.0 −1213.0 −308.0 5526.0 5005.0 6251.0 6283.0 6645.0 6073.0 3906.0 3087.0 3232.0 5418.0 6506.0 6399.0
0.0 0.0 0.0 0.0 0.0 154.7 1245.2 813.4 454.0 484.0 0.0 0.0 0.0 0.0 0.0 1488.0 1425.0 1003.0 1788.0 491.0 196.0 90.0 0.0 10.0 5526.0 5005.0 6251.0 6283.0 6645.0 6073.0 1573.0 2438.0 3232.0 5418.0 6506.0 6399.0
70.0 78.4 97.1 69.1 62.6 45.0 40.4 55.2 140.5 325.0 480.0 514.0 429.0 456.0 532.0 604.0 619.0 802.0 1118.0 1446.0 1566.0 1461.0 1319.0 1247.0 748.0 747.0 760.0 741.0 750.0 737.0 271.0 180.0 663.0 571.0 761.0 981.0
103.3 96.8 −120.7 −316.4 −40.4 −20.0 −19.4 29.8 1303.8 510.0 570.0 182.0 −177.0 595.0 686.0 214.0 −311.0 1533.0 3703.0 993.0 1592.0 −690.0 3053.0 −4110.0 −2824.0 2678.0 0.0 1625.0 −1377.0 2043.0 −3363.0 914.0 −4258.0 −2100.0 841.0 −1956.0
598.2 626.2 654.3 819.7 970.0 1049.2 665.7 949.7 1052.0 1524.0 2023.0 2053.0 2251.0 2507.0 2703.0 1210.0 1229.0 1606.0 964.0 2283.0 2575.0 2675.0 905.0 575.0 −4954.0 −4319.0 −5488.0 −5469.0 −5786.0 −5164.0 −2950.0 −2082.0 −2147.0 −4244.0 −5256.0 −4716.0
(2) a strong interdependence is evident between the investment and new debt financing decisions; and (3) there is no compelling evidence to suggest that in estimating the structure one ought to use full information methods. That is, we find similar statistically significant relationships using ordinary least squares analysis and when using limited information and full information methods. The three aspects of the firm’s behavior on which this study is focused have been studied in the literature with varying degrees of intensity. Thus investment behavior has perhaps been studied most intensively. The integration of investment
The Corporate Sector as a Net Exporter of Funds
39
theory with the neoclassical theory of the firm can be traced to Tinbergen (1938). An extensive survey of the work of the last two authors can be found in Meyer and Kuh (1957).5 It is perhaps accurate to say that the main results of such studies lie in providing tests concerning the empirical relevance of the accelerator, capacityaccelerator, or profits (or rate of profit) theories of investment. The issue is not yet satisfactorily resolved, but it appears that neither the capacity accelerator nor the profits theory is alone sufficient. Rather, instead, a combination of elements of both is probably necessary to provide a satisfactory account of the empirical behavior of investment. Finally, external finance is more or less residually derived, and hence would depend on investment and dividends. Kuh’s estimation is carried out by single-equation methods, and his results seem to bear out the capacity accelerator theory of investment and the homogeneous Lintner hypothesis on dividends. In the case of dividend behavior, the main work on the subject (Lintner’s) views dividend disbursals as totally divorced from investment considerations. Miller and Modigliani (1961) formulated the perfect markets hypothesis, in which the dividend decision is independent of the investment decision, by deriving that the valuation process of the firm is independent of dividend policy and that firm value is dependent upon investment opportunities to produce earnings, dividends, or cash flow. The firm’s dividend policy is generally maintained until a permanent change in operations (earnings) has occurred. New capital issues raise funds from which research and development, dividends, and investments are undertaken. It is assumed that dividends and investments increases lead to new capital issues. Modigliani and Miller only allow the interdependence of the investment and new capital issues functions. As an empirical matter, Meyer and Kuh (1957) report that 75% of investment in manufacturing is internally “financed.” Brealey, Myers, and Allen (2005) and Dhrymes and Guerard (2005, Investment, Dividend, and External Financing Behavior of Firms: A Further Examination, Unpublished) report numbers indicating that internal financing may account for as much as 85% of capital investment. The same authors list an impressive catalogue of reasons why such a preference might exist. While this phenomenon may, in some part, be due to the peculiarities of the tax structure of the United States, it may also reflect imperfection in the capital market. The view taken in this paper is simply put as follows. Quite generally a firm faces an outflow of funds represented by its variable and fixed costs, its tax and dividend payments, as well as by its investment activities. On the other hand, it can rely on an inflow of funds represented chiefly by its sales and proceeds through various forms of external finance, viz., by bond or stock flotation. To the extent that a plausible objective for a firm is to grow, provided that its operations are profitable and that the capital market is less than perfect, it would follow that investment and dividend outlays are quite clearly competitive. Dhrymes and Kurz (1967) modeled the interdependence of the dividend, investment, and new debt decisions of 181 industrial and commercial firms during the 1947–1968 period and found the following: (1) strong interdependence exists between the investment and dividend decisions; new debt issues result from increased investments and dividends but do not directly affect them; (2) the interdependence among the two-stage least squares residuals
40
J.B. Guerard
compels the use of full information (three-stage least squares regression methods); and (3) the accelerator as well as profit theory is necessary to explain investment. The Dhrymes and Kurz study generated much interest in testing the perfect markets hypothesis. Mueller (1967) found significant interdependence among the research, advertising, dividend, and investment decisions in 67 manufacturing firms for the 1957–1968 period. Higgins (1972) examined the Dhrymes and Kurz rejection of the perfect markets hypothesis and produced a study showing independence of the investment and dividend decisions. Fama (1974) employed time series methodology and McDonald, Jacquillat, and Nussenbaum (1975) employed a cross-sectional analysis of French firms to find little evidence of imperfect markets. McCabe (1979) criticized the Higgins and Fama studies and using firms during the 1961–1970 period found evidence rejecting the perfect markets hypothesis. Studies by Grabowski and Mueller (1972), Peterson and Benesh (1983), Switzer (1984), and Jalilvand and Harris (1984) found evidence supporting the interdependence of the investment, dividend, and financing decisions. Guerard and McCabe (1992) and Guerard, Bean, and Andrews (1987) employed a diversified 303-firm sample for the 1975–1982 period and found significant interdependencies among investment and new debt and R&D and dividends. Thus, the evidence on the perfect markets hypothesis is mixed.
The Model The Structure of the Model The general (schematic) structure of the model is as follows : I1 = f1 (I2 , D, EF1, EF2; X1 , X2 , . . . Xn ) I2 = f2 (I1 , D, EF1, EF2; X1 , X2 , . . . Xn ) D = f3 (I2 , I2 , EF1, EF2; X1 , X2 , . . . Xn ) EF1 = f4 (I1 , I2 , D„ EF2; X1 , X2 . . . Xn )
(2)
EF2 = f5 (I1 , I2 , D, EF1; X1 , X2 , . . . Xn ) where I1 is investment in fixed assets; I2 is inventory and other short term investments; D is common stock dividends paid; EF1 is (net) external finance obtained by borrowing; EF2 is (net) external finance obtained by stock flotation; the Xi are predetermined variables, i = 1, 2, . . . n. The predetermined variables may include profits, depreciation, sales, long-term debt outstanding, etc., and will be introduced explicitly as the occasion arises. In addition, the firm faces the “budget constraint” I1 + I2 = EF1 + EF2 + P − D + Dep, where P and Dep denote, respectively, profits and depreciation allowances.
(3)
The Corporate Sector as a Net Exporter of Funds
41
If we use the constraint in (2), we can eliminate one of the (endogenous) variables of the system in (1). We have chosen to eliminate EF2; this was done chiefly because data on this variable were very difficult to obtain with any reasonable degree of reliability At any rate, stock flotation as a source of external finance, while not negligible, is of minor significance compared with bond flotation for most years and industries in our analysis. New bond floatations exceeded new equity floatations by a 7:1 margin over the 1950–2002 period (Guerard & Schwartz, 2007). Since our sample will consist largely of manufacturing and retail trade firms with a rather small representation of mining firms, it follows that our selection of bond finance as the principal source of external funds to be studied is not likely to lead to serious deficiencies. Finally, we have chosen to regard short-term investment as a predetermined variable, so that our final system of equations to be estimated is reduced to three, viz., the dividend, the (fixed) investment, and the external finance equations. This last decision carries with it, in principle, serious deficiencies. There is unfortunately no detailed breakdown of the short-term investment series available, so we have chosen to treat it as a predetermined variable. Thus, the model finally estimated is of the form D = g1 (I, EF1; X1 , X2 , . . . ,Xn ,) I = g2 (D, EF1;X1 , X2 , . . . ,Xn ,) EF1 = g3 (D, I; X1 , X2 , . . . ,Xn ,)
(4)
General Comments on the Form of the Equations The Dividend Equation. One can look upon dividend disbursals as conveying information to the market on the inherent profitability of the disbursing firm as Modigliani and Miller (1961) inter alios have argued. In fact, they would contend that the dividend series contains “more information” than the profit series. Hence, it would appear that it is the policy of firms to maintain a steady dividend per share and to adjust it, upward and downward, only when a “permanent” change in their economic environment has taken place. As a matter of fact, it is more or less common for firms to maintain a constant dividend per share. But this in no way implies constancy in the dividend-profit ratio. It is reasonable to suppose that dividend disbursals will depend on the rate of profit of the firm, its investment plans, and the external finance obtained through the bond market; the rationale for this last variable would be that external finance will enable the firm to carry out its planned dividend disbursals even when the rate of profit is low and investment programs are extensive. The Investment Equation. The foundations of investment theory in the theory of the firm are too well-known to require repetition here. Clearly from this point of view investment would depend either on changes in the volume of output or on its rate of profit, which may be taken to lead to changes in the expected profitability
42
J.B. Guerard
of new investment. These two considerations are not totally unrelated, especially if the firm is assumed to operate with a neoclassical production function allowing substitution; if no substitution is allowed, then it is not clear that the rate of profit has any place in the investment function. Our innovation here consists in introducing the other two jointly dependent variables, dividends and external finance. We have already given some indication as to why we consider these variables relevant. Clearly, dividend disbursals and investment outlays represent competing demands on the resources available to the firm. Thus it would be quite plausible to suppose that the investment activities of the firm will be affected by its dividend activities. Postponement or curtailment of investment could conceivably result because of inability of the firm to carry out a given investment program, “optimally” determined by some “rational” criteria, and at the same time continue to make “satisfactory” dividend payments. It would also be of interest to inquire whether such variables as depreciation (see Meyer and Kuh (1957)) are significant determinants of investment; if depreciation is an accurate index of deterioration of the capital stock due to its employment in the productive process, then depreciation would accurately describe that part of investment undertaken for replacement purposes. There are good reasons to believe, however, that depreciation does not accurately measure the using up of capital, and hence its introduction in the investment equation would only serve to portray more accurately the resources available to the firm for investment and dividend outlays. In addition, there is the question of the proper lags operating in the investment process; thus it would be of interest to ascertain whether lagged rates of change of sales or past rates of profit significantly affect the decision to invest. The introduction of the bond finance variable here has a motivation best understood in terms of imperfect capital markets. Thus, if in a given universe all firms belong to more or less the same uncertainty class, then market discrimination might be expected to take the form of restricting the amount a firm can borrow without raising the cost of obtaining long-term funds. Hence, we may conjecture, ceteris paribus, that the easier the access to this market—either in the amounts or in the terms on which the loans are granted—the larger the investment program a firm may undertake. Thus, in the investment equation dividends may be expected to have a negative impact, while external finance will have a positive one. The External Finance Equation. Enough has been said in connection with the other two equations to make the hypothesized form of the external finance equation clear. One would expect to have this variable depend positively on investment, negatively on the market interest rate, and negatively on depreciation and profits. The relationship of external finance to dividends, however, is not very clear-cut. Thus, it is possible to argue that because of a budgetary constraint, more dividends, other things being equal, mean more borrowing. But it is equally plausible to argue that for firms that are no longer growing rapidly, more dividends need not induce further borrowing simply because their investment activities are somewhat restricted. Thus, there should not be any feedback from dividends to external finance.
The Corporate Sector as a Net Exporter of Funds
43
Empirical Estimation of the Equation System The sample on which our study is based consists of firms with assets exceeding $200 million in 2002 for which a continuous satisfactory record exists between 1952 and 2002. These firms are largely manufacturing and retail trade ones, although several are chiefly engaged in mining activities. The sources of our data are the balance sheets and income statements of individual firms appearing in the Compustat database. Our sample does not weigh very heavily any particular classification, although the firms in this study tend to be mostly medium- and large-sized ones. The sample begins with 156 companies in 1952 and concludes with 3762 companies in 2002. Our sample in 1952 of 156 firms is very similar to the Dhrymes and Kurz (1967) sample of 181 companies for the same year. The econometric results are summarized in Dhrymes (1974). Our study is the largest (in scope) and longest (in time period) analysis of the interdependencies of financial decisions. At this stage it is desirable to catalog and explain briefly the nature of the variables entering into our investigation. The following basic variables are employed: St = Sales at time t, undeflated (EF1)t = long-term borrowing—external finance—at time t; EF1 is the first difference of the book value of long-term debt outstanding and thus represents net current long-term borrowing; it should be remembered that this measure is somewhat biased by the transfer of maturing long-term debt to the short-term category Dt = dividends (common) paid at time t It = gross fixed investment at time t Kt = book value of the capital stock at beginning of time t Pt = net profits (after taxes) at time t, undeflated (LTD)t = net long-term debt outstanding at time t, in nominal terms (Dep)t = depreciation allowances at time t Nt = net current position of the firm at time t, defined as the excess of inventories, cash, short-term securities, and accounts receivable over accounts payable and other short term liabilities Rt = interest payments at time t, on long-term debt outstanding. This is admittedly a very poor measure of the relevant interest rate, but it is the only one available In carrying out the empirical implementation of the model, we have chosen to normalize the jointly dependent variables by St . This was done for two reasons: first, it tends to reduce heteroscedasticity and hence make the stochastic characteristics of our sample correspond more closely to the standard specification of simultaneous equation models; second, since our objective is to isolate the determinants of the investment, dividend, and external financing decisions process, this procedure prevents our results from being unduly influenced by large firms simply because of sheer size. Another related reason is the fact that one would not expect the
44
J.B. Guerard
relation between investment and the appropriate accelerator variable to be identical in the case of a retail store and an aircraft manufacturer. By relying on “intensive” variables, one tends to overcome such problems. A list of the predetermined variables employed is given below: N/K enters the model as a consequence of the use of the budget constraint to eliminate one of the equations of the system; the normalization employed here is to some extent motivated by portfolio theory considerations. Dep/K represents the portion of the book value of the capital stock written off as depreciation charges. It would have been better perhaps to have defined the numerator of this fraction as profits plus depreciation plus interest charges on the ground that, since it measures the (average) rate of return on the firm’s capital resources, this ought to be measured gross of irrelevant bookkeeping items such as depreciation and interest charges. P/K is the rate of profit. S2 ∗ = (St -St-3 )/ (St-3 ) is the usual accelerator variable except that it is normalized by a lagged value of sales. It was felt, however, that it is the pressure of sustained relative increases in sales that affects investment. LTD/(K - LTD) is the leverage variable employed to test the principle of increasing risk. It is probably not a very accurate one, the rationale behind it being that businessmen are influenced by book rather than market value considerations.
Main Empirical Results We present annual cross-section regression for selected years, one per decade, over the 1952–2002 period. We chose to report non-recessionary years, 1955, 1968, 1978, 1986, and 1998, as defined by the Conference Board (Zarnowitz, 1992 and 2001; Montgomery, Zarnowitz, Tsay, & Tiao, 1998; and Zarnowitz & Ozyildirim, 2001).6 We will send interested readers the all-regression results. Let us address here the ordinary least squares regression (OLS) results and the two-stage least squares (2SLS) regression results for the dividend, investment, and external financing equations. The regression results are shown in Tables 5, 6 and 7 for the respective equations. The primary, statistically significant determinants of dividends, shown in Table 5, are profits (positive), net current position (negative), and investments (negative). The external financing variable is statistically significant in about half of the annual dividend equations. The primary, statistically significant determinants of dividends, in 2SLS shown in Table 5, are, again, profits (positive), net current position (negative), and investments (negative). The OLS and 2SLS dividend coefficients and their corresponding t-statistics are virtually identical. The external financing variable is incorrectly negative and statistically significant in many of the annual dividend equations.
The Corporate Sector as a Net Exporter of Funds
45
Table 5 Dividends selected years 1955
1968
1978
1986
1998
2006
0.036 (4.18) 0.042 (3.08) −0.026 (−2.11) −0.059 (−3.09) 0.069 (1.46) 0.10 5.6
0.027 (5.69) 0.073 (6.71) −0.045 (−6.65) −0.090 (−5.62) 0.120 (7.19) 0.25 33.4
0.018 (6.74) 0.039 (6.00) −0.019 (−5.16) −0.041 (−4.92) 0.010 (3.35) 0.08 14.5
0.012 (9.82) 0.017 (3.66) −0.002 (−1.85) −0.030 (−.75) −0.009 (−2.61) 0.02 5.3
0.026 (6.34) 0.001 (0.13) −0.019 (−3.19) −0.054 (−2.92) 0.009 (3.43) 0.01 6.4
0.01 (10.17) 0.007 (2.55) −0.002 (−1.76) −0.014 (−1.72) 0.001 (0.09) 0.00 2.7
0.037 (3.35) 0.043 (2.90) −0.027 (−1.89) −0.059 (−3.02) 0.057 (0.46) 0.09 5.1
0.024 (3.86) 0.074 (6.72) −0.043 (−6.68) −0.091 (−5.63) 0.158 (3.28) 0.19 22.9
0.008 (1.7) 0.033 (3.80) −0.004 (−.59) 0.001 (0.03) −0.050 (−2.31) 0.05 8.2
0.013 (9.65) 0.017 (3.64) −0.004 (−2.42) 0.000 (−.06) −0.029 (−2.54) 0.02 5.0
0.016 (10.49) 0.003 (1.43) −0.004 (−1.79) −0.020 (−2.20) −0.028 (−1.82) 1.00 4.5
0.01 (4.81) 0.012 (1.81) −0.003 (−1.42) −0.034 (−1.52) −0.090 (1.25) 0.00 1.1
OLS Variable intercept t PK CAK IK FS Adj.R–Square F 2SLS Variable intercept t PK CAK IK FS Adj.R–Square F
The statistically significant investment determinants are the net current position (negative) and new debt issues (positive). The lagged profits, sales accelerator variable, and dividend variables are not statistically significant in the investments equation, shown in Table 6. The statistically significant 2SLS investment determinants are the net current position (negative) and new debt issues (positive). The lagged profits variable is not consistently positive and statistically significant in the investments equation, nor is the sales accelerator variable. Indeed, the sales accelerator variable is more often negative and statistically significant rather than the presumed positive coefficient. The dividend variable is often negative and statistically significant in the investments equation, shown in Table 6. The OLS external financing determinants are capital investments. The debtto-equity ratio, depreciation, profits, and interest payments are not statistically significant in the external financing equation. The reader is referred to Table 7 for the external financing results. Moreover, the dividend variable is not consistently positive or negative in sign in its coefficient. The 2SLS external financing determinants are capital investments. The debt-to-equity ratio, depreciation, profits, and
46
J.B. Guerard Table 6 Investment selected years 1955
1968
1978
1986
1998
2006
0.162 (10.67) −0.021 (−.72) −0.221 (−9.61) 0.078 (−3.09) −0.039 (−.25) 0.226 (2.13) 0.46 29.6
0.172 (12.6) 0.011 (0.28) −0.255 (−11.12) 0.006 (−5.62) −0.112 (−.63) 0.334 (5.10) 0.33 38.5
0.178 (13.0) 0.046 (1.24) −0.247 (−11.74) −0.009 (−4.92) 0.178 (0.65) 0.141 (7.27) 0.24 37.2
0.082 (10.4) 0.033 (1.84) −0.006 (−1.35) −0.003 (−.75) 0.235 (1.11) 0.254 (12.15) 0.16 30.7
0.145 (15.64) −0.010 (−1.32) −0.096 (−5.71) −0.063 (−2.92) −0.094 (−1.25) 0.110 (12.91) 0.11 44.2
0.074 (10.17) −0.002 (−.39) −0.012 (−2.93) 0.001 (−1.72) 0.105 (1.01) 0.019 (3.44) 0.01 4.3
0.49 (2.18) 0.179 (1.04) −0.513 (−2.25) 0.560 (1.46) −0.876 (−.52) −0.550 (−1.27) 0.03 2.0
0.157 (3.92) 0.212 (2.45) −0.262 (−4.69) −0.265 (−3.06) −3.941 (−4.52) 2.138 (4.03) 0.15 14.9
0.269 (5.28) 0.155 (1.15) −0.358 (−5.84) −0.359 (−3.63) −0.375 (−.12) 1.274 (4.62) 0.07 10.0
0.122 (2.54) 0.286 (3.57) −0.004 (−3.31) −0.096 (−2.19) −8.179 (−2.40) 1076.000 (4.17) 0.02 4.1
0.218 (2.69) 0.022 (0.64) −0.228 (−2.73) −0.076 (−0.71) −10.485 (−3.48) 1.195 (5.1) 0.02 7.0
0.126 (0.84) −0.019 (−.3) −0.065 (−1.19) 0.014 (0.35) −7.946 (−.57) 2.800 (2.12) 0.00 1.0
OLS Variable intercept t PKL CAK D2SALES DS FS Adj.R–Square F 2SLS Variable intercept t PKL CAK D2SALES DS FS Adj.R-Square F
interest payments are not statistically significant in the external financing equation (see Table 7). As was the case in the OLS external financing estimated equation, the dividend variable is not consistently positive or negative in sign in its coefficient. The primary difference in the OLS and 2SLS regression results is the negative, and statistically significant, interdependency between investments and dividends. Investments and dividends are alternative uses of funds.
Conclusion Schwartz and Aronson were correct that the corporate sector was, and is, a net exporter of funds. Moreover, as we address the issue of why firms issue debt, we must determine the structure underlying the dividend-investment-external
The Corporate Sector as a Net Exporter of Funds
47
Table 7 External financing selected years 1955
1968
1978
1986
1998
2006
0.007 (0.69) 0.000 (−.18) 0.092 (−9.61) −0.006 (−.60) 0.055 (2.31) 0.187 (1.43) 0.236 (4.74) 0.14 5.7
0.042 (3.81) 0.000 (−.51) −0.001 (−11.12) −0.145 (−1.22) −0.045 (−1.40) 0.812 (6.08) 0.199 (5.37) 0.19 15.6
−0.056 (−1.80) 0.000 (0.81) −0.004 (−11.74) −0.590 (−1.68) 0.185 (1.92) 0.185 (1.92) 0.490 (5.83) 0.06 7.0
−0.018 (−1.30) −0.001 (−.98) 0.000 (−1.35) −0.018 (−.35) 0.048 (1.20) −0.934 (−2.75) 0.637 (12.28) 0.16 26.9
0.073 (2.88) −0.001 (2.54) 0.000 (−5.71) −0.454 (−1.92) −0.125 (−2.29) 0.623 (3.09) 0.790 (12.8) 0.01 34.2
−0.001 (−.70) 0.000 (0.10) −0.001 (−2.93) −0.093 (−1.04) 0.009 (0.16) 0.269 (0.65) 0.301 (3.4) 0.00 2.3
−0.029 (−1.28) 0.000 (−.09) 0.399 (1.99) 0.063 (0.46) 0.090 (0.60) 1.218 (2.34) 0.171 (2.54) 0.06 2.9
0.043 (3.51) 0.212 (−.55) −0.003 (−.06) −0.133 (−.94) −0.048 (−1.29) 0.818 (2.72) 0.188 (3.60) 0.07 5.8
0.061 (1.26) 0.155 (−.24) 0.007 (0.90) −1.367 (−2.59) 0.561 (3.71) −11.674 (−4.68) 0.570 (3.16) 0.04 4.9
0.040 (1.98) 0.286 (−.93) 0.003 (0.87) −0.052 (−.84) 0.105 (1.88) 6.957 (−2.62) 0.413 (3.08) 0.02 3.8
0.063 (0.98) 0.022 (0.53) 0.000 (−.14) −0.309 (−1.18) −0.163 (−2.67) 4.340 (2.22) 0.566 (5.09) 0.03 8.9
0.126 (0.75) −0.019 (−.11) −0.001 (−.25) −0.177 (1.17) 0.072 (0.68) −6.129 (−.71) 0.221 (1.46) 0 1.0
OLS Variable intercept t DE INTE DEPK PK DS IS Adj.R–Square F 2SLS Variable intercept t DE INTE DEPK PK DS IS Adj.R-Square F
finance triad of decision-making processes. Moreover, we were concerned with demonstrating the simultaneous character of these decision-making processes. The sample employed was a cross-sectional one involving manufacturing, mining, and retail trade firms with assets exceeding $200 million in 2002, with data during the period 1952–2002. The method of investigation consisted of estimating the structure of our model successively for the years 1952–2002. The main findings are in brief the following:
48
J.B. Guerard
1. There is a significant degree of interdependence between the investment and dividend decision-making processes, with the implication that if dividend policies are very rigid as some allege, then this rigidity may tend to hamper the investment activity of firms. On the other hand, our results show that the investment requirements of firms tend to have a significant effect on their dividend behavior. 2. The external finance activity of firms seems to be strongly affected by their investment policies, but less so by their dividend policies.
Notes 1. We examine the 1971–2006 period because Compustat does not maintain debt and equity issuance and repurchases prior to 1971. 2. Stock repurchases rose substantially following the crash of October 1987. 3. Lakonishok and Vermalen reported premiums of 21.79, 24.09, and 18.54% on tender offers during the 1962–1986, 1962–1979, and 1980–1986 periods, respectively. They also reported cumulative abnormal returns of 12.54, 14.58, and 9.78% to non-tendering stockholders during the corresponding periods. Smaller firms produced the highest abnormal returns. 4. If the market re-evaluates a stock favorably, assigning it plus factors for future growth, increased dividends and buybacks and greater stability all at once, the stock can show a rapid rise in price in a brief span of time. However, the market for common stock is notoriously volatile, and a wave of optimism as to the future of the economy and the share of corporate earnings can send the general level of stock prices to unsustainably high levels. Widespread pessimism can have the opposite effect. Three constraining factors prevent the growth period from running for an indefinite length. First, it is apparent that no option can keep generating a return that grows forever at a rate that exceeds the growth rate of the economy. A super-growth enterprise that grows at a rate exceeding that of the GDP becomes a larger and larger part of the GDP; it becomes a bigger and bigger proportion of the total economy. Taking this proposition to the logical extreme, when the super-growth enterprise becomes the total GDP, it cannot grow faster than the GDP. More realistically, we may note that as the enterprise becomes a larger part of the economy, its rate of growth is constrained by the limits of the system in which it operates. The second limiting factor involves the relation of the level of the interest rate to available super-growth projects. Theoretically a super-growth investment of infinite duration has an infinite value. However, if there were a prevalence of super-growth options, the demand for capital would take a quantum leap. But savings, the supply of capital, is not infinitely elastic. Presumably in any society people must consume some resources, eat, clothe themselves, and find shelter. They cannot save all income. The price for savings, i.e. the interest rate, must rise, and equilibrium is reached when the interest rate becomes higher than the rate of growth. Last, we may take a look at the actual financial side of a super-growth process. Presumably, a super growth situation requires an increase in net capital per period which is in excess of the investment in the previous period and which requires that in each period the return on the total capital continues at a rate in excess of the cost of capital. Although, given some initial superior advantage in technology or product, an enterprise might enjoy super-growth for a considerable number of years; such growth is not likely to go on indefinitely on an increasing capital base. When in the particular situation the enterprise begins to evidence a declining marginal return to capital, the period of super-growth must begin to slow and eventually cease. 5. In this connection it seems appropriate to cite, in some detail, a very important study by Kuh (1963), which is, in some respects, similar to the one we propose to pursue here. Kuh
The Corporate Sector as a Net Exporter of Funds
49
investigates the investment, dividend, and external finance aspects of the firm’s behavior in the following context. His basic sample consists of sixty industrial firms for which a continuous satisfactory record exists over the years 1935–55. This sample was arrived at from a larger one by a process of selection which eliminated firms that were merged into others over the sample period, as well as those that were “too large”—owning gross assets over $120 million in 1953. The work is divided into two distinct parts, a theoretical and an empirical one. In the theoretical part the interdependence of the three decisions is quite clearly recognized and an integrated model is presented combining the capacity-accelerator model of investment with the Lintner (1956) and Fama and Babiak (1968) models of dividend behavior. There, investment is given by the usual capacity-accelerator model, so that it depends on the capital stock and sales, as well as on the observed sales-capital ratio, the latter being an approximation to the desired output-capital ratio of the accelerator theory. In this connection models are also tried in which the sales variable is replaced by a profit variable. It is found, however, that the former models are more in accord with the data than the profit ones. They are thus made to depend on profits and past dividends, the model being essentially an adaptation of the usual flexible accelerator model of investment; here the role of the capital coefficient is played by the desired dividend-payout ratio. Finally, external finance behavior is derived residually through the budgetary requirement that investment expenditures must equal retained earnings plus depreciation allowances plus external finance. Thus, in this context there is a certain direction of causality: investment is independent of dividends and external finance and dividends depend only on profits (and lagged dividends) which may depend on investment although the dependence is not explicit. 6. We report 2002 despite its recessionary environment because the 2000–2005 period does not include a non-recessionary year.
References Bierman, Jr., H. (2001). Increasing Shareholder Value: Distribution Policy, a Corporate Finance Challenge, Boston: Kluwer Academic Publishers. Bierman Jr., H. (2001). “Valuation of Stocks with Extraordinary Growth Prospects.” Journal of Investing, 10, 23–26. Brealey, R., Myers, S.C., & Allen, F. (2005). Principles of Corporate Finance, New York: McGraw-Hill/Irwin. Dann, L. (1981). “Common Stock Repurchases: An Analysis of Returns to Bondholders and Stockholders,” Journal of Financial Economics, 9, 115–38. Dhrymes, P.J. (1974). Econometrics: Statistical Foundations and Applications, New York: Springer-Ver1ag. Dhrymes, P.J., & Kurz, M. (1967). “Investment, Dividends, and External Finance Behavior of Firms,” in Robert Ferber (Ed.), Determinants of Investment Behavior, New York: Columbia University Press. Fama, E.F. (1974). “The Empirical Relationship Between the Dividend and Investment Decisions of Firms,” American Economic Review, 63, 304–18. Fama, E.F., & Babiak, H. (1968). “Dividend Policy: An Empirical Analysis,” Journal of the American Statististical Association, 63, 1132–61. Grabowski, H.G., & Mueller, D.C. (1972). “Managerial and Stockholder Welfare Models of Firm Expenditures,” Review of Economics and Statistics, 54, 9–24. Guerard, Jr., J.B., Bean, A.S., & Andrews, S. (1987). “R&D Management and Corporate Financial Policy,”Management Science, 33, 1419–27. Guerard, Jr., J.B., & McCabe, G.M. (1992). “The Integration of Research and Development Management into the Firm Decision Process,” in D.F. Kocaoglu (Ed.), The Management of R&D and Engineering, Amsterdam: North-Holland. Guerard, Jr., J.B., & Schwartz, E. (2007). Quantitative Corporate Finance, New York: Springer.
50
J.B. Guerard
Higgins, R.C. (1972). “The Corporate Dividend-Saving Decision,” Journal of Financial and Quantitative Analysis, 7, 1527–41. Jalilvand, A., & Harris, R.S. (1984). “Corporate Behavior in Adjusting to Capital Structure and Dividend Targets: An Econometric Study,” Journal of Finance, 39, 127–45. Kuh, E. (1963). Capital Stock Growth: A Micro-Econometric Approach, Amsterdam: NorthHolland. Lakonishok, J., & Vermaelen, T. (1990). “Anomalous Price Behavior Around Repurchase Tender Offers,” Journal of Finance, 45, 455–77. Lintner, J. (1956). “Distributions of Incomes of Corporations Among Dividends, Retained Earnings and Taxes,” American Economic Review, 46, 97–118. Meyer, J.R., & Kuh, E. (1957), The Investment Decision, Cambridge: Harvard University Press. McCabe, G.M. (1979). “The Empirical Relationship Between Investment and Financing: A New Look,” Journal of Financial and Quantitative Analysis, 14, 119–35. McDonald, J.G., Jacquillat, B., & Nussenbaum, M. (1975). “Dividend, Investment, and Financial Decisions: Empirical Evidence on French Firms,” Journal of Financial and Quantitative Analysis, 10, 741–55. Miller, M., & Modigliani, F. (1961). “Dividend Policy, Growth, and the Valuation of Shares,” Journal of Business, 34, 411–33. Montgomery, A.L., Zarnowitz, V., Tsay, R.S., & Tiao, G.C. (1998). “Forecasting the U.S. Unemployment Rate,” Journal of the American Statistical Association, 93, 478–93. Mueller, D.C. (1967). “The Firm Decision Process: An Econometric Investigation,” Quarterly Journal of Economics, 81, 58–87. Peterson, P., & Benesh, G. (1983). “A Reexamination of the Empirical Relationship Between Investment and Financing Decisions,” Journal of Financial and Quantitative Analysis, 18, 439–54. Schwartz, E., & Aronson, J.R. (1966). “The Corporate Sector: A Net Exporter of Funds” Southern Economic Journal, 33, 252–57. Schwartz, E., & Aronson, J.R. (1967). “The Corporate Sector: A Net Exporter of Funds: Reply” Southern Economic Journal, 34, 153–54. Switzer, L. (1984). “The Determinants of Industrial R&D: A Funds Flow Simultaneous Equation Approach,” Review of Economics and Statistics, 66, 163–68. Tinbergen, J. (1938). “Statistical Evidence on the Accelerator Principle,” Econometrica, 10, 164–176. Tinbergen, J. (1938). Statistical Testing of Business Cycles, Geneva: The League of Nations. Zarnowitz, V. (2001). “The Old and the New in the U.S. Economic Expansion,” The Conference Board. EPWP #01–01. Zarnowitz, V. (1992). Business Cycles: Theory, History, Indicators, and Forecasting, Chicago: University of Chicago Press. Zarnowitz, V., & Ozyildirim, A. (2001). “On the Measurement of Business Cycles and Growth Cycles,” Indian Economic Review, 36, 34–54.
Piscal Folicy 101—Economic Policy Meets Partisan Politics John Hilley
The world of political Washington is leavened by a wonderful troupe called the Capitol Steps, an ensemble whose mission is to poke fun at our political follies. Their signature rap is called “Lirty Dies,” open-ended verse in which the first letters of words are strategically reversed, causing outbursts of laughter and insight. Our friend Eli Schwartz has always been that inquisitive guy with the twinkle in his eye. He has a deep interest in the world of economic policy, the subject of this brief essay. In Capitol-Steps speak, he can flo with the go, deploying his mealthy hind in areas such as fublic penance and beveraged lie-outs. In the spirit of smiling while soldiering on, I recount and draw lessons from three episodes of economic policy-making in the world of partisan politics—Piscal Folicy 101. The first was the battle waged over President George H.W. Bush’s desire to reduce the capital gains tax rate—a battle that ended in a defeat for the president. The second episode is again one of failure, Bill Clinton’s 1993 attempt to pass a stimulus bill in advance of his major deficit reduction effort. The last example is one of success, a rare political moment when Republicans and Democrats joined together to balance the federal budget while enacting major policies in the national interest. In my conclusion I address opportunities lost as well as some policy challenges awaiting the next set of national leaders.
The Battle for a Capital Gains Rate Reduction Winning the election in 1988, George H.W. Bush was caught betwixt and between on fiscal policy. As the inheritor of the supply-side faith, the president was eager to reduce taxes. But after a decade of large budget deficits, politics and economics were strongly pointed toward deficit reduction. Mainstream economists were lined up behind fiscal prudence, seeing benefit in lower interest rates, more private sector investment, a less overvalued dollar, and a more balanced sharing of government’s costs between current and future generations. For their part, congressional Democrats and Republicans had spent the decade of the 1980s trying to stem the deficit hemorrhage opened by the 1981 Reagan tax cuts. J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_5, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
51
52
J. Hilley
A reduction in the capital gains rate had been the long-sought Holy Grail of the Republicans. It was an integral part of the supply-side faith, as well as a powerful means to reward its upper-income constituency. As part of President Reagan’s 1981 tax cuts, the capital gains rate was lowered from 28 to 20%, leaving a substantial difference between the capital gains rate and the top rate on ordinary income of 50%. But that differential was eliminated in 1986 as part of a bipartisan tax reform effort that lowered income tax rates in exchange for closing “tax loopholes.” A central part of the compromise was to do away with the capital gains preference, leaving ordinary and capital gains to be taxed at the same rate. President George H.W. Bush reopened the capital gains issue in 1989, advocating a 15% rate. The problem, of course, was how to enact the tax cut while paying homage to the goal of deficit reduction. Having defeated Democratic contender Michael Dukakis largely on the basis of his “no new taxes” pledge, President Bush needed a way to advance his proposal without an offsetting tax increase. His finesse was a “self-financed” capital gains proposal. The required holding period for capital gains treatment would be one year initially, and then rise to 2 years before settling at three. It was a clever inducement for asset holders to realize capital gains and pay taxes—not only to take advantage of the new lower rate, but to beat the longer holding period requirements on the horizon. If enacted, the policy would have flooded the Treasury with revenue in the early years, causing a major budgetary drain only later. This proposal was clever in another respect. In early 1989, President Bush reached out to Democrats, seeking a bipartisan solution to the deficit problem. By mid-April, bipartisan negotiations yielded an agreement to reduce the deficit by just over $20 billion, including a $5.3 billion tax increase. And lo and behold, the Bush administration had a ready-made way to raise that much revenue—via its proffered capital gains rate reduction. If enacted, the proposal would have given President Bush the trifecta—a cut in the capital gains rate, a reduction in the near-term deficit, and fealty to his “no new taxes” pledge. Through some adept efforts and good fortune, the Bush administration was partially granted its capital gains wish by the Democratic-controlled House.1 Things were different in the Senate where Democratic leader George J. Mitchell and a majority of Democrats were determined to defeat the proposal. To them, the capital gains provision made the deficit problem worse in the long run and was badly skewed to favor the wealthy. The battle came to a head on the Senate floor where Republican leader Bob Dole claimed to have a majority of votes for a Republican proposal to lower the capital gains rate. But the proposal increased the deficit beyond the 5-year budget window, giving leader Mitchell and the Democrats a potential way to defeat the proposal. Under the Senate’s so-called Byrd rule, a legislative provision that increased the deficit beyond the 5-year budget window was considered extraneous and would require 60 votes to pass. Even at the higher threshold, the vote would have been very close. Instead of directly taking on the capital gains issue, Democrats smartly launched a flanking move, a high-road attack designed to kill the capital gains rate reduction by calling for a clean deficit reduction bill, stripped of all extraneous and special
Piscal Folicy 101
53
interest provisions. The call to purity hit a resonant chord with Democrats and many Republicans and on October 13, 1989, a clean reconciliation bill was passed in the Senate by 87—7. The conference with the House was gridlocked for several weeks, but Mitchell and the Senate Democrats held firm. In early November, Republican leaders informed President Bush that it was time to get on with the nation’s business and abandon capital gains for the year. But the defeat in 1989 did nothing to lessen the ardor of the Bush administration for a lower capital gains rate. In 1990, it launched a reconfigured “self-financed” capital gains proposal. The capital gains issue was in play throughout the year as the 1990 bipartisan Budget Summit went through its various ups and downs. Mitchell and other Democrats made it clear that a capital gains cut would need to be offset by an increase in the top income tax rate—to at least 33%. At the end of the day, the Republicans would only agree to increase the top rate to 31%—and the Democrats would only reduce the capital gains rate to 28%. The mere three percent differential favoring capital gains was widely interpreted as a defeat for President Bush—compared to his proposal for a 19.6% rate. Despite the perceived defeat of his capital gains proposal as part of the 1990 agreement, President George H.W. Bush provided the political leadership that allowed the passage of that major bipartisan deficit reduction package. In an act of responsibility and political courage, he saw the $500 billion deficit reduction agreement to its conclusion, despite the defection of House Republicans over the tax increases contained in the bipartisan compromise—which made up thirty percent of the budget savings.2
The Storal of this Mory? Things often turn out far differently than anticipated. The quest for a deficitincreasing capital gains rate reduction propelled the largest deficit reduction package in U.S. history—one that contained significant tax increases even as the president’s capital gains proposal was substantially defeated.
The Clinton 1993 Stimulus Bill Bill Clinton unseated incumbent George H.W. Bush in 1992 by crystallizing public sentiment against the Bush administration’s stewardship of the economy. With unemployment at 7.5%, Americans responded to the message of “It’s the Economy, Stupid,” by electing the first democratic president since 1976. When President Clinton came to office in January 1993, the fiscal outlook was bleak. The Congressional Budget Office was projecting annual deficits that would average more than $300 billion a year for years to come. It was a stunning amount of red ink that the new President inherited—a projected $1.8 trillion over 6 years.3 He did not shrink from the challenge, vowing to cut the deficit in half in his first term.
54
J. Hilley
Despite his high-minded desire—and ultimate success—in dealing with the nation’s fiscal problems, President Clinton’s first foray into fiscal policy was a disaster.4 The problem confronting the new administration was that while deficit reduction was the correct long-run fiscal policy, the nation was just emerging from recession. The fear was that the contractionary effects of deficit reduction might put the country back into recession. The administration’s solution was born of textbook economic advice that was oblivious to political reality. The Clinton administration proposed a two-pronged approach: first invigorate the economy with fiscal stimulus and then follow with a major long-term deficit reduction package. What this theoretically sound plan ignored was just about everything important: the difficulty of legislating two major fiscal policy bills in 1 year; the miniscule economic effect of the stimulus package versus its large political cost; and the eagerness of the Republican minority to assert itself, demonstrating that it could be a political force even if the Democrats controlled both ends of Pennsylvania Avenue. The Clinton stimulus bill proposed to add $16 billion of federal spending to the economy. Given the speed with which the legislation would need to be passed, that spending would have to be added to existing programs. There was too little time to invent targeted programs for those impacted by the recession. But, of course, spending money is in Congress’s wheelhouse. The speed at which the appropriations committees can spend $16 billion dollars is a paragon to government efficiency, even if the uses to which it is put are not. The stimulus bill was manna from heaven for Senate Republicans. If deficit reduction was the overriding issue—one the Clinton White House had campaigned on and was planning furiously—here was a chance to demonstrate fiscal responsibility by defeating the Democratic spending initiative. Senate Republicans were greatly aided by the specifics of the bill; the appropriations committee had funded well-known programs such as community block grants that gave recipients wide leeway to allocate funds. While numbers and statistics are important to decisions in Washington, the anecdote is king and “few things are harder to put up with than the annoyance of a good example.”5 Republican Senators turned their staffs loose, easily digging up example after example of wasteful “pork-barrel” spending, even if these had previously been approved in an accommodating bipartisan manner. Righteous indignation flowed across the floor and into the press galleries. How could the Democratic majority justify spending for midnight basketball, or swimming pools, and on and on? The Republicans had more than enough ammunition to justify their filibuster of the bill, proudly claiming a win for responsible government as the bill was taken down.
The Storal of this Mory? Ivory tower plus Green Acres equals political disaster. Politics delimits the economic policies that can be enacted.
Piscal Folicy 101
55
The Balanced Budget Agreement of 1997 In 1997, responsible leaders reached across party lines to balance the federal budget while enacting significant policies in the national interest.6 That rare political moment was made possible by a number of factors. Thanks to the politically perilous but productive deficit reduction efforts of Presidents Bush in 1990 and Clinton in 1993, as well as a robust economy that was filling the Treasury with revenues, the long-sought goal of a balanced budget was in reach.7 As important, the political stars were aligned. Each party had taken its turn at deficit reduction while also trying to advance its political goals. But both the Democratic package of 1993 and the Republican Contract with America in 1995 were rewarded with political defeat.8 What both parties learned the hard way was that behind the laudable goal of balancing the budget were a bunch of tough policy choices guaranteed to make many important constituencies unhappy. Both Democrats and Republicans realized that they had severely depleted their political capital trying to turn a difficult problem to partisan advantage. Bipartisanship, they decided, was a better way. We were also fortunate to have seasoned and responsible leaders on both sides of the aisle.9 Although they were tough partisans, each understood that an important national objective could be achieved by working together. None of the central players was so rigid that he could not negotiate in good faith and remain open to the viewpoint of the other side. These were also professionals who knew how to drive the legislative process. They understood the political and personal leanings of their colleagues and the policies they would and would not support. That knowledge and their leadership skills provided the foundation for an agreement that commanded the overwhelming support of all four congressional caucuses. Most important, these partisans gave themselves the political room to find the middle. They pledged that all the policies had to be within a balanced budget box. This self-imposed discipline forced them to make better choices, throwing out policies on the extremes and those which were budget busters. They treated policy differences as founded in principle; they kept their disagreements out of the glare of press and politics; and always kept the lines of communication open. Committed to the common enterprise, they came to trust each other and act as teammates, understanding that winning was finding common ground, not imposing losses on the other side.
The Storal of this Mory? It is possible for responsible leaders to deal with national problems in a constructive bipartisan manner.
Opportunities Lost and the Road Ahead As a result of the Balanced Budget Agreement of 1997, the federal budget went into surplus the very next fiscal year and was followed by three more years of growing
56
J. Hilley
budget surpluses. At that time, the hope and expectation of responsible leaders in both parties was that these budget surpluses would allow them to deal with the enormous fiscal challenges facing the nation, largely driven by a demographic wave that would soon be putting enormous stress on the economy and many government programs, particularly Social Security and Medicare. Unfortunately, that was not the path taken by our elected leaders. In 2000, the electorate chose George W. Bush as president, giving Republicans control of all branches of government. The surpluses that were bequeathed by responsible leaders of both parties have been overwhelmed by a tide of red ink—projected to be over $2 trillion during President George W. Bush’s two terms in office.10 The restraints placed on federal spending as part of the 1997 budget agreement expired in fiscal year 2002 and in the subsequent 5 years, federal spending increased by 35% in nominal terms. At the same time, taxes were reduced in multiple rounds of legislation. And during this period no significant actions were taken to address the fiscal challenges brought by the impending retirement of the baby boomers. In sum, America has treated itself to nearly a decade of fiscal irresponsibility. The reason, of course, is politics. Two decades of budget politics have taught our elected leaders that stepping up to raise taxes or cut spending is the quick route to retirement. Rather than making policy and budgetary choices within the discipline of a balanced budget, our country and our leaders have opted for the path of least political resistance. They found it is easier to put the bill on the tab and push the costs into the future. The next best chance to take on the nation’s major problems is 2009. No matter which party prevails in the 2008 elections, we will have an opportunity to reassess the challenges we face as a nation and a government. Elections offer new beginnings and this one is very much needed. Here is a short list of major policy challenges that the new president will face: • • • • •
Restoring fiscal responsibility. Assuring the long-term solvency of the Social Security system. Reforming the health care system. Strengthening national security. Reforming the tax system.
Social Security Over the 5-year budget window from 2010 to 2014, Social Security surpluses will average about $250 billion annually. However, those large surpluses that are part of the unified budget are temporary. In 2017, the Social Security outlays will begin to exceed Social Security revenues. Social Security surpluses will decline slowly as earnings on the trust fund’s Treasury securities provide a temporary cushion. By 2040, however, the trust fund will be exhausted. At that point and if no changes are
Piscal Folicy 101
57
made to current spending and revenue streams, the system will be able to pay only 74% of promised benefits. There are three compelling reasons to tackle the Social Security issue. First, this isn’t rocket science. This is mainly a financing issue—a politically-charged financing issue to be sure—but not a problem caused by major structural flaws as is the case with our nation’s health care system. For Social Security, the cure is mostly a matter of holding hands and making reasonable adjustments to a hugely successful program. Second, shoring up Social Security would alleviate financial pressure on the rest of the budget for the long run. The smaller the deficits (or the larger the surpluses) in Social Security, the less has to be made up out of the rest of the budget to achieve overall balance. Third, the demographic wave is upon us, and delay simply means making larger and more painful adjustments later.
Health Care The need for health care reform is compelling. Unlike Social Security, however, this is fundamentally a policy problem and secondarily a financing one. As a nation, we spend about $2 trillion annually on health care. But the system is under stress from numerous directions: lack of coverage; the cost of health care; dependence on a fractured mix of employer, government, and individual financing; and perverse incentives with their resulting inefficiencies and waste. The policy problem is so imposing that policy makers could benefit by agreeing to a workable budget box for the reform effort. I suggest that in 2009, the president and Congress agree to a 5-year budget path for the federal government’s health care spending and revenue streams— consistent with overall fiscal responsibility. That path could be above or below baseline amounts. It would become the basis for the health care reform budget box—a box requiring that health care reform take place on a deficit neutral basis relative to those amounts. Federal health spending could be raised or reduced— likewise for federal revenue to support the health care system. But these changes in federal health care spending or revenues would have to fit in the agreedupon box. There are three good reasons to take this approach. First, it would discipline decisions, forcing them to be made in a flexible but fiscally responsible way. Second, the politics of tax and spend has been a major factor behind our nation’s inability to address many of its problems. If increased revenues were to be a part of health care reform, there would be a clearer link between costs and benefits, increasing the likelihood of public support. Third, a budget box would force us to confront the key financing issue: the distribution of health care responsibilities among individuals, employers, governments, and charities. It seems to me that this approach strikes the right balance between flexibility and discipline. It allows all reform ideas to be put on the table but serves notice that any changes in federal responsibility must be deficit neutral.
58
J. Hilley
National Security Our nation remains unprepared to propel the peace as well as to secure ourselves against multiple threats—man-made and not. Congress and the Executive branch have the institutional ability to make the needed changes provided the President and Congress act constructively together. Those choices would be better made within a long-term fiscal framework that enabled needed reforms—in the division, coordination, and oversight of national security responsibilities; in our strategic posture; in our diplomatic, humanitarian, and development efforts; in our procurement policies; and in making our homeland truly secure. Similar to the budget box for health care reform, improvements in our national security could be aided by quickly adopting a long-term defense spending path within which trade-offs and choices would be made.
Tax Reform As for taxes, there is room to work, thanks to provisions that will soon expire. These include tax rates that apply to ordinary income, as well as provisions affecting capital gains, dividends, and estates, the child credit, and amounts paid by joint filers. The pending expiration of these provisions puts a lot of money in play. Offsetting this is the impact of the alternative minimum tax (AMT) on upper- and middle-income taxpayers. The AMT was designed to limit deductions by high-income taxpayers—such as for state and local taxes, real property taxes, personal exemptions and standard deductions, and several other important tax preferences. Taxpayers are required to calculate income tax liabilities under both the regular income tax and the AMT and then pay the higher amount. But the AMT is not indexed for inflation and its bite is being felt further down the income ladder, effectively increasing tax payments for large numbers of taxpayers. As a result of the expiring provisions, several key revenue policies are on the table, including tax rates, the kinds of deductions allowed, the viability of a parallel tax system in the form of the AMT, and key provisions affecting capital gains and dividends. There is an obvious and compelling need to straighten this all out, reforming and simplifying the tax system while keeping an eye on the bottom line of what it takes to pay for what we say we want. In 2009 and beyond, we have the opportunity to solve a number of important national challenges. There are sound policies to address these concerns; the difficult issue will be creating the needed political shelter. Partisans can work together across party lines if they endorse overarching national goals and commit to bipartisan solutions. Those efforts could be aided by using fiscal discipline as a tool to direct and discipline the effort. Most importantly, our leaders need to agree to create the political room to lock arms and give each other the political cover to act responsibly. Our storal mory should be to take that opportunity.
Piscal Folicy 101
59
Notes 1. Six conservative Democrats joined Republicans in rolling the Ways and Means Committee Chairman, Dan Rostenkowski, passing a capital gains provision that lowered the rate to 19.6% through 1991. After that, the rate would go to 28%, but the value of assets would be indexed to inflation. Like President Bush’s proposal, it would induce behavior that would make the tax cut self-financing in the short term. Although the House Democratic leaders opposed the provision, it sailed through the House, awaiting a conference with the Senate. 2. The deficit reduction package enacted under President George H.W. Bush had projected 5-year deficit savings of $496.2 billion, of which $146.3 billion was in the form of increased revenues. See 1990 Congressional Quarterly Almanac, p. 112. 3. The projected deficits faced by President Clinton in January 1993 were based on Congressional Budget Office estimates, The Economic and Budget Outlook: Fiscal Years 1994 to 1998, p. 28, Table 2–1. 4. The American electorate has the unfortunate habit of electing governors to be president. The reasons are obvious. The governor has been a chief executive, seemingly qualified to run the country. In addition, elected representatives serving in Washington will have been subjected to a large number of votes on issues of national importance, many of which are ripe for partisan plucking. Governors are isolated from these forced revelations of policy positions and also from the scrutiny of a sophisticated press corps. Combine that with a fundraising base leveraged off the governorship, the ability to run as an outsider, and highly effective campaign managers for hire, governors have a decided advantage over those with Washington experience. But the common result of electing governors is to subject the nation to neophytes who must quickly learn on the job. The smart ones like Bill Clinton overcome their inexperience. Others don’t do so well. 5. As Samuel Clemens so aptly put it. 6. The full story is told in my book, The Challenge of Legislation: Bipartisanship in a Partisan World, Brookings Institution Press, 2007. 7. The 5-year savings in the 1990 and 1993 budget agreements were the two largest deficit reduction efforts in U.S. history. The 1993 Democrats alone-deficit reduction package had projected 5-year savings of $496 billion, of which $240 billion was from increased revenues. See 1993 Congressional Quarterly Almanac, p. 108. 8. The Contract with America laid out a sweeping, but fairly specific, agenda including a balanced budget, tax cuts, welfare reform, crime prevention, policies for families and children, national security, Social Security, regulatory relief, and congressional reform and term limits. See Congressional Quarterly Almanac 1995, p. 1–10. 9. On the Republican side, House Speaker Newt Gingrich and Senate Republican leader Trent Lott were instrumental to this success. On the Democratic side, President Clinton’s efforts were greatly aided by Senate Democratic leader Tom Daschle. 10. Congressional Budget Office, The Budget and Economic Outlook: An Update, August 2007, p. 6, Table 1–3.
A Taxonomy of Utility Functions Benjamin J. Gillen and Harry M. Markowitz
Introduction Modern financial theory began with the publication of two articles in 1952: Markowitz (1952a) and Roy (1952). The principal difference between the Roy article and the Markowitz article is that Roy recommended a specific portfolio from the mean-variance frontier, namely, the one which maximizes R=
E−d σ
where E and σ are portfolio expected return and standard deviation, and d is a “disaster” level. Markowitz, instead, presented the investor with a mean-variance frontier and allowed the investor to choose from this. These proposals for choosing efficient portfolios from a mean-variance frontier are now frequently contrasted with behavioral finance proposals which assume that the investor acts with various sorts of irrationality. The behavioral finance literature is typically dated from prospect theory as published by Kahneman and Tversky (1979). While Kahneman and Tversky are considered the “fathers” of behavioral finance, a recent three-volume handbook on the subject by Shefrin (2001) cites Roy (1952) and Markowitz (1952b) as essentially the “grandfathers” of behavioral finance. The present paper analyzes a class of functions which generalize the utility-ofwealth function in Markowitz (1952b). The latter function is drawn with “customary wealth” at the origin. Customary wealth equals current wealth except when the economic agent has experienced a recent windfall gain or loss. The utility function is concave to the left of the origin and convex to the right.1 Thus, if there has been no recent windfall gain or loss, the agent will buy insurance to protect against downside loss, and also buy lottery tickets, chancing a small loss in hope of a large gain. However, the function becomes concave as gain increases further; in fact, it is bounded above to avoid the St. Petersburg Paradox. Conversely, to the left (as ever The authors are honored to be invited to submit an article to this volume for Eli Schwartz, and are delighted to do so. J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_6, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
61
62
B.J. Gillen and H.M. Markowitz
greater losses are contemplated) the function becomes convex and, in fact, bounded below. We will refer to a utility function with the properties postulated in the Markowitz (1952b) “Utility of Wealth” article as a UW function. We introduce here a generally broader class of functions which we refer to as GUW (generalized UW) functions. Like the UW functions, the GUW functions are measured from some “reference” wealth level, e.g., the customary wealth of the UW functions. Also like the UW functions, all GUW functions are bounded above and below. By definition, any GUW function reaches its asymptote at some finite (but perhaps extremely large) gain or loss. This may or may not happen with a UW function. However, the chief difference between the UW and GUW functions is that, whereas a UW function has three inflection points (one at customary wealth, one below and one above this), a GUW function may have any odd number of inflection points. These inflection points may be distributed arbitrarily with respect to reference wealth. As with UW functions, GUW functions are drawn with reference wealth at the origin O, i.e., at X = O. The qualitative properties of a utility function U in GUW depend on the number and placement of its inflection points. For example, a function with one inflection point placed to the left of the origin O is “locally concave” at O. An extreme case is one in which the inflection point is close to the point at which U touches its lower bound. An agent with such a U-function would be risk-averse unless the agent is so far below the reference wealth that only an act of desperation has a chance (albeit small) of saving it. If the single inflection point is at O, then the utility function is essentially that proposed by Kahneman and Tversky. If the inflection point is to the right of O, the agent is locally a gambler. In addition to inflection points (where U" changes sign), a U in GUW may have one or more linear regions with U = 0. For example, if the linear region includes O, then the agent is locally risk-neutral at O. If U is linear to the right of O and concave to the left, then the agent is risk-neutral with respect to gains and riskaverse with respect to losses. In general, U" may or may not have a different sign to the left and right of a linear region. Since the utility functions in various subsets of the class GUW have such different behavior implications, this class is not proposed as a hypothesis about action, but as a way of classifying utility functions depending on the number and placement of their inflection points. It is analogous to the classification of auto-regressive models as AR (1), AR (2), etc. The next section reviews the UW class of functions and their relation to the Friedman-Savage (1948) hypothesis. Following sections define the class, provide notations for specifying subclasses, and explore the characteristics of some of these.
The Friedman and Savage Hypothesis Friedman and Savage (1948) seek to present a utility function consistent with the existence of both gambling and the buying of insurance. To do so they propose the function in Fig. 1. The X-axis represents the wealth of the decision maker; the Y-axis the utility of this wealth level. The Friedman-Savage (FS) function has two
A Taxonomy of Utility Functions
63
Fig. 1 Friedman-Savage Utility Function
U
A
B
W
inflection points—at A and B. The FS curve is concave below A, convex between A and B, and concave above B. Markowitz (1952b) presented the following objections to the Friedman-Savage utility function. Draw a line tangent to the two “humps” of the function, namely, tangent to the points C and D as in Fig. 2. Point C is a low level of wealth, below which economic agents are considered poor. Point D is a high level of wealth, above which agents are considered rich. Consider two people whose wealth lies halfway between these two levels: W = (C + D)/2. There is no fair bet which two such individuals would prefer to one that would send one to the poor level C and the
U
Fig. 2 Markowitz (1952b) Construction on the Friedman-Savage Function
C
D
W
64
B.J. Gillen and H.M. Markowitz
other to the rich level D. Such a bet would be preferred by each of the individuals to staying at (C + D)/2 with certainty. But one rarely if ever finds “middle income” individuals taking such bets. Further complaints about the FS utility function are that below C the individual does not gamble at fair odds. In particular, the individual would not buy a lottery ticket even if the lottery sponsor made no money. But casual observations suggest that poor people are large buyers of lottery tickets even when the lottery sponsor makes a profit. Also, if a person has wealth just short of the “rich” level D, then there is no fair bet that the individual would prefer to one which, if the economic agent wins, will bring it up to D, and if lost will plunge it down to C. Again, this is not generally observed. The only level of wealth which seemed plausible to Markowitz (1952b) was the inflection point at A. A person at that wealth level would buy insurance against a loss, yet spend a small amount for a small chance of a large gain and a large chance of a small loss. Markowitz hypothesized that this inflection point was at or near the investor’s current wealth. He further hypothesized that, in the case of windfall gains or losses, the investor could move temporarily from this inflection point. The inflection point thus represented what Markowitz called “customary wealth” rather than current wealth necessarily. Specifically, if the individual had a recent windfall gain, he or she would move into the convex portion to the right of the inflection point and become a bit more devil-may-care in choosing among probability distributions of wealth; if they had a windfall loss they would move temporarily into the concave portion and become more cautious. While the utility function postulated by Markowitz (1952b) was concave to the left and convex to the right in the neighborhood of customary wealth, it was further postulated that it would become concave and, in fact, bounded-above to the right, and convex and bounded-below on the left. The reason for these assumptions was the St. Petersburg Paradox on the right and a similar argument on the left.
Relationship between UW, GUW and Behavioral Finance Shefrin (2001) considers Roy (1952) and Markowitz (1952b) to be essentially the grandfathers of behavioral finance. Specifically, concerning the Markowitz (1952b) article Shefrin says the following: Markowitz (1952) truly qualifies as a behavioral work, with its focus on how people actually behave. Markowitz addresses a classic question posed by Friedman and Savage (1948): why do people simultaneously purchase insurance and lottery tickets? Friedman and Savage proposed a solution to this question, a solution that Markowitz criticized on behavioral grounds. In arguing against the Friedman-Savage solution, Markowitz described the results of how people behave, citing an experiment conducted by Mosteller and Nogee (1951) about how people bet. . . . Looked at in hindsight, Markowitz showed amazing insight. . . . His discussion about the difference between present wealth and customary wealth gave rise to the coding of gains and losses relative to a reference point. He recognized that losses loom larger than gains. He proposed a utility function with three inflection points to capture the idea that attitude or risk varied with the situation being faced. In this respect he emphasized the importance of whether a gamble is framed in terms of gains or losses, as well as whether
A Taxonomy of Utility Functions
65
the stakes are small or large. His discussion touches on aspiration points, the preference for positive skewness, and a property Thaler and Johnson (1991) subsequently called the “house money effect.”
Concerning the move by Markowitz (1952b) towards behavioral finance, a crucial step is that of going from measuring wealth along the X-axis to that of measuring deviations from some reference wealth along this axis. The next section defines a class (GUW) of utility functions which also measure deviations from a reference wealth on the X-axis. In the case of UW, reference wealth is customary wealth. We can imagine situations in which reference wealth has a different meaning. Consider the case of a rogue trader whose account is millions (or billions) of dollars behind. It might be most natural and convenient to use getting back to “break-even” as the reference wealth.
The Generalized UW Class of Functions We consider a class (GUW) of functions that are not only bounded on the left and the right as in Markowitz (1952b), but that eventually become flat, i.e., there exists levels c and d such that c < 0 and U (X) = U (c) for all X ≤ c
(1)
d > 0 and U (X) = U (d) for all X ≥ d.
(2)
In other words, above some very large wealth one can assume that it is of no further value to have another billion dollars; and below some negative wealth there is no way to pay off the debt anyway, so being another billion dollars poorer is of no consequence. As with the utility-of-wealth (UW) class of functions, for a (GUW) function X represents deviations from some reference value, but not necessarily with the same interpretation as with UW in Markowitz (1952b), as noted in the preceding section. We further assume U (X) > 0 for all X (c, d)
(3)
U exists everywhere in (c, d) and changes signs only finitely often.
(4)
Since U is the integral of U , it is continuous. Thus (3) implies that, in any closed interval [e, f ] with c < e < f < d U (X) achieves a minimum greater than zero on the interval; i.e., there is an X0 [e, f ] such that 0 < U (X0 ) ≤ U (x) for all X [e, f ]
66
B.J. Gillen and H.M. Markowitz
Rather than deciding that two particular values on the X-axis have zero and unit utility (as we may without loss of generality), we will assume that reference wealth, X = 0, has U (0) = 0, and
(5)
U (0) = 1.0.
(6)
As we move either to the right or left of X = 0, U is the integral of U , and U the integral of U : U (X) = 1 +
x
U (t)dt
0
x U (X) =
U (t)dt
0
where X may be negative. Thus, starting with the values of U and U at X = 0, the values of U and U can be determined from U (X) elsewhere on the line until X = c and X = d are reached.
Subsets of GUW Inevitably GUW functions have a convex portion to the left and a concave portion to the right. Thus, the sign of U is eventually positive on the left and negative on the right. From left to right, in Markowitz (1952b) the sign of U is first positive, then negative, then then positive, and finally negative. Thus every UW function changes signs three times. GUW functions are only required to start with U > 0 and end with U < 0. This could happen with one change of sign and, in general, any odd number of sign changes. GUW functions may also have segments in which U = 0. U may or may not have a different sign to the left and the right of a segment with U = 0. The properties of GUW utility functions depend on the number of their convex (U" > 0), concave (U" < 0 ) and linear (U" = 0 ) regions, and how these are placed with respect to reference wealth (X = O). We adopt the following notation as abbreviations for the regions of a GUW function: + convex region U > 0, − concave region U < 0, 00 linear region U = 0 R reference wealth X = 0
A Taxonomy of Utility Functions
67
For example + − R denotes a function with one inflection point of the left of reference wealth; whereas + 0R0 − has a linear region to left and right of reference wealth. This section considers various subsets with “signatures” such as the above. (+ - R) Investors with this U-function are locally risk averse. If the one inflection point is far to the left, the economic agent will only gamble if desperate. (+ R -) This is essentially the Kahneman and Tversky utility function, except that theirs has a kink at X = 0. A GUW function cannot have a kink in (c, d) since U is continuous. However, a kink can be approximated by having U change (continuously but) very rapidly in the neighborhood of X = 0. Such a “prospect theory” utility function does not explain the buying of insurance by an agent at X = 0. In case of insurance, all outcomes (i.e., incurring an uninsured loss, paying an insurance premium, or not paying or losing anything) are in a convex region. Therefore, the risk is preferred to the certainty: the insurance is not bought. In the case of the lottery ticket, this may or may not be bought, since the region of the small loss (buy the ticket and not win) is in a convex region, while the win is in a concave region. However, because of the kink (almost) at X = 0, presumably the lottery ticket (like the insurance) is not bought. ( + 0 R 0 - ) This is a case with only one change of sign, with a linear segment that includes An investor with this utility function is locally risk-neutral. The extreme case has the linear region from almost c to almost d. The investor then is risk-neutral from almost the “desperation point” to almost the “satiation point.” ( + - R 0 0 -) This is also a case with only one change in sign, but with a linear segment immediately to the right of X = 0. Levy and Markowitz (1979) show empirically that the behavior of the (+ – R) investor is approximately meanvariance efficient if probability distributions are not spread out “too much.” A similar analysis would presumably show that the (+ – R00–) investor would be locally mean-semivariance efficient. (See Markowitz, 1959, Chapters 9 and 13; Journal of Investing, 1994; and Sortino and Satchell, 2001). ( + - R + - ) This is the signature of the UW investor.
General Considerations A possible objection to any utility function that is asymptotically horizontal to the left is that the fair bet which an economic agent would find most attractive is one which is increasingly large (without bound) to the left. But if this bet is lost, there
68
B.J. Gillen and H.M. Markowitz
is no way that the individual can pay it, so that no informed counter-party will ever accept this bet. In the case of a rogue-trader, the counter-party is not informed. Under certain circumstances it is plausible for utility functions to suddenly change signature. If some miraculous good luck saves a hopelessly behind roguetrader with a (+ R + -) or (+ 0 R 0 -) utility function, he might promise some saint (or some AA club for gamblers) that his utility function will henceforth be (+ - R -) with the inflection point far to the left.
Summary In this paper we have considered a class (GUW) of utility functions that generalize the Markowitz (1952b) “Utility of Wealth” function. The GUW class includes functions that permit the simultaneous purchase of gambling and insurance, cautious utility functions which diversify for wealth levels within a large area of current wealth, utility functions that are risk-neutral or risk-seeking near current wealth (when the latter is the reference wealth), or are approximately the Kahneman and Tversky utility function with the kink smoothed. Unlike the UW (utility of wealth) functions, the class is too broad to have interesting, verifiable implications. Rather, various subclasses have such implications. We have presented notation to specify various subclasses and have explored the properties of some of these.
Notes 1. Markowitz (1952b) refers to convex line segments as concave and concave segments as convex. This was so because Friedman and Savage (1948) refer to convex segments as concave from above (and sometimes as just concave) and conversely for concave segments.
References Friedman, M., & Savage, L.P. (1948). “The Utility Analysis of Choices Involving Risk,” Journal of Political Economy, 56, 279–304. Journal of Investing (1994), Vol. 3, (3), Fall. Kahneman, D., & Tversky, A. (1979). “Prospect Theory: An Analysis of Decision Under Risk,” Econometrica, 47(2), 263–91. Levy, H., & Markowitz, H.M. (1979). “Approximating Expected Utility by a Function of Mean and Variance,” American Economic Review, 69(3), 308–17. Markowitz, H.M. (1952a). “Portfolio Selection,” The Journal of Finance, 7(1), 77–91. Markowitz, H.M. (1952b), “The Utility of Wealth,” The Journal of Political Economy, 60, 152–58. Markowitz, H.M. (1959), “Portfolio Selection: Efficient Diversification of Investments,” New Haven: Wiley, Yale University Press, 1970, 2nd ed., Basil Blackwell, 1991. Roy, A.D. (1952). “Safety First and the Holding of Assets,” Econometrica, 20, 431–49.
A Taxonomy of Utility Functions
69
Shefrin, H. (2001). (Ed.), Behavioral Finance, Vol. 3, Northhampton, MA: Edward Elgar Publishing Co. Sortino, F. & Satchell, S. (2001). Managing Downside Risk in Financial Markets: Theory, Practice and Implementation. Burlington, MA: Butterworth-Heinemann.
Equilibrium and Disequilibrium Growth: a Comment on a Comment Robert Solow
An old cartoon by James Thurber shows one of his overpowering women, evidently an actress, sitting down next to one of his mousy men, and saying: “Now let’s talk about you. What did you think of my performance?” With that lesson in mind, my favorite way to pay tribute to an old friend like Eli Schwartz on occasions like this is to reread one of his works and try to hold up my end of an imaginary conversation. What came readily to hand was his 1993 book Theory and Application of the Interest Rate. The tone of this book is distinctive. The mainstream theoretical core is always evident, but the text is full of practical advice and numerical examples. I will try to keep with that way of proceeding. The part of the book where I thought I might have something to add is the final Chapter 10 on “Growth, Profits and the Interest Rate.” (From now on I will refer to the author, rather artificially, as “ES.” The natural “Eli” is too informal for print, and the conventional “Schwartz” is impossibly formal.) There are a few places where I would be inclined to amend ES’s discussion. To begin with, he says pithily, without much in the way of reference: “Thus, in the pure models, the rate of return on capital is equal to the interest rate; the rate of growth is equal to the rate of savings relative to the capital stock; the amount of savings equal to investment is equal to property income; thus, by definition, the growth rate is equal to the rate of return on the capital stock, or finally the growth rate, assuming neutral technological progress, is equal to the interest rate.” I think this takes too seriously the special case in which all property income is saved and all other (i.e., wage) income is consumed. I don’t think this assumption was ever thought to represent the “pure” or general case. It is an extreme, and possibly illuminating, special case, and it is the so-called “golden rule” that leads to the highest feasible level of steady-state consumption per person in a growing economy. But it is not a good vehicle for more general theorizing. If all wages are consumed and a fraction s of property income is saved, then ES’s final statement is modified to say that the growth rate is s times the interest rate. And, by the way, it is very important to realize that this equation is to be understood as determining the (steady-state equilibrium) interest rate from the growth rate and s, not the other way around. It is only fair to say that in the next paragraph ES remarks that the “savings equals property income” assumption, with possible Marxian overtones, has little to recommend it. In fact, the literature treats it as a pedagogical special case.
J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_7, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
71
72
R. Solow
Later on in the chapter, in empirical comments that I will come to soon, ES takes pains to point out that actual experience strongly rejects this special case. In fact, property income consistently exceeds investment. It should: the essence of the “golden rule” is the implication that then a higher saving-investment rate would lead eventually to higher sustainable consumption per head. (And, in the opposite inefficient case, if saving exceeded property income, a lower sustained saving rate would increase steady-state consumption per head.) This is fundamentally a trivial matter, and a digression. ES’s real interest lies elsewhere. Investment is driven by economic profit (or quasi-rent), the excess (or deficiency) of a firm’s net income over (or below) the cost of capital. So how can it be that, in a growing economy, always in equilibrium, the rate of return on capital should just equal the interest rate? The latter can be taken as the cost of capital (with proper adjustment for depreciation, etc.); but then there is no room for any positive profit. What motivates the investment that is needed to keep the economy growing? There is an easy answer to this question in pure theory, and ES had more or less given it just a page earlier in a brief discussion of the classical stationary state. There no net saving is occurring, and one has to wonder why the equilibrium rate of interest should not be zero (except for possible risk premia to safeguard principal). ES goes on at that point: “The only reason for the existence of any net (i.e., positive) interest in such circumstances would rest on the existence of a strong time preference, such that if there were no interest, some of the present stock of capital would be consumed.” The same sort of reasoning applies in a growing steady state. If investment were to fall short of what is required to keep the return on capital just equal to the interest rate, the capital-labor ratio would fall, the rate of return on capital would rise a little, some positive gap would open up, and additional investment would be stimulated until the pure profit disappeared. If investment were to exceed the steady-state requirement, the same process would occur with all the signs changed. The practical side of ES would notice at once that this is a little too good to be true. Everyday dynamics would see to it that the observed course of events would exhibit at least small fluctuations around the steady state, with the return on capital sometimes above and sometimes below the rate of interest. One force that might keep the deviations small is that some firms would be experiencing positive quasi-rents while other firms are facing negative quasi-rents. This is roughly where ES’s thoughts go, though he doesn’t put them in exactly that way. A deeper macroeconomic question, far beyond the scope of ES’s book, is whether there are forces in a modern capitalist economy that tend to convert these inevitable small deviations occasionally or perhaps regularly into larger self-reinforcing fluctuations around the steady-state growth path. Schumpeter gave one answer to that question, not much accepted these days. Axel Leijonhufvud sketched a quite different version in his notion of a “corridor,” within which the growth equilibrium is stable but outside of which all bets are off; this idea also failed to attract attention. (One reason may be that a positive theory of macroeconomic policy might be a necessary part of such an investigation.)
Equilibrium and Disequilibrium Growth
73
ES is interested in profits and interest, not in trends and business cycles. So he proceeds to focus on those temporary deviations around a path of (otherwise) equilibrium growth. The result is interesting and, I think, valid. Growth theorists, including me, working mainly though not exclusively with continuous-time models, have worked out the basic mechanics of an economy whose growth is driven by technological progress. Always, however, it has been “disembodied” technological progress; no innovation-specific investment is required to make new technology effective. My own attempt, 50 years ago, to model growth as requiring the “embodiment” of new technology in tangible investment petered out for lack of either empirical confirmation or good follow-up ideas. ES, without thinking in those terms, arrives at roughly the same place. By thinking in discrete-time terms and imagining technological progress as a series of discrete innovations, he represents each innovating firm as needing also to make a standard choice-of-capital-intensity investment decision. It will invest to the point where the marginal return on further investment will equal the cost of capital. There is no reason why the usual concavity conditions should not apply, so the supra-marginal bits of investment will earn more than the cost of capital. These quasi-rents will eventually be eroded by competition with newer, more productive, technology; but there would always be a moving float of quasi-rents over and above interest earnings. ES doesn’t think about this whittling away of profits, but taking account of it would not undermine his point. (Schumpeter did, of course, focus famously on the erosion of profits—“creative destruction”—but in his story profits are a temporary monopoly return, whereas ES’s just reflect the normal excess of average over marginal returns to investment.) This is a useful idea. It could use more working-out than ES can give it in a couple of pages. From the growth-theoretic standpoint it introduces a long-run difference between a model in which technological progress is disembodied and one in which it is embodied in innovation-specific capital investment. Previously there seemed to be only short-run differences, and that is one reason why the disembodied case, though it violates common sense, seemed to be an adequate vehicle for theory. Notice that the point here does not turn on whether technological change is exogenous, but only on whether it has this disembodied or “atmospheric” character. But the more immediate implication, and the one that interests ES, is primarily empirical. The fact that the aggregate return on capital, as measured by corporate profits, say, is consistently larger than implicit interest on the stock of corporate capital is sometimes taken as at least mildly paradoxical. ES cites James Tobin as fretting that corporations seem to be able to earn 10% a year after tax even while families are prepared to save and acquire assets that pay much less. Doesn’t that suggest that the socially desirable rate of investment must be much larger than the current rate? Tobin apparently attributed a substantial part of the gap to monopolistic distortions. ES first calls attention to the important, though unmeasured, role of quasi-rents in guiding the level and direction of business investment. He also endorses, as anyone would, the Knightian view that (entrepreneurial) profits are the reward for the residual bearing true uncertainty. Tobin would have accepted the general point, but
74
R. Solow
would have pointed out that “social” uncertainty is probably much less than private uncertainty, because much of the latter is zero-sum, losses here being the mirror image of profits there. “Society” is the all-inclusive diversifier. In considering the relevant empirical magnitudes, ES tends to revert to the national income and product accounts. He insists correctly that it is really the corporate sector that we should be looking at in this connection. One could go further and take the non-financial corporate sector as the appropriate part of the economy. There is a convenient table that appears monthly in Economic Indicators and annually in the Economic Report of the President and decomposes the gross value added of non-financial corporate business (about half of the GDP). In a year like 2000, net interest (and miscellaneous payments) amounted to 3.6% of gross value added. (If all taxes, including corporate income taxes, are deducted from gross value added, net interest amounts to 4.1% of what remains.) Keep in mind that this is net interest as a fraction of value added, not as a ratio to capital. In the same year, profits after tax were 5.8% of value added. (These figures are cyclically unstable: by 2006, after-tax profits had risen to 7.8% of gross value added, while net interest had fallen all the way to 1.9%; clearly long-term bond rates had not fallen by that much, but value added per unit of capital had risen during the business-cycle upswing.) The fact that these figures are all given per unit of value added is very inconvenient for making and judging comparisons; but at least we know that in a normal year profits after tax can be half again as large as net interest paid by the nonfinancial corporate sector. That ratio would remain valid if both interest and profits were referred to the same stock of capital. If we identify after-tax profits with ES’s quasi-rents, the interesting question is whether that is a large or a small discrepancy. We do not have to choose among the possible accounts of the origin of economic profits: Schumpeterian monopoly, good old monopoly, Knightian uncertainty-bearing, and the supra-marginal returns on investment emphasized by ES. All could be playing a role, and probably are. At a purely terminological level, I would be inclined to regard those quasi-rents tied to capital-embodied technological change as equilibrium rather than disequilibrium occurrences. What happens in the aggregate is of course the sum of microeconomic events. If there is a reasonably steady flow of innovations, some of them no doubt more capital-intensive than others, there is likely to be a continuing and perhaps fluctuating flow of net quasi-rents. (I say “net” because there will be some negative quasi-rents associated with obsolescence.) For the individual firms involved, there will be disappointed or unusually favored expectations; and the process may be felt as disequilibrium. For the observing macroeconomist, however, the whole process can be thought of as a sort of stochastic equilibrium. I mentioned at the beginning that the 1993 book represents a distinctive way of doing economics, with easy transitions among theory, facts, real and made-up examples. It strikes me as a good way to do economics, and it is certainly characteristic of Eli Schwartz. I wish it were characteristic of more economists.
Reputational Risk and Conflicts of Interest in Banking and Finance: The Evidence So Far Ingo Walter
Financial services comprise an array of “special” businesses. They are special because they deal mainly with other people’s money, and because problems that arise in financial intermediation can trigger serious external costs. In recent years the role of various types of financial intermediaries has evolved dramatically. Capital markets and institutional asset managers have taken intermediation share from banks. Insurance activities conducted in the capital markets compete with classic reinsurance functions. Fiduciary activities for institutional and retail clients are conducted by banks, broker-dealers, life insurers and independent fund management companies. Intermediaries in each cohort compete as vigorously with their traditional rivals as with players in other cohorts, competition that has been intensified by deregulation and rapid innovation in financial products and processes. Market developments have periodically overtaken regulatory capabilities intended to promote stability and fairness as well as efficiency and innovation. It is unsurprising that these conditions would give rise to significant reputational risk exposure for the financial firms involved. For their part, investors in banks and other financial intermediaries are sensitive to the going-concern value of the firms they own, and hence to the governance processes that are supposed to work in their interests. Regulators in turn are sensitive to the safety, soundness and integrity of the financial system, and from time to time will recalibrate the rules of the game. Market discipline, operating through the governance process, interacts with the regulatory process in ways that involve both costs and benefits to market participants and are reflected in the value of their business franchises. The first section of this paper defines reputational risk and outlines the sources of reputational risk facing financial services firms. The second section considers the key sources of reputational risk in the presence of transactions costs and imperfect information.1 The next section surveys available empirical research on the impact of reputational losses imposed on financial intermediaries, including the separation of reputational losses from accounting losses. The fourth section builds a link between exploitation of conflicts of interest and reputational risk. The fifth section considers managerial requisites for dealing with both reputational risk and conflicts of interest. The last section concludes.
J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_8, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
75
76
I. Walter
What Is Reputational Risk? Reputational risk in banking and financial services is associated with the possibility of loss in the going-concern value of the financial intermediary—the risk-adjusted value of expected future earnings. Reputational losses may be reflected in reduced operating revenues as clients and trading counterparties shift to competitors, increased compliance and other costs required to deal with the reputational problem—including opportunity costs—an increased firm-specific risk perceived by the market. Reputational risk is often linked to operational risk, although there are important distinctions between the two. According to Basle II, operational risks are associated with people (internal fraud, clients, products and business practices, employment practices and workplace safety), internal processes and systems, and external events (external fraud, damage or loss of assets, and force majeure). Operational risk is specifically not considered to include strategic and business risk, credit risk, market risk or systemic risk, or reputational risk.2 If reputational risk is bracketed-out of operational risk from a regulatory perspective, then what is it? A possible working definition is as follows: Reputational risk comprises the risk of loss in the value of a firm’s business franchise that extends beyond event-related accounting losses and is reflected in a decline in its share performance metrics. Reputation-related losses reflect reduced expected revenues and/or higher financing and contracting costs. Reputational risk in turn is related to the strategic positioning and execution of the firm, conflicts of interest exploitation, individual professional conduct, compliance and incentive systems, leadership and the prevailing corporate culture. Reputational risk is usually the consequence of management processes rather than discrete events, and therefore requires risk control approaches that differ materially from operational risk.
According to this definition, a reputation-sensitive event might occur which triggers an identifiable monetary decline in the market value of the firm. After subtracting from this market capitalization loss the present value of direct and allocated costs such as fines and penalties and settlements under civil litigation, the balance can be ascribed to the impact on the firm’s reputation. Firms that promote themselves as reputational standard-setters will, accordingly, tend to suffer larger reputational losses than firms that have taken a lower profile—that is, reputational losses associated with identical events according to this definition may be highly idiosyncratic to the individual firm. In terms of the overall hierarchy of risks faced by financial intermediaries, reputational risk is perhaps the most intractable. In terms of Exhibit 1, market risk is usually considered the most tractable, with adequate time-series and cross-sectional data availability, appropriate metrics to assess volatility and correlations, and the ability to apply techniques such as value at risk (VaR) and risk-adjusted return on capital (RAROC). Credit risk is arguably less tractable, given that many credits are on the books of financial intermediaries at historical values. The analysis of credit events in a portfolio context falls short of market risk, although many types of credits have over the years become “marketized” through securitization structures such as asset-backed securities (ABS) and collateralized loan obligations (CLOs) as well
Reputational Risk and Conflicts of Interest
77
Exhibit 1 A hierarchy of risks confronting financial intermediaries
as derivatives such as credit default swaps (CDs). These are priced in both primary and secondary markets, and transfer some of the granularity and tractability found in market risk to the credit domain. Liquidity risk, on the other hand, has both pluses and minuses in terms of tractability—in continuous markets liquidity risk can be calibrated in terms of bid-offer spreads, although in times of severe market stress and flights to quality, liquidity can disappear. If the top three risk domains in Exhibit 1 show a relatively high degree of manageability, the bottom three are less so. Operational risk is a composite of highly manageable risks with a robust basis for suitable risk metrics together with risks that represent catastrophes and extreme values—tail events that are difficult to model and in some cases have never actually been observed. Here management is forced to rely on either simulations or external data to try to assess the probabilities and potential losses. Meanwhile, sovereign risk assessment basically involves applied political economy and relies on imprecise techniques such as stylized facts analysis, so that the track record of even the most sophisticated analytical approaches is not particularly strong—especially under conditions of macro-stress and contagion. As in the case of credit risk, sovereign risk can be calibrated when sovereign foreigncurrency bonds and sovereign default swaps (stripped of non-sovereign attributes like external guarantees and collateral) are traded in the market. This leaves reputational risk as perhaps the least tractable of all—with poor data, limited usable metrics, and strong “fat tail” characteristics.
78
I. Walter
The other point brought out in Exhibit 1 relates to the linkages between the various risk-domains. Even the most straightforward of these—such as between market risk and credit risk—are not easy to model or to value, particularly in a bidirectional form. There are 36 such linkages, exhibiting a broad range of tractability. We would contend that the linkages which relate to reputational risk are among the most difficult to assess and to manage.
Sources of Reputational Risk Where does reputational risk in financial intermediation originate? We argue that it emanates in large part from the intersection between the financial firm and the competitive environment, on the one hand, and the direct and indirect network of controls and behavioral expectations within which the firm operates on the other, as depicted generically in Exhibit 2.3 The franchise value of a financial institution as a going concern is calibrated against these two sets of benchmarks. One of them, market performance, tends to be relatively transparent and easy to reward or punish. The other—performance against corporate conduct benchmarks—is far more opaque but potentially critical as a source of risk to shareholders.
Exhibit 2 Performance gaps, competition and conflict
Management must work to optimize against both sets of benchmarks. If it strays too far in the direction of meeting the demands of social and regulatory controls, it runs the risk of poor performance in the market, punishment by shareholders, and possibly a change in corporate control. If it strays toward unrestrained market performance and sails too close to the wind in terms of questionable market conduct, its behavior may have disastrous results for the firm, its managers and its shareholders. In the end, striking this balance is a key corporate governance issue.
Reputational Risk and Conflicts of Interest
79
Such are the rules of the game, and financial intermediaries have to live with them. But they are not immutable. There is constant tension between firms and regulators about appropriate constraints on corporate conduct. Sometimes financial intermediaries win battles (and even wars) leading to periods of deregulation. Sometimes it’s possible to convince the public that self-regulation and market discipline are powerful enough to obviate the need for external control. Sometimes the regulators can be convinced, one way or another, to go easy. Then along comes another major transgression and the constraint system reacts and creates a spate of new regulations. A wide array of interests get into this constant battle to define the rules under which financial business gets done—managers, politicians, the media, activists, investors, lawyers, accountants—and eventually a new equilibrium gets established which will define the rules of engagement for the period ahead. There are some more fundamental things at work as well. Laws and regulations governing the market conduct of firms are not created in a vacuum. They are rooted in social expectations as to what is appropriate and inappropriate, which in turn are driven by values imbedded in society. These values are rather basic. They deal with lying, cheating and stealing, with trust and honor, with what is right and what is wrong. These are the ultimate benchmarks against which conduct is measured and can be the origins of key reputational losses. But fundamental values in society may or may not be reflected in people’s expectations as to how a firm’s conduct is assessed. There may be a good deal of slippage between social values and how these are reflected in the public expectations of business conduct. Such expectations are nevertheless important and the build-up of adverse opinion in the media, the formation of special-interest lobbies and pressuregroups, and the general tide of public opinion with respect to one or another aspect of market conduct can be reputationally debilitating. Moreover, neither values nor expectations are static in time. Both change. But values seem to change much more gradually than expectations. Indeed, fundamental values such as those noted above are probably as close as one comes to “constants” in assessing business conduct. But even in this domain things do change. As society becomes more diverse and mobile, for example, values tend to evolve. They also differ across cultures. And they are sometimes difficult to interpret. Is lying to clients or to trading counterparties wrong? What is the difference between lying and bluffing? Is it only the context that determines how particular behavior is assessed? The same conduct may be interpreted differently under different circumstances—interpretation that may change significantly over time and differ widely across cultures, giving rise to unique contours of reputational risk. There is additional slippage between society’s expectations and the formation of public policy, and the activities of public interest groups. Things may go on as usual for awhile despite occasional media commentary about inappropriate behavior of a firm or an industry in the marketplace. Then some sort of social tolerance limit is reached. A firm goes too far. A confluence emerges among various groups concerned with the issue. The system reacts through the political process and a new set of constraints on firm behavior emerge, possibly anchored in legislation, regulation,
80
I. Walter
and bureaucracy. Or the firm is subject to class action litigation.4 Or its reputation is so seriously compromised that its share price drops sharply. As managers review the reputational experiences of their competitors, they cannot escape an important message. Most financial firms can endure a credit loss or the cost of an unsuccessful trade or a broken deal, however large, and still survive. These are business risks that the firms have learned to detect and limit their exposure to before the damage becomes serious. Reputational losses may be imposed by external reactions that may appear to professionals as unfocused or ambiguous, even unfair. They may also be new—a new reading of the rules, a new finding of culpability, something different from the way things were done before. Although regulators and litigants, analysts and the media are accepted by financial professionals as a fact of life, such outsiders can become susceptible to public uproar and political pressure, during which times it is difficult to take the side of an offending financial firm.5 In the United States, for example, tighter regulation and closer surveillance, aggressive prosecution and plaintiff litigation, unsympathetic media and juries, and stricter guidelines for penalties and sentencing make it easier to get into trouble and harder to avoid serious penalties. Global brokerage and trading operations, for example, involve hundreds of different, complex and constantly changing products that are difficult to monitor carefully under the best of circumstances. Doing this in a highly competitive market, where profit margins are under constant challenge and there is considerable temptation to break the rules, is even more challenging. Performance-driven managers, through compensation and promotion practices, have sometimes unwittingly encouraged behavior that has inflicted major reputational damage on their firms and brought some of them down. The reality is that the value of financial intermediaries suffers from such uncertain reputation-sensitive conditions. Since maximizing the value of the firm is supposed to be the ultimate role of management, its job is to learn how to run the firm so that it optimizes the long-term trade-offs between profits and external control. It does no good to plead unfair treatment—the task is for management to learn to live with it, and to make the most of the variables it can control. The overall process can be depicted in a graphic such as Exhibit 3, representing the firm and its internal governance processes in the center and various layers of external controls affecting both the firm’s conduct and the reputational consequences of misconduct, ranging from “hard” compliance components near the center to “soft” but potentially vital issues of “appropriate” conduct in the periphery. Clearly, serious reputational losses can impact a financial firm even if it is fully in compliance with regulatory constraints and its actions are entirely legal. The risk of reputational damage incurred in these outer fringes of the web of social control are among the most difficult to assess and to manage. Nor is the constraint system necessarily consistent, with important differences in regulatory regimes (as well as expectations regarding responsible conduct) across markets in which a firm is active—so that conduct which is considered acceptable in one environment may give rise or significant reputational risk in another.
Reputational Risk and Conflicts of Interest
81
Exhibit 3 Reputational risk and the external control web
Valuing Reputation Risk Recent research has attempted to quantify the impact of reputation risk on share prices during the 1980s and 1990s.6 Given the nature of the problem, most of the evidence has been anecdotal, although a number of event studies have been undertaken in cases where the reputation-sensitive event was “clean” in terms of the release of the relevant information to the market. Exhibit 4 summarizes shareholder value losses in a reputation-sensitive situation involving the aforementioned sources of loss: (1) client defections and revenue erosion; (2) increases in monetary costs comprising accounting writeoffs associated with the event, increased compliance costs, regulatory fines and legal settlements
Exhibit 4 Reputational-sensitive events in a simple going-concern valuation framework
82
I. Walter
as well as indirect costs related to loss of reputation such as higher financing costs, contracting costs and opportunity costs—including “penalty box” suspension by the regulators from particular business activities; and (3) an increase in firm-specific (unsystematic) risk assigned by the market as a result of the reputational event in question. In order to value the pure reputational losses, it is necessary to estimate the overall market value loss of the firm to a reputation-sensitive event and subtract from it the monetary losses identified in italics in Exhibit 4. Consider the following example:7 On December 28, 1993, the Bank of Spain took control of the country’s fourth largest bank, Banco Español de Crédito (Banesto). Subsequently, shares of JP Morgan & Co., a U.S. bank holding company closely involved with Banesto, declined dramatically. Such a reaction appeared inconsistent with market rationality, given that the impact of the event on Morgan’s bottom line was trivial (the accounting loss to Morgan was unlikely to exceed $10 million after taxes). Perhaps something more than the underlying book value of JP Morgan & Co. was moving the price of the stock, i.e., the central bank takeover of Banesto may have affected the value of Morgan’s corporate franchise in some of the firm’s core business areas, notably securities underwriting, funds management, client advisory work and its ability to manage conflicts of interest that can accompany such activities in non-transparent environments. JP Morgan was involved in Banesto in four ways, in addition to normal interbank transactions relationships.8 First, in May 1992, it began raising funds for the Corsair Partnership, L.P., aimed at making non-controlling investments in financial institutions. By February 1993, Morgan had raised over $1 billion from 46 investors that included pension funds and private individuals. Morgan served as General Partner and fund manager, with an investment of $100 million. The Corsair Partnership’s objective was to identify troubled financial institutions and, by improving their performance, earn a significant return to shareholders in the fund. The Corsair Partnership’s first investment, undertaken in February 1993, was a share purchase of $162 million in Banesto—giving Morgan a $16.2 million equity stake in the Spanish bank. Second, a vice-chairman of JP Morgan served on the Spanish bank’s board of directors. Third, Morgan was directly advising Banesto on its financial and business affairs. Fourth, as part of an effort to recapitalize Banesto, Morgan was lead underwriter during 1993 of two stock offerings that totaled $710 million. Corsair Partnership, L.P. was intended to search for troubled financial institutions in the United States and abroad. The objective was to restructure such institutions by applying Morgan’s extensive expertise and contacts. Morgan indicated that Corsair investors could expect a 30% annual return over 10 years. Although Morgan had a separate investment banking subsidiary (JP Morgan Securities Inc.), Corsair was believed to be the first equity fund organized and managed by Morgan since the Glass-Steagall Act separated banking and securities activities in 1933—a separation which eventually lasted until 1999. The business concept of searching for troubled financial institutions emerged from a time of turmoil in the U.S. and foreign banking sectors. When the U.S. banking industry started to improve as a result of a favorable interest rate environment, Corsair ventured abroad. Corsair’s first stake in Banesto was taken in February 1993. By August 1993, it had invested $162 million (23% of
Reputational Risk and Conflicts of Interest
83
the funds raised) in the Spanish bank. The overall JP Morgan-Banesto relationship is depicted in Exhibit 5.
Exhibit 5 Reputational risk exposure—JP Morgan and Banco Espanol de Credito, 1993
Banesto’s problems stemmed from rapid growth and a convoluted structure of industrial holdings followed by a serious downturn in the Spanish economy. The bank’s lending book increased from Pta.4 trillion in 1988 to Pta2.3 trillion in 1991, a period when its competitors were growing at a quarter of that rate. Banesto bid aggressively for deposits, increasing interest rates by 51% while competitors increased theirs by 40%. When the Spanish economy weakened, the bank was stuck with an array of bad loans, and the group was further burdened by losses on its industrial holdings. In October 1992, a partial audit by the Bank of Spain was forced to lend the troubled institution “a substantial amount.” An audit released at the end of December 1993 revealed that Banesto assets of Pta5.5 trillion ($385 billion) were overvalued in excess of Pta50 billion ($3.5 billion). In April 1994, Banesto was bought for $2.05 billion by Banco Santander, leaving costs of $3.7 billion to be borne by the Spanish banks and by taxpayers. Morgan had been advising Banesto on various deals since 1987. In July 1992, Morgan’s involvement became more intensive when it began advising Banesto on how to raise capital. By August 1993, Morgan had assisted Banesto in two rights issues to raise $710 million. During the time of these rights issues, Corsair invested $162 million in Banesto. In a letter dated December 27, 1993, Morgan wrote to the Bank of Spain’s Governor, outlining how Banesto could continue to raise capital, including a bond issue that Morgan was planning to launch in the first quarter of 1993.
84
I. Walter
Instead, the Bank of Spain took control of Banesto on the following day, December 28, 1993. Citing mismanagement and reckless lending, the Governor justified the action in order to avoid a run on the deposits of the bank, whose share prices were falling sharply on the Madrid Exchange. Given Morgan’s multifaceted involvement in Banesto and potential conflicts imbedded in that relationship, the announcement of the takeover could have had a large effect on the value of Morgan’s reputation and business franchise and hence its stock price. In order to test the impact of the Banesto case on the JP Morgan share price, we use conventional event study methodology.9 We create a sample prediction of returns on Morgan stock and compare the predicted returns with actual returns on Morgan shares after the Banesto event announcement. The difference is considered the excess return attributable to the event. In order to create this prediction, we regress the daily return of Morgan stock on the daily return on the market index as well as on an industry-group index.10 We use data from 300 to 50 days prior to the announcement date (December 28). The resulting coefficients are then multiplied by the returns on the market and industry indices from 50 days prior to 50 days after the announcement, in order to obtain an estimation of the daily stock return during this period. Then, the excess return is calculated at the “predicted” return minus the actual Morgan stock returns for the period, and the cumulative excess return is plotted. In order to translate these results into the monetary effect on JP Morgan stock, the cumulated excess return is multiplied by the total market value of equity (shares outstanding times price per share) 50 days before the announcement. In effect, this amount represents the difference between what shareholders would have received had they sold their shares in the market 50 days prior to the announcement and the industry’s stocks, and what they would have received if they had sold them on subsequent days. If the reputation-effect hypothesis is correct, the market response to the Bank of Spain’s announcement on 28 December 1993 should have significantly exceeded the firm’s book exposure to Banesto. We regressed Morgan’s stock returns against the value-weighted NYSE index11 and the industry group composed of 20 banking and securities firms. We obtained the following model, estimated over days –300 to –50 prior to the announcement date: RJPMt = −0.00014 + 0.5766∗ RMt + 0.2714∗ RGt + ut where RJPMt = Return on JP Morgan stock; RMt = Return on NYSE composite (value-weighted) index; RGt = Return on group of companies in the same industry. The excess return attributable to the event is the calculated residual (ut ) from 50 days prior to 50 days after the announcement.12 Prior to the announcement, Morgan stock behaved as predicted based on its behavior during the 250 days before the event period. A few days before announcement, the return began to decline. Thereafter, an essentially steady decline occurred. A cumulative loss of 10% of
Reputational Risk and Conflicts of Interest
85
shareholder equity value is apparent 50 days after the announcement. In monetary terms, the 10% loss in shareholder value translates into a loss in JPM market capitalization of approximately $1.5 billion versus a maximum direct loss of only $10 million from the Banesto failure. This analysis suggests that the loss of an institution’s franchise value can far outweigh an accounting loss when its reputation is called into question, a finding similar to that of Smith (1992) in the case of Salomon Brothers, Inc. Reasons for the adverse market reaction can only be conjectured. The takeover of Banesto could have been seen as compromising Morgan’s reputation in precisely those areas key to its future. Inability to turn Banesto around may have called into question Morgan’s ability to successfully advise clients. Banesto as the dominant participation in the Corsair portfolio may have suggested flaws in Morgan’s ability to organize and manage certain equity funds. Underwriting stock and placing them with important investor clients raises questions about its ability to judge risks in underwriting securities. Service on Banesto’s Board suggests problems with monitoring, and the configuration of Morgan’s various involvements with Banesto suggests the potential for conflicts of interest or lack of objectivity. Whatever the linkages, here was a case of a financial services firm of exceedingly high standing, which in no way violated any legal or regulatory constraints but whose shares nevertheless appeared to have been adversely affected by the market reaction to the way a high-profile piece of business was handled. In recent years, event studies such as this have yielded a growing body of evidence as to the share price sensitivity to reputational risk. For example, Cummins, Lewis, and Wei (2004) undertook a large sample study of operational TM and reputational events contained in the Fitch OpVar database. Exhibit 6 shows
Exhibit 6 Cumulative abnormal returns (CARs) for banks and insurers in a large-sample study of operational and reputational events (three-factor models)
86
I. Walter
the results in terms of the magnitude of the losses using three-factor estimation models in terms of cumulative abnormal returns (CARs) and number of trading days before and after the announcement. The authors, however, do not distinguish between operational losses and reputational losses, as defined above. TM De Fontnouvelle et al. (2006) use loss data from the Fitch OpVar and SAS TM OpRisk databases to model operational risk for banks that are internationally active. In a series of robust statistical estimates, they find a high degree of regularity in operational losses that are both quantifiable and consistent with large amount of capital increasingly reported by major banks as being held against operational risk—which in many cases exceeds capital held against market risk—see Exhibits 7 and 8. The paper also segments the losses by event type and by activity line, as well as whether or not the operational losses occurred in the United States. The largest losses involved retail and commercial and retail/private banking activities in terms of type of event. As in the case of other studies, the authors do not distinguish the associated accounting losses due to legal settlements, fines, penalties and other explicit operational risk-related costs from reputational losses, and as such these estimates are relevant from a regulatory perspective but probably materially understate the losses to shareholders.
Exhibit 7 Operational losses by event type
In a pilot study of 49 reputation-sensitive events, using the aforementioned definition and excluding operational events, we find negative mean CARs of up to 7% and $3.5 billion, depending on the event windows used. Exhibit 9 shows the results graphically and the tables in Exhibits 10 and 11 show the numerical results. We
Reputational Risk and Conflicts of Interest
87
Exhibit 8 Operational losses by business line
do not, however, distinguish between the associated monetary losses and the pure reputational losses.13 The only study so far which attempts to identify pure reputational losses is Karpoff, Lee and Martin (2006). The authors attempt to distinguish book losses from reputational losses in the context of US Securities and Exchange Commission enforcement actions related to earnings restatements—“cooking the books.” The authors review 2,532 regulatory events in connection with all relevant SEC enforcement actions during 1978–2002 and the monetary costs of these actions in the ensuing period, through 2005. These monetary costs are then compared with the cumulative abnormal returns estimated from event studies to separate them from the reputational costs. The results are depicted in Exhibit 12—note that the reputational losses (66%) are far larger than the cost of fines (3%), class action settlements (6%) and accounting writeoffs (25%) resulting from the events in question.
Reputational Risk and Conflicts of Interest One of the key sources of reputational risk in the financial services sector is the exploitation of conflicts of interest. Potential conflicts of interest are a fact of life in the financial services industry and always will be. The question is whether they are exploited, and thereby impose agency costs on others. In recent years, the role of banks, securities firms, insurance companies, and asset managers have become
88
I. Walter
Exhibit 9 Reputational impact and shares prices pilot study—49 events, 1998–2005 (unweighted mean CARs)
Exhibit 10 Relative CARs—reputational loss pilot study
enmeshed in alleged abusive practices — as facilitators in various corporate scandals, acting simultaneously as principals and intermediaries in various high-profile transactions, for example. This has turned the attention of supervisory authorities, public prosecutors, legislators and the conflict-of-interest issue.14
Reputational Risk and Conflicts of Interest
89
Exhibit 11 Absolute CARs—reputational loss pilot study
Exhibit 12 Decomposing CARs related to earnings restatements
In this section we note that conflict-of-interest exploitation requires information asymmetries, transaction costs and market frictions—in perfect markets, conflicts of interest cannot be exploited. We argue that conflict of interest exploitation
90
I. Walter
is sensitive to the strategic positioning of the financial intermediary, as well as strategic execution and the intensity of performance pressure imposed on individual business units. Finally, we suggest that appropriate conflict of interest diagnostics can promote sensible safeguards against the reputational exposure involved. There are essentially two kinds of conflicts of interest confronting firms in the financial services industry. Type-1 conflicts are those between the firm’s own economic interests and the interests of its clients. Examples include exploiting conflicts of interest in order to enhance the firm’s profitability or market-share, or to transfer risk. Type-2 conflicts of interest may develop between the firm’s clients (or between types of clients) which place the firm in a position of favoring one at the expense of another. Behavior that systematically favors corporate clients over retail investors in the presence of asymmetric information is an example of this type of conflict. (Walter, 2004) Each of these types of conflicts of interest may arise either in inter-professional activities carried out in wholesale financial markets or in activities involving retail clients. The distinction between these two domains is important because of the key role of information asymmetries and transactions costs, which differ dramatically between the two broad types of market participants. Consequently, their vulnerability to conflict-exploitation differs significantly, and measures designed to remedy the problem in one domain may be inappropriate in the other. In addition there are what we term “domain-transition” conflicts of interest, which run between the two domains and whose impact can be particularly troublesome. Conflicts of interest in wholesale financial markets are depicted in the left panel of Exhibit 13. For example, a financial intermediary may be involved as a principal with an equity stake in a transaction in which it is also serving as adviser, lender or underwriter, creating an incentive to act in its own interest and against those of its clients or third parties. Or a financial intermediary may use its lending power to influence a client to use its securities or advisory services as well—or the
Exhibit 13 Conflict of interest domain-mapping
Reputational Risk and Conflicts of Interest
91
reverse, denying credit to clients that refuse to use other (more profitable) services. Or the asset management unit of a financial institution may be pressured by a corporate banking client into proxy-voting of shares in that company for management’s position in a contested corporate action. Or a multifunctional financial firm may act as trading counterparty for its own fiduciary clients, as when the firm’s asset management unit sells or buys securities for a fiduciary client while its affiliated broker-dealer is on the other side of the trade. Or a financial intermediary may exploit institutional, corporate, or other wholesale clients by executing proprietary trades in advance of client trades that may move the market. All of these represent exploitation of Type-1 conflicts, which set the firm’s own interest against those of its clients in wholesale, inter-professional transactions. Type-2 conflicts dealing with differences in the interests of multiple wholesale clients center predominantly on two issues: (1) a financial intermediary may obtain private information about a client, which in turn may be used in ways that harm the interests of that client; or (2) a financial firm may have a relationship with two or more clients who are themselves in conflict. Conflicts of interest in retail financial markets are depicted in the right panel of Exhibit 13. All such conflicts appear to be Type-1 conflicts, which set the interests of the financial intermediary itself against the interests of its clients. They include biased client advice based on a “salesman’s stake” in promoting high-margin “house” products over lower-margin third-party products, based on incentives that are rarely transparent to the retail client. Or retail clients may be pressured to acquire additional financial services on unfavorable terms in order to access a particular product, such as the purchase of credit insurance tied to consumer or mortgage loans. Or a financial firm that is managing assets for clients may exploit its agency relationship by engaging in excessive trading which creates higher costs. Or clients may be encouraged to leverage their investment positions through margin loans from the firm, exposing them to potentially unsuitable levels of market risk and high credit costs. Or there may be misuse of personal information by a firm under intense pressure to cross-sell. Conflict-of-interest exploitation may also transition the wholesale and retail domains, as depicted in the center panel of Exhibit 13. This can involve either Type-1 or Type-2 conflicts, and sometimes both at the same time. One example is the classic conflict between a firm’s “promotional role” in raising capital for clients in the financial markets and its obligation to provide suitable investments for retail clients.15 Or a financial firm that is acting as an underwriter or has acquired securities in a secondary market trade may be unable to resell them at an acceptable price and may seek to cut its exposure to loss by allocating unwanted securities to investment accounts over which it has discretion. Or analysts working for sell-side firms in diverse and fundamentally incompatible roles may encounter intractable conflicts in taking views on listed equities. Or a bank with credit exposure to a client whose bankruptcy risk has increased (to the private knowledge of the banker) may have an incentive to assist the corporation in issuing bonds or equities to the general public, with the proceeds used to pay-down the bank’s loans.
92
I. Walter
Aside from this basic taxonomy of conflict of interest exploitation, we posit that the broader the range of a financial intermediary’s activities, (1) the greater the likelihood that the firm will encounter exploitable conflicts of interest, (2) the higher will be the potential agency costs facing its clients, and (3) the more difficult and costly will be the safeguards necessary to prevent conflict of interest exploitation. If this proposition is correct, agency costs associated with conflicts of interest can easily offset the realization of economies of scope in financial services firms—scope economies that are supposed to generate benefits on the demand side through crossselling (revenue synergies) and on the supply side through more efficient use of the firm’s business infrastructure (cost synergies). As a result of conflict exploitation, the firm may win and clients may lose in the first instance; but subsequent adverse reputational and regulatory consequences (along with efficiency factors such as the managerial and operational cost of complexity) can be considered diseconomies of scope. Breadth of engagement with clients may create conflicts of interest that can be multidimensional, and involve a number of different stakeholders at the same time. Several examples came to light during the corporate scandals in the early 2000s. Following the $103 billion bankruptcy of WordCom in 2002, for example, it appeared that Citigroup—a multifunctional, global financial conglomerate—was serving as equity analyst supplying assessments of WorldCom to institutional and (through the firm’s brokers) retail clients, while simultaneously advising WorldCom management on strategic and financial matters. Citigroup’s equity analyst at times participated in WorldCom’s board meetings. As a major telecommunications-sector commercial and investment banking client, Citigroup maintained an active lending relationship with WorldCom and successfully competed for its securities underwriting business. At the same time, Citigroup served as the exclusive pension fund adviser to WorldCom and executed significant stock option trades for WorldCom executives, while at the same time conducting proprietary trading in WorldCom stock and holding a significant position in the company’s stock through its asset management unit. Additionally, Citigroup advised the WorldCom CEO, financed margin purchases of company stock, and provided loans for one of his private businesses. On the one hand, Citigroup was very successfully engaged in the pursuit of revenue economies of scope (cross-selling), simultaneously targeting both the asset and liability sides of its client’s balance sheet, generating advisory fee income, managing assets, and meeting the private banking needs of WorldCom’s CEO. On the other hand, that same success caught the firm in simultaneous conflicts of interest relating to retail investors, institutional fund managers, WorldCom executives and shareholders as well as Citigroup’s own positions in WorldCom credit exposure and stock trades. WorldCom’s bankruptcy triggered a large market capitalization loss for Citigroup’s own shareholders, only part of which can be explained by a $2.65 billion civil settlement the firm reached with investors in May 2004.16 It seems plausible that the broader the range of services that a financial firm provides to a given client in the market, and the greater the cross-selling pressure, the greater the potential likelihood that conflicts of interest will be compounded in
Reputational Risk and Conflicts of Interest
93
any given case and, when these conflicts of interest are exploited, the more likely they are to damage the market value of the financial firm’s business franchise once they come to light. Similarly, the more active a financial intermediary becomes in principal transactions such as affiliated private equity businesses and hedge funds, the more exposed it is likely to be to reputational risk related to conflicts of interest.
Controlling Conflicts of Interest Mechanisms to control conflicts of interest are based on either regulation, civil litigation, or market discipline—often a combination. These external controls, in turn, form the basis for a set of internal controls, which can be either prohibitive or affirmative, involving in the first instance the behavioral “tone” and incentives set by boards and senior management together with reliance on the loyalty and professional conduct of employees. They are fundamentally matters of sound corporate governance. Regulatory control of conflicts of interest tends to be applied through both SROs and public agencies, and is generally anchored in banking, insurance, securities, and consumer protection legislation that is supposed to govern market practices. Its failure to prevent serious exploitation of conflicts of interest became evident in the US and elsewhere during the early 2000s with serial revelations of misconduct by financial intermediaries. The corrective initiative in this instance was taken not by the responsible SROs (the NYSE or the NASD, for example) or by the national regulators (such as the SEC), but in large measure by the New York State Attorney General under the Martin Act, a 1921 state law aimed at securities fraud which survived all subsequent banking and securities legislation and was bolstered in 1955 with the addition of criminal penalties. The Act contains extremely broad “fraud” provisions and conveys unusually wide discovery and subpoena power, but had been largely dormant until the 2001–02 revelations of the excesses in financial market practices and corporate governance. The de facto ceding of enforcement actions by the SROs and the SEC to a state prosecutor (later joined by several others) focused attention on gaps in regulation and led to a burst of activity by the SROs, the regulators and the Congress, including the 2002 Sarbanes-Oxley Act. It also led to a large number of nolo contendere settlements by major banks and securities firms, none of which created any useful legal guidance for the future. Civil litigation proved to be an important component of external control of financial intermediary conduct, especially when linked to regulatory sanctions. Despite high costs and occasional unjust outcomes, US tort litigation is arguably an important adjunct to market discipline with respect to exploitation of conflicts of interest. Nevertheless, both regulatory action and civil litigation are blunt instruments in dealing with exploitation of conflicts of interest in financial intermediaries, conflicts that are often extremely granular and sometimes involve conduct that is “inappropriate” or “unethical” rather than “illegal” in terms of Exhibit 3. So market discipline via reputational impacts on share prices may provide a more consistent and durable basis for internal defenses against exploitation of conflicts of interest than those
94
I. Walter
mandated by the regulators or implemented through the compliance infrastructure by legal staff reporting to senior management—or the threat of litigation. There are several linkages that can be identified. First, market discipline can leverage the effectiveness of regulatory actions. When they are announced, regulatory actions can have an adverse effect on a financial firm’s share price and competitive advantage linked to the share price such as the cost of capital, the ability to make acquisitions and vulnerability to takeover, and management compensation. As noted, any such share-price effects reflect the market’s response to the prospective combined impact of regulatory actions on revenues, costs, and exposure to risk. In addition, regulatory actions or abrupt share price declines can trigger derivative civil litigation. In extreme cases the firm could be taken over, broken up, or go out of business. Awareness of these risks on the part of boards and managements ought to be reflected in compensation arrangements as well as organizational structure—effective separation in wholesale financial intermediation of trading, asset management, and corporate finance, for example. Second, even in the absence of regulatory constraints, actions that are widely considered to be “unfair,” “unethical,” or otherwise contrary to the external constraint system discussed earlier will tend to be subject to market discipline through its reputational impacts. In a competitive context, this will affect firm valuation through the revenue and risk dimensions identified in Exhibit 4, in particular. That is, even in the absence of regulatory constraints, management ought to be aware that efforts to avoid conflict of interest exploitation and other sources of reputational damage are likely to reinforce the value of the firm as a going concern and, with properly structured incentives, their own rewards. Since they tend to be more granular and applied in a real-time context, market discipline constraints can reach the more opaque risks to reputational capital, including conflict of interest exploitation. They can identify such issues as they occur in real time, which external regulation normally cannot do. Third, since market-conduct regulation tends to be linked to information asymmetries and transactions costs, optimum regulation should be carefully tailored to market domain—notably the wholesale and retail domains. Often this is not possible, resulting in overregulation in some areas and underregulation in others. Market discipline-based constraints can help alleviate this problem by permitting lower overall levels of regulation. Particularly in the case of conflicts of interest that bridge the wholesale and retail domains, market discipline can be effective in dealing with fault-lines across financial market segments. And, just as market discipline can reinforce the effectiveness of regulation, it can also serve as a precursor of sensible regulatory change. Fourth, market structure and competition across strategic groups can help determine the effectiveness of market discipline. For example, inside information accessible to a bank as lender to a target firm would almost certainly preclude its affiliated investment banking unit from acting as an adviser to a potential acquirer. An entrepreneur may not want his or her private banking affairs handled by a bank that also controls his or her business financing. A broker may be encouraged by a firm’s compensation arrangements to sell in-house mutual funds or
Reputational Risk and Conflicts of Interest
95
externally-managed funds with high fees under “revenue-sharing” arrangements, as opposed to funds that would better suit the client’s needs. Market discipline that helps avoid exploitation of such conflicts of interest may be limited if most of the competition is coming from financial conglomerates that face the same issues. But if the playing field is also populated by aggressive insurance companies, brokerdealers, fund managers and other “monoline” specialists, market discipline may be much more effective.
Conclusions We have attempted to define reputational risk and to outline the sources of such risk facing financial services firms. We then considered the key drivers of reputational risk in the presence of transactions costs and imperfect information and surveyed available empirical research on the impact of reputational losses imposed on financial intermediaries. We then developed the link between reputational risk and exploitation of conflicts of interest, arguably one of the most important threats to the reputational capital of a financial intermediary. Finally, we considered managerial requisites for dealing with both reputational risk and conflicts of interest. We conclude that market discipline, through the reputation-effects on the franchise value of financial intermediaries, can be a powerful complement to regulation and civil litigation. Nevertheless, market discipline-based controls remain controversial. Financial firms continue to encounter serious instances of reputation loss due to misconduct despite its effects on the value of their franchises. This suggests material lapses in the governance process. Dealing with reputational risk and controlling exploitation of conflicts of interest can be an expensive business, with compliance systems that are costly to maintain and various types of walls between business units and functions that impose significant opportunity costs due to inefficient use of information within the organization. Moreover, management of certain kinds of conflicts in multifunctional financial firms may be sufficiently difficult to require structural remediation. On the other hand, reputation losses associated with conflict of interest exploitation can cause serious damage—as demonstrated by reputation-sensitive “accidents” that seem to occur repeatedly in the financial services industry. Indeed, it can be argued that such issues contribute to market valuations among financial conglomerates that fall below valuations of more specialized financial services businesses (Laeven & Levine, 2005; Schmid & Walter, 2006).17 Managements and boards of financial intermediaries must be convinced that a good defense is as important as a good offense in determining sustainable competitive performance. This is something that is extraordinarily difficult to put into practice in a highly competitive environment for both financial services and for the highly skilled professionals that comprise the industry. It seems to require an unusual degree of senior management leadership and commitment (Smith & Walter, 1997). Internally, there have to be mechanisms that reinforce the loyalty and professional conduct of employees. Externally, there has to be careful and sustained
96
I. Walter
attention to reputation and competition as disciplinary mechanisms. In the end, it is probably leadership more than anything else that separates winners from losers over the long term—the notion that appropriate professional behavior reinforced by a sense of belonging to a quality franchise constitutes a decisive comparative advantage.
Notes 1. Earlier studies focusing on reputation include Chemmanur & Fulghieri (1994), Smith (1992), Walter & De Long (1995) and Smith & Walter (1997). 2. Basle II at http://www.bis.org/publ/bcbs107.htm. 3. For an early discussion of external conduct benchmarks, see Galbraith (1973). 4. For a discussion, see Capiello (2006). 5. For a full examination of these issues, see Smith & Walter (1997). 6. For one of the early studies, see Smith (1992). 7. See Walter and De Long (1995) 8. For a journalistic account, see The Wall Street Journal (1994) and Eurom 9. De Long & Walter (1994). For event study methodology, see Brown and Warner (1985). 10. The industry group index included 20 financial institutions with characteristics showing some degree of overlap with those of JP Morgan. This is the unweighted average of share prices for Banc One, BankAmerica, Bank of Boston, Bank of New York, Bankers Trust NY, Barnett Bank, Bear Stearns, Chase Manhattan, Chemical Bank, Citicorp, Continental Bank, First Chicago, First Fidelity Bancorp, First Virginia, Merrill Lynch, Morgan Stanley Group, NationsBank, Paine Webber Group, Salomon Inc. and Wells Fargo. 11. While autocorrelation can be a problem in using daily stock returns, JP Morgan stock was heavily traded, so that daily carryover is unlikely to be significant. Indeed, when we controlled the industry for this potential problem by including the lagged market index as a regressor, the resulting coefficient was negative and statistically insignificant. 12. A careful search was undertaken covering the 50 days after the event to check for further announcements that could have affected the JP Morgan share price. On January 14, 1994, Morgan announced that net earnings for the fourth quarter of 1993 were up by 77%. Despite an increase in earnings, therefore, stock prices fell during the period examined. Besides the statement on earnings, no other announcements occurred during the period. 13. Ongoing empirical work on reputation-sensitive financial services events with Gayle De Long and Anthony Saunders. 14. A survey of the economics of conflicts of interest can be found in Demsky (2003). Analysis of conflicts of interest applied to the financial services industry includes Herman (1975), Krozner & Strahan (1999), Saunders (1985) and Schotland (1980). 15. A well-known version of this conflict involves biased research. See Attorney General of the State of New York (2003). 16. Similar issues surfaced in the case of the 2001 Enron bankruptcy. See Batson (2003) and Healey & Palepu (2003). 17. See also Kanatas & Qi (2003) and Saunders & Walter (1997).
References Attorney General of the State of New York (2003). Global Settlement: Findings of Fact (Albany: Office of the State Attorney General). Batson, N. (2003). Final Report, Chapter 11, Case No. 01-16034 (AJG), United States Bankruptcy Court, Southern District of New York, July 28.
Reputational Risk and Conflicts of Interest
97
Brown, S.J. & Warner, J.B. (1985). “Using Daily Stock Returns: The Case of Event Studies,” Journal of Financial Economics, 14, 3–31. Capiello, S. (2006). “Public Enforcement and Class Actions against Conflicts of Interest in Universal Banking—The US Experience Vis-à-vis Recent Italian Initiatives.” Bank of Italy, Law and Economics Research Department, Working Paper. Chemmanur, T.J. & Fulghieri, P. (1994). “Investment Bank Reputation, Information Production, and Financial Intermediation,” Journal of Finance, 49, March 57–79. Cummins, J.D., Christopher, M.L., & Ran, W. (2004). “The Market Impact of Operational Risk Events For US Banks and Insurers.” The Wharton School, Working Papers. Available at
http://papers.ssrn.com/sol3/ papers.cfm?abstract_id=640061 De Fontnouvelle, P., DeJesus-Rueff, V., Jordan, J.S., & Rosengren, E.S. (2006). “Capital and Risk: New Evidence on Implications of Large Operational Losses.” Federal Reserve Bank of Boston. Working Paper. September. De Long, G. & Walter, I. (1994). “J.P. Morgan and Banesto: An Event Study.” New York University Salomon Center. Working Paper. April. Demsky, J.S. (2003). “Corporate Conflicts of Interest,” Journal of Economic Perspectives, 17(2), Spring. Galbraith, J.K. (1973). Economics and the Public Purpose (New York: Macmillan). Herman. E.S. (1975). Conflicts of Interest: Commercial Banks and Trust Companies (New York: Twentieth Century Fund). Healey, P.M. & Palepu, K.G. (2003). “The Fall of Enron,” Journal of Economic Perspectives, 17(2), Spring. Kanatas, G. & Qi, J. (2003). “Integration of Lending and Underwriting: Implications of Scope Economies,” Journal of Finance, 58(3). Karpoff, J.M., Lee, D.S., & Martin, G.S. (2006). “The Cost to Firms of Cooking the Books.” University of Washington Working Papers. Available at SSRN: http://ssrn.com/abstract=652121 Krozner, R.S. & Strahan, P.E. (1999). “Bankers on Boards, Conflicts of Interest, and Lender Liability,” NBER Working Paper No. W7319, August. Laeven, L. & Levine, R. (2005). “Is There a Diversification Discount in Financial Conglomerates?” C.E.P.R. Discussion Papers No. 5121. Saunders, A. (1985). “Conflicts of Interest: An Economic View,” in Ingo Walter (ed.). Deregulating Wall Street (New York: John Wiley). Saunders, A. & Walter, I. (1997). Universal Banking In the United States: What Could We Gain? What Could We Lose? (New York: Oxford University Press). Schmid, M.M. & Walter, I. (2006). “Do Financial Conglomerates Create or Destroy Economic Value?” SSRN: http://ssrn.com/abstract=929160. Schotland, R.A. (1980). Abuse on Wall Street: Conflicts of Interest in the Securities Markets (Westport, Ct.: Quantum Books). Smith, C.W. (1992). “Economics and Ethics: The Case of Salomon Brothers,” Journal of Applied Corporate Finance, 5(2), Summer. Smith, R.C. & Walter, I. (1997). Street Smarts: Linking Professional Conduct and Shareholder Value in the Securities Industry (Boston: Harvard Business School Press). Walter, I. (2004) “Conflicts of Interest and Market Discipline in Financial Services Firms,” in Claudio Borio, William Curt Hunter, George G. Kaufman, and Kostas Tsatsaronis (eds.), Market Discipline Across Countries and Industries (Cambridge: MIT Press). Walter, I. & DeLong, G. (1995). “The Reputation Effect of International Merchant Banking Accidents: Evidence from Stock Market Data,” New York University Salomon Center, Working Paper.
Seeking Common Ground on Globalization Murray Weidenbaum
Globalization is one of the great and controversial developments of our time. It is an ongoing process that more closely integrates the many local, regional, and national markets. Globalization is characterized by a rising tendency for national borders to be crossed by people, goods, services, money, information, and ideas. Many specific factors are involved in the fundamental shifts that are occurring in the world economy. A vital force at play is technological advance, which is a key to the pace of globalization as well as a source of public concern. Transportation and communication barriers between nations have fallen as new technology has dramatically reduced the expense and time required to move people, goods, and information across borders. As a result, in recent decades international trade has been expanding twice as fast as domestic production. Direct overseas investment is growing about three times as rapidly as world trade. Yet the key changes in international relations have not been economic, although the economic ramifications are powerful. The world has witnessed not only the breakup of the Soviet Union, but also the reunification of Germany, the expansion of the European Union, and the rise of China and India as emerging world powers. Not all of the changes have been positive. The Middle East and portions of Africa and South Asia have witnessed violent responses by people who, in large measure, do not participate in the modern economy or who believe that they do not benefit from globalization. It is useful, therefore, to look at the various sides of the debate on globalization.
The Pros and Cons of Globalization Most economists and business leaders focus on the benefits of globalization, and they are substantial (U.S. Trade Deficit Review Commission, 2000; Rodrik, 1997). A greater flow of international trade and investment stimulates economic growth. World Bank studies show that developing countries that were globalizing in the 1990s grew twice as fast as the developed economies and also reduced their poverty rates at the same time. Increasing output requires more employment and income J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_9, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
99
100
M. Weidenbaum
payments and thus generates higher living standards for consumers (Sachs and Warner, 1996). Rising living standards in turn increase the willingness of the society to devote resources to the environment and other important social goals. The litany on the part of the proponents of globalization goes on. Global competition also keeps domestic businesses on their toes, forcing them to innovate and improve product quality and industrial productivity. According to this line of thinking, competition is good and spreading it out internationally must therefore be even better. More fundamentally, rapidly developing economies tend to generate a new middle class, and that is the bulwark of support for personal liberty as well as economic freedom. The most powerful benefit of the global economy may not be economic at all. It is the ability of people to exchange the most strategic of all factors—new ideas. That process empowers individuals in ways never before possible. Historical experience demonstrates that economic isolationism does not work. The most striking example was sixteenth-century China, where one misguided emperor abruptly cut off trade and commercial intercourse with other nations. China had been the wealthiest, most technologically advanced, and arguably the most powerful nation on the face of the globe. Yet it promptly went into a decline from which it has yet to fully emerge (Weidenbaum and Hughes, 1996). The real shortcoming of this line of thinking is not that the facts or analysis are wrong, but that it does not respond to the genuine concerns of the critics of globalization. Those other voices in the globalization debate emphasize the dark side. It is ironic that the portion of the American labor force that currently fears losing employment is now about two to three times higher than in 1981–1982 when the unemployment rate was twice as high (10% compared to 5%). Workers feel threatened by unfair competition from low-cost “sweatshops” overseas. Other citizens worry about the conditions in those factories, especially the presence of children in the workplace. People who care about the environment see the pollution caused by the long-distance movement of goods as well as the shift of production to overseas localities with low or no environmental standards. Simultaneously, financial crises arise in many parts of the globe, while mass starvation occurs amidst the collapse of whole societies in Africa. Meanwhile, concerns abound about the supposed growing inequality of income around the world. Apparently, the poor are getting poorer while the rich are getting richer. Globalization may be good for the compiler of economic statistics; but, according to this viewpoint, it is the antithesis of justice and fairness. At the same time, some government officials fear the loss of sovereignty and many individuals see an erosion of liberty with the rise of large multinational corporations and international agencies such as the World Bank, the International Monetary Fund, and the World Trade Organization. Moreover, global networks of transportation and communication also result in faster transmission of bad news as well as good. Wall Street’s woes are quickly shared by such diverse entities as Swiss insurance companies and Arab princes. The financial problems of Japanese banks translate directly into unemployment in South
Seeking Common Ground on Globalization
101
Korea and Thailand. The economy of Finland becomes tied to the fortunes of the U.S. technology sector, in bad times as well as good. All citizens face the rising power of international crime syndicates and spreading epidemics of AIDS and other threatening diseases. Of special concern is the rise of global terrorist groups that take advantage of the availability of low-cost international transportation and communication.
Trends in Globalization It is useful to note that economic historians tell us that, measured by trade, labor, and investment flows, the world economy may have been more integrated in the nineteenth century than it is today. For example, before passports were generally required for crossing borders, people were freer to travel and to migrate than they are now (Bordo, 2002). The extent of globalization—economic interdependence across national boundaries—did not decline in the early twentieth century because of mass protests or a bad press. The precipitating factors were far more fundamental—World War I, the worldwide depression of the 1930s, and the subsequent separation of the major nations into democratic and totalitarian camps that culminated in World War II. That long period was a time of rising isolationism, both political and economic. Without responding in detail to the critics, we can conclude that, on balance, globalization is neither the bright sun nor the dark side of the moon. Most economic analyses show that multinational corporations (MNCs) are effective sources of economic development in the poorer countries, providing new technology as well as the investment capital to apply it. U.S. companies, as well as other MNCs, are usually the leaders in offering higher wages (about double the average in low-income nations) and in setting more enlightened business standards (“Globalization and Its Critics,” 2001). The countries that have not kept up with the progress of the world economy are primarily those that have been bypassed by globalization, rather than having been exploited by the “greedy” MNCs. It seems to be clear that the substantial internationalization of business activity that has occurred in recent decades is a continuing phenomenon. However, the costs of globalization, even if they are much less than the public believes, are far more visible than the benefits. The companies and workers hurt by imports know who they are, while the beneficiaries of international commerce are widely distributed all through society. The modern global economy is characterized by a combination of substantial involvement of governmental agencies as well as large discretion on the part of the managers of individual business firms. As in the domestic economy, the involvement of government in business is not a static situation but an evolving phenomenon. Moreover, as in domestic regulation, the way that business responds to the various issues that arise strongly influences the government’s decisions on whether and how to intervene in economic matters. To the extent that public policy pays more attention to those who do not fully share the benefits, the likelihood of a serious backlash against globalization will be reduced.
102
M. Weidenbaum
The increasingly negative public reaction to the rapid and pervasive changes generated by globalization is beginning to overtake the positive aspects, at least in terms of perception. The instinctive response by economists is to correct the substantial amount of misinformation that has fueled the backlash to opening markets and to expanding the reach of competition. Frankly, no amount of technical brilliance is going to convince the people who are genuinely concerned over the dark side of globalization.
Developing Some Common Ground It is not clear how those who are seriously concerned with world commerce should proceed under these circumstances. Obviously, the terrorists are not going to be helpful. It is their diabolical use of modern technology that has been pacing the rise of globalization, which has enabled them to be so effective in the devastation that they generate. As economists have to reluctantly add, their efforts have been very cost-effective. It is the two more conventional sides of the globalization debate—those who favor freer markets and those who have other peaceful priorities—that we can turn to. However, even here a greater degree of trust and open-mindedness is necessary for the educational approach to succeed. It is tempting to respond directly to some of the more naïve protesters of globalization. My favorite sign is the one that blithely proclaims, “Food is for people, not for export.” It is painful to think of the malnutrition and worse that would follow from a ban on the export of food. Instead, let us try to identify some common ground on which people of good will on both sides of the heated controversy on globalization might possibly agree. The focus should be on useful but undramatic policy changes, what can be called the “nuts and bolts” of problem solving. We can note in passing that the greatest opposition to globalization comes from those who believe that they have no stake in it. So here are five suggestions for developing common ground: 1. Reform the World Trade Organization Many of the criticisms of the WTO are on target and they deserve a positive response. This UN-sponsored agency has become too closed and too bureaucratic. But there is little value in trying to shut it down. Its fundamental notion of advancing the rule of law on an international front is still an appealing idea. It does not diminish the adherence to free and open trade to state that the WTO has become inbred and rigid in its operations. For starters, the general sessions of trade negotiations should be open to the public, like Congress is. So should the hearings at which the various interest groups present their views. Yet, like legislative committees, the members should be expected to go into closed executive sessions when they begin to do the actual negotiations and drafting of trade agreements. That is a common sense distinction which experience teaches us is practical and workable.
Seeking Common Ground on Globalization
103
Similarly, the WTO’s critical process for settling trade disputes should be opened up so that the critics can see for themselves the nitty gritty of the workings of the WTO rules. The hearings of the dispute panels should be open to government and private observers as courtroom proceedings typically are. This does not require that the deliberations of the panels be public events. 2. Do Not Take a “Holier Than Thou” Position on Trade Issues The United States is not an island of free trade in a world of protectionism. Of course, other nations erect trade barriers—although the trend has been to cut back these obstacles to commerce. But our hands are not as clean as we like to think. There is no shortage of U.S.-imposed restrictions on importers trying to ship their products into this country. These exceptions to free trade come in all shapes, sizes, and varieties (Weidenbaum, 2004). U.S. import barriers include the following and more. Buy-American laws give preference in government procurement to domestic producers. Many states and localities show similar favoritism. In Michigan, for example, preference is given to in-state printing firms. At the federal level, the Jones Act prohibits foreign ships from engaging in waterborne commerce between U.S. ports. Other statutes limit the import of specific agricultural and manufactured products, ranging from sugar to pillowcases. The United States also imposes selective high tariffs on numerous specific items, notably textiles. Often state and local regulatory barriers, such as building codes, are aimed at protecting domestic producers. It does seem strange that consumer groups and consumer activists tend to be mute on the subject of trade restrictions. After all, it is the American consumer who has to pay higher prices as a result of all of these special interest laws. Nevertheless, the United States should adopt a tougher position on trade issues. The concern over the lack of fairness in trade rules and procedures is often justified, even if we disagree with the specific response of restricting trade. Promoting free trade is not just a matter of opening our markets to the products of other countries. Those nations need to do a better job of opening their markets to our products. Free trade, properly conducted, is a two-way street. Trade barriers by foreign governments offend the sense of fairness of Americans who see the greater openness of our economy. The United States should strengthen its efforts to monitor and enforce trade agreements. Responsibility for enforcement should be elevated within the two trade policy agencies, the Department of Commerce and the Office of the Trade Representative. Sufficient numbers of highly qualified staff should be assigned to this task, which is now the case. This, however, is treacherous ground. At times there is a fine line between keeping out blatantly improper imports (e.g., those made by forced child labor) and responding to pressures to exclude products just because they sell at prices lower than the cost of producing in the United States. The United States should work harder to get the nations we trade with to fully enforce the trade agreements they have entered into. Signing agreements to open borders to the products of other countries should be followed by action to carry
104
M. Weidenbaum
out those commitments. Sadly, this has not always been true. For example, at times China has responded to a commitment it makes to eliminate some paperwork obstacle to imports by substituting a new paperwork burden. 3. Help the People Hurt by Globalization Every significant economic change generates winners and losers. It does not satisfy the people hurt by globalization to tell them that far more people benefit from international trade and investment. Although that response is technically accurate, it is so cavalier that it is bound to infuriate those concerned with the dark side of globalization. A two-prong approach is needed. The United States must do a better job of helping the people who lose their jobs due to imports or the movement of factories to overseas locations. Simultaneously, we must grapple with the issue of the labor and environmental standards that are followed in poor (and thus usually low cost) countries by the companies that provide products for export to developed nations. Many of these overseas factories are either owned by companies in the developed world or they sell the bulk of their output to those western companies. Let us take up each of these two issues in turn. In the developed nations, such as our own, the most effective adjustment policy to help those who lose their jobs due to globalization—or for other reasons, notably technological advance—is to achieve a growing economy that generates a goodly supply of new jobs. In the absence of a successful macroeconomic policy, no adjustment programs will work well. Nevertheless, some more specific and constructive actions can be taken to improve the adjustment process. Often laid-off workers need just a modest bit of help, but they need it quickly. This is especially true for the people who went straight from school to work and never had to conduct a serious job hunt. For them, the most effective assistance is modest but essential: help in locating a new job and learning how to prepare for a job interview, including filling out a job application. That may seem elementary to any college graduate, but we are dealing with people who feel they have been treated badly by an economic system they do not really understand (Decker and Corson, 1995; Marcal, 2001). Many other unemployed people find that their job skills are obsolete or that much of their knowledge is only useful to their previous employer. They may be long on what we can call institutional information, but extremely short on the mathematics and the language capabilities required for many new and well paying jobs. These people could benefit from some pertinent education and training. Such “trade adjustment” programs have existed in the United States for many decades, but their track record is not very inspiring. Those public sector adjustment assistance programs need to be adjusted. They should be made more user friendly. “One stop” registration should replace the current uncoordinated array of assistance. It is disheartening for a newly unemployed worker to feel like a ping-pong ball being tossed from bureau to bureau, rarely encountering a government official who seems to have any real interest in his or her situation.
Seeking Common Ground on Globalization
105
In the United States, there is a type of educational and training institution that can be geared to serving unemployed blue-collar and white-collar workers. It is the network of community and junior colleges who serve a very different group of people than do the more prestigious senior colleges and universities. Older workers present an especially difficult challenge. They have limited motivation to undertake training programs that, at best, will prepare them for positions that pay much less than their customary wages—and in a labor market where they will compete against youngsters half their age. Some innovations are needed. One example of fresh thinking is the idea of providing “wage insurance” to pay a major portion of the difference in earnings between the new job and the previous position. The idea is to give the older workers the incentive to get back to work quickly before their skills become rusty. To the extent that such older workers demonstrate to their new employers their greater worth in terms of seasoned judgment and good work habits, they may find the wage gap between the old and new jobs narrowing. This would minimize the need to draw on the wage insurance plan. Congress has established a modest pilot program along these lines and the results should be analyzed when they become available. An even more contentious area is the issue of establishing labor and environmental standards for overseas locations that make products for export to the developed nations. There are several contending groups involved, each with its own goals and objectives. The labor unions in the industrialized countries resent the competition from workers in countries with lower costs of production and hence lower working standards. Certainly compared with pay and factory conditions in the United States, it is easy and sometimes accurate to label these places as “sweatshops.” As would be expected, employers have a somewhat different attitude toward the matter. They view low-cost production sites overseas as necessary to meet competition. Many Western firms report that the factories they own or buy from in developing countries pay their workers substantially above locally prevailing wages. They also claim to maintain above-average working conditions (Sethi et al, 2000). There is a third force in this debate, which really complicates the issue. It consists of the governments of the developing countries, such as India and Brazil. They openly resent what they describe as the newly-formed concern on the part of Westerners with the working conditions in their countries. They see that interest as a poorly disguised form of protectionism designed to keep their products out of the markets of the advanced economies. This cynical attitude was conveyed by Youssef Boutros-Ghali, Egypt’s trade minister, “Why, all of a sudden, when Third World labor has proved to be competitive, do industrial countries start feeling concerned about our workers?” (Sanger, 2001). In terms of action on globalization matters, most unions and many environmental organizations insist on making labor and environmental standards a part of any new international trade agreement. Products produced in violation of the standards would be barred from entering other nations. The opposition to that approach is hardly limited to teachers of international trade theory. The most vehement opponents are business interests in the developed nations and governments in developing
106
M. Weidenbaum
countries. The result is often a standoff almost ensuring the failure of any new round of substantial trade liberalization. 4. Strengthen the International Labor Organization (ILO) One global organization that warrants more attention is the ILO. It is the only international agency in which labor is fully represented. Yet unions are reluctant to use the ILO to enforce international labor standards—and for good reason. When it comes to ensuring compliance with the enlightened standards it adopts, the ILO has been a paper tiger. Worse yet, the U.S. Congress has not gotten around to approving all of the four “core” labor standards the ILO has promulgated—the right to form unions, ridding the workplace of discrimination in employment, and eliminating child and forced labor. Ironically, compared to our own detailed and pervasive labor laws and regulations, the core ILO standards are basic but far more limited. We must acknowledge, however, the problems that have arisen from the bureaucratic language that has found its way into the ILO process. However, this issue is not an insurmountable obstacle, but does require some significant high level attention. The U.S. Congress should quickly endorse all four of the ILO core labor standards. To follow up, the United States should take the lead in urging the other industrialized nations to join us in providing adequate resources and support to the ILO. The focus should not be on expanding the ILO staff. Large bureaucracies are notoriously unresponsive. Rather, funds should be increased for such enlightened activities as the special program which provides financial assistance to very poor families in developing nations whose children are taken off factory employment. The idea is to enable those youngsters to stay in school. In so many cases, the families cannot afford to pay for such education on their own—or even to forego the money the children had been earning (Anker, 2000). 5. Give People a Voice Via the Internet There is a way of promoting adherence to the ILO labor standards without resorting to trade sanctions or other forms of compulsion. The ILO should post on the Internet the names of the countries that are not complying with the core labor standards. Such a “seal of disapproval” should be widely publicized. Consumers would then be encouraged not to buy goods made in those nations. This approach does not provide the entertainment of puppet-parading protestors against globalization. However, it may be more effective in the long run. To be successful, the information approach requires a citizenry that takes the pains to inform itself and then acts voluntarily on an individual basis. Given the widespread access to the Internet, such a consumer effort could be powerfully effective. There is some attraction for the idea of consumers making up their own minds rather than relying on the compulsion of government or even the intimidation of group pressure. As the various sides in the debate on globalization continue to harden their positions, any movement to the high middle ground becomes increasingly difficult. The development of a feeling of trust, or at least common understanding, is a badly needed precondition. Some movement along the lines suggested here should help.
Seeking Common Ground on Globalization
107
Meanwhile, one modest change should be made. The various participants in the often-heated discussions on globalization should moderate their vocabularies. So often the argument seems to be carried on between “greedy, profiteering monopolists” and “impractical free-trade theorists,” on one side, and “environmental whackos” and “corrupt union bosses” on the other. The introduction of a bit of mutual good will would surely help. In any event, the serious concerns generated by a more closely linked global marketplace must be faced. More of us need to understand and deal with both the dark side and the bright side of globalization. Real improvement in government policies on international commerce will not take place until we respond constructively to the genuine concerns of the other voices in the globalization debate.
The Future Global Economy As noted earlier, in the long sweep of world history, there have been cycles or waves of globalization. In the first decade of the twenty-first century, however, the trend of the past century to a more open global economy seems likely to continue, at least for the foreseeable future. The most positive factor in this regard may have very mixed effects for the companies operating in the United States and for their work forces—the rapid rise of the Chinese and Indian economies and their increased participation in the global marketplace. In the short run the relatively low costs of doing business in those two nations presents a difficult challenge for companies operating in industrial economies, such as the United States. The advantages of low cost industrial production in China are well known as is the attractiveness of the service activities in India. Chinese industry especially is moving up the production chain from merely making low-cost and low-tech components to designing and producing higher-tech and more sophisticated end products. Macroeconomic data do not suffice to explain the unprecedented trend underway. On a typical day, 10,000 new cars appear on the streets of Beijing. Every week, a new power station is built. One in every two building cranes in the world is standing on a construction site in China (Bruton, 2007). India began its current expansion later than China, but its longer-run prospects may even be brighter.As these two giant societies continue to expand rapidly, their emerging middle classes will represent new market potentials for the more sophisticated goods and services produced in the United States and other developed nations. Globalization in all its dimensions will continue to represent threat as well as opportunity to American companies, workers, and investors.
References Anker, R. (2000). “The Economics of Child Labour,” International Labour Review, 139(3), 259. Bordo, M.D. (2002). “Globalization in Historical Perspective,” Business Economics, 37(1), 20–29.
108
M. Weidenbaum
Bruton, J. (2007). “How Globalization Will Impact the Future of EU/US Economic and Political Relations,” Vital Speeches of the Day, 242–46. Decker, P. & Corson, W. (1995). “International Trade and Worker Displacement,” Industrial and Labor Relations Trade Review, 48, 758–74. “Globalization and Its Critics,” The Economist, September 29, 2001, p. S13. Marcal, L.E. (2001). “Does Trade Adjustment Assistance Help Displaced Workers?” Contemporary Economic Policy, 19(1), 59–72. Rodrik, D. (1997). Has Globalization Gone Too Far? Washington, DC: Institute for International Economics. Sachs, J. & Warner, A. (1996). “Economic Reform and the Process of Global Integration,” Brookings Papers on Economic Activity, No. 1, p. 36. Sanger, D.E. (2001). “A Grand Trade Bargain,” Foreign Affairs, 66–67. Sethi, S.P., Weidenbaum, M., & McCleary, P. (2000). “A Case Study of Independent Monitoring of U.S. Overseas Production,” Global Focus, 12(1), 142–47. U.S. Trade Deficit Commission (2000). The Trade Deficit, Washington, DC: U.S. Government Printing Office. Weidenbaum, M. (2004). Business and Government in the Global Marketplace, Seventh edition, Upper Saddle River, NJ: Pearson Prentice Hall, Chapter 11. Weidenbaum, M. & Hughes, S. (1996). The Bamboo Network, New York: Free Press.
Tax Reform Then and Now J. Richard Aronson
The U.S. tax structure has been reformed and simplified many times in the last 30 years but the need for reform and simplification remains greater than ever. (For a detailed summary of recent tax acts see C.E. Steuerle, 1992, 2004). Should the tax structure be built on a foundation of comprehensive income or on comprehensive consumption? (Examples of excellent tax reform analysis are Bradford, 1984 and Aaron and Gale, 1996.) Should wealth transfer taxes be inheritance taxes, estate taxes or repealed entirely? Should payroll taxes apply to all earned income or should the tax base be capped? These are all tax issues upon which reasonable people can differ. Unanimity of thought is not expected. But there is at least one area of tax reform upon which even those of opposing economic philosophies can agree. The corporation income tax needs to be fixed. By taxing all shareholders at a common rate, the tax causes inequities; and by taxing dividends both at the corporate and personal level, it creates inefficiencies. Corporate earnings should be taxed, either when earned (The Haig-Simons comprehensive income approach) or when consumed (The Kaldor, Fisher comprehensive consumption approach). It is the separate treatment of personal income and corporate income, however, that is the villain. In 1972 Eli and I published a paper with the title “How to Integrate Corporate and Personal Income Taxation” (Schwartz & Aronson, 1972). We criticized the separate treatment of these two taxes on the following four counts: 1. The corporate income tax is regressive between shareholders since the imputed burden is ordinarily a larger share of a poorer stockholder’s total income than of a richer stockholder’s income. 2. Variations in corporations’ dividend policy can have an independent effect on the relative tax burdens of individual shareholders who would otherwise be in the same bracket. 3. The corporate tax can distort the corporation’s choice between debt and equity financing. 4. The separate tax on corporation income may encourage retained earnings over dividends, thereby retarding capital mobility and distorting the individual choice between consumption and savings (Schwartz & Aronson, 1972, p. 1073). J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_10, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
109
110
J.R. Aronson
We also discussed and described several methods of integration and concluded that the withholding method suggested by the Carter Commission of Canada (Canada, 1966) was best. The topic remains relevant today. Although the two taxes are at least partially integrated, Harvey Rosen feels it appropriate to conclude Chapter 17 of his very popular textbook, Public Finance (Rosen, 2005), with a discussion of corporate tax reform through integration with the personal tax. In what follows I, once again, describe the various methods of integration and once again conclude that the withholding method is superior. (I’m sure Eli will agree—well, I’m reasonably sure.)
Methods of Integration The various techniques of integration are best described by introducing Joseph Pechman’s concept of the “added burden” of a corporate tax (Pechman, p. 180). In all examples and in Tables 1, 2, 3 and 4 we consider a situation in which the corporate tax rate is 35% and in which the firm has a dividend payout rate of 100%. The added burden of the corporate income tax is the difference between that amount of taxes an individual owes in corporate income tax plus the amount he owes in personal income tax on dividends less the amount of tax that he would owe if the individual’s share of corporate income were taxed at the appropriate personal income tax rate. Table 1 Added burden of the corporate income tax (1) Marginal Tax Rate on Income (%) 0 10 15 25 28 33 35
(2) (3) (4) (5) (6) (7) Corp. Inc. Corp. Tax Div. Personal Inc. Total Tax Burdena Added Burdenb Tax. ($)
($)
($)
($)
($)
($)
100 100 100 100 100 100 100
35 35 35 35 35 35 35
65 65 65 65 65 65 65
0.00 6.50 9.75 16.25 18.20 21.45 22.75
35.00 41.50 44.75 51.25 53.20 56.45 57.75
35.00 31.50 29.75 26.25 25.20 23.45 22.75
Source: Based on J. Pechman, 1987, pp. 180–85. a Col. (3) + Col. (5) b Col. (6) – Col. (1) × $100
Table 1 shows the anatomy of the added burden. Column (1) contains the marginal tax rate structure of the current personal income tax. Thus, in a world of complete separation of corporate from personal taxes, an individual in the 10% tax bracket who earned $100 million corporate income owes $35 in corporation taxes
Tax Reform Then and Now
111
Table 2 Integration through the dividend received deduction method (1) Marginal Tax Rate on Personal Income (%)
(2) Total Tax Burdena
(3) Added Burdenb
($)
($)
0 10 15 25 28 33 35
35 35 35 35 35 35 35
35 25 20 10 7 2 0
Source: Based on J. Pechman, 1987, pp. 180–85. tax + personal tax b Col. (3) – Col. (1) × $100 a Corporation
Table 3 Integration through the dividend paid deduction method (1) Marginal Tax Rate on Personal Income (%)
(2) Corp. Tax
(3) Personal Tax on $70
(4) Total Tax Burden
(5) Added Burdena
($)
($)
($)
($)
0 10 15 25 28 33 35
30 30 30 30 30 30 30
0.00 7.00 10.50 17.50 19.60 23.10 24.50
30.00 37.00 40.50 47.50 49.60 53.10 54.50
30.00 27.00 25.50 22.50 21.60 20.10 19.50
Personal tax on $100 0 10 15 25 28 33 35
0 0 0 0 0 0 0
0 10 15 25 28 33 35
0 10 15 25 28 33 35
0 0 0 0 0 0 0
Source: Based on J. Pechman, 1987, pp. 180–85. (4) – Col. (1) × $100
a Col.
(0.35 × $100) plus 10% of the $65 received in dividends. The total tax paid, corporate plus personal, shown in col. (6) is $41.50. If the $100 earned by the individual through the corporation were simply taxed at 10%, the taxpayer’s liability would be $10 rather than $41.50. The Pechman measure of added burden for this person
112
J.R. Aronson Table 4 Integration through the withholding method
(1) Marginal Tax Rate on Personal Income (%)
(2) Amount Withheld
(3) Refund
(4) Div
(5) Tax Burden after Credit
($)
($)
($)
($)
0 10 15 25 28 33 35
35 35 35 35 35 35 35
35 25 20 10 7 2 0
65 65 65 65 65 65 65
0 10 15 25 28 33 35
Source: Based on Schwartz & Aronson, December 1972.
shown in col. (7) is thus $41.50–$10.00 or $31.50. Table 1 contains similar calculations for taxpayers in each of the personal income tax brackets. The calculations reported in col. (7) expose the regressive nature of the added burden. The added burden falls from $35 for a person in the zero tax bracket to $22.75 for an individual in the top bracket of 35%. The question now is how to reduce or eliminate this added burden. The “partnership” method is perhaps the easiest to understand and at the same time the most radical. Under this scheme the corporation tax would be abolished and each stockholder would be taxed on his share of corporate income. In other words the individual would simply include his corporate income in his personal income tax return. The stumbling block to the partnership method has always been the uncertainty of the individual’s cash flow. Dividends cannot be counted on to provide the funds needed to meet the individual’s incremental personal income tax liability, and thus it would be necessary to sell assets or draw on other funds to meet the tax bill. Two methods of integration that leave the corporate tax in place but avoid the cash flow problem are the dividend received deduction method and the dividend paid deduction method. The dividend received deduction achieves integration by adjusting the tax liability associated with the personal income tax. At the extreme, suppose that all dividends were simply made free of taxation. The so-called double taxation of dividends would be eliminated. Nobody would owe tax on dividends, and as a result the total taxes paid by each person at any bracket level would be $35 (the corporation tax rate). The added burden of the person in the 10% bracket would be $25 (i.e. $35.00–$10.00) which is $6.50 less than when there was no integration (i.e. $31.50–$25.00). The Added Burden column in Table 2 shows that the dividend received deduction favors those in the higher tax brackets. In fact, people in the zero percent bracket receive no benefit while those in the 35% bracket find that their added burden has been completely eliminated. The dividend received deduction thus provides partial integration at best.
Tax Reform Then and Now
113
The dividend paid deduction method integrates the two taxes through adjustments to the corporate income tax. The idea is to allow the firm to deduct dividends in calculating its corporate taxable income. Table 3 shows the effect on the added burden of lowering corporation taxes. For example suppose the right to deduct dividends lowered the corporate tax burden from $35 to $30 per $100 of earnings. Column (5) shows the resulting added burden. When compared to the added burden shown in Table 1, each person’s added burden has been lowered by 14.3%. Under complete separation (Table 1) the person in the 10% bracket faced an added burden of $31.50, but under the dividend paid deduction the added burden is $27 (i.e. $37–$10), a reduction of 14.3%. Under the dividend paid deduction it is theoretically possible to eliminate the entire added burden for all taxpayers. If the deduction is high enough to eliminate the corporate tax entirely, the only tax remaining is the personal income tax; and as a result by definition there is perfect integration. Finally there is the withholding method. The concept is simple. Set the corporate tax rate at the level of the maximum personal tax rate and have the corporation act as a tax collector, remitting these receipts to the Treasury on behalf of the individual. Thus, in col. (2) of Table 4 we see that the amount withheld is $35 for each person no matter what his tax bracket. Each person now files his personal income tax form including his or her $100 of corporate income. Since $35 has already been withheld, there is no cash flow problem. The person in the 10% bracket owes $10 in tax, but $35 has already been withheld. This person is actually entitled to a refund of $25. Similarly the person in the 35% bracket owes no additional taxes; $35 is already withheld and $35 is owed. This person gets no refund. The withholding method solves the cash flow problem and integrates as completely as the partnership method.
Conclusion There has always been a strong case for integrating corporate and personal income taxes. Recent attempts have relied on the dividend received deduction approach, but as shown above this works in favor of those in the higher tax brackets and does little for those who face lower rates. A better case can be made for using the dividend paid deduction method, which lowers the burden of all taxpayers by the same proportion. Withholding, however, still remains the preferred technique, especially at this time when the corporate tax rate is equal to the maximum personal tax rate.
References Aaron, H. & Gale, W. (1996). Economic Effects of Fundamental Tax Reform, The Brookings Institution, Washington. Bradford, D. (1984). Blueprints for Basic Tax Reform, 2, Tax Analysts, Arlington, VA. Canada (1966). The Report Commission on Taxation.
114
J.R. Aronson
Fisher, I. & Fisher, H. (1942). Constructive Income Taxation: A Proposal for Reform. Harper and Brothers, New York. Kaldor, H. (1955). An Expenditure Tax, Allen and Unwin, London. Pechman, J.A. (1987). Federal Tax Policy, The Brookings Institution, Washington, DC. Rosen, H. (2005). Public Finance, 7, McGraw Hill-Irwin, Boston. Schwartz, E. & Aronson, J.R. (1972). “How to Integrate Corporate and Personal Income Taxation,” Journal of Finance, December, 27(5). Simons, H. (1938). Personal Income Taxation, University of Chicago Press, Chicogo. Steuerle, C.E. (1992). The Tax Decade, The Urban Institute Press, Washington, DC. Steuerle, C.E. (2004). Contemporary Tax Policy, The Urban Institute Press, Washington, DC.
Joseph A. Schumpeter: Not Guilty of Plagiarism but of “Infelicities of Attribution” Nicholas W. Balabkins
Throughout his academic life, Eli Schwartz’s field of specialization has been Corporate Finance. In 1962 he published his successful text, Corporate Finance, and in 2007, with John B. Guerard, Jr., he co-authored Quantitative Corporate Finance. For decades he has ruled supremely in this field at Lehigh’s College of Business; and today generations of his former students are plying the skills they acquired from him throughout the U.S. Professor Schwartz’s analytical finance skills straddle a broad base of the social sciences. He knows economic theory, the forms it takes and the role it plays in economic analysis. He also respects and uses the adjoining fields of economic theory, such as statistics, economic sociology, and economic history. Throughout his teaching career, Eli Schwartz has shown a unique interest in economic thought, methodology, and even “welfare state” economic systems, the Cinderella fields of mainstream economics of today. He has also taught economic thought, and one of his role models is Joseph A. Schumpeter and his concept of the innovator. For this reason, it seems only proper to devote this festschrift essay, which is meant to express my esteem for him, to the analysis of the initial emergence of Schumpeter’s innovator of 1912.
My Arrival at Lehigh University In the early spring of 1957, I received a message from Donald Tailby alerting me to a vacancy in Lehigh’s economics department. Don had been a Ph.D. classmate of mine at Rutgers and we had kept in touch. I did not hesitate long. I flew to Bethlehem and spent a day with members of the economics department. The chairman of the economics department at that time, Herbert Diamond, wanted to know about my background and my education at the University of Göttingen in West Germany and at Rutgers. He offered me a job as an assistant professor and almost doubled my salary. Thereafter, I remember meeting Professor Elmer C. Bratt, the nationally known business cycle and forecasting specialist. I also shook hands with Professor Finn B. Jensen, and a young and attractive and fast-talking Dr. Dudley Johnson. I also met a number of instructors who were all ABDs (all but the doctoral thesis) J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_11, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
115
116
N.W. Balabkins
working toward their Ph.D.s at universities other than Lehigh. These were young “eager beavers” so to say, but their minds were on holding classes, correcting exams, and finishing their dissertations. In the finance department, the Chair was held by Professor Frederick Bradford, well-known for his textbook in the field of money and banking. The young Eli Schwartz was also a member of the finance department. He asked me a few searching questions, we shook hands, and I left the campus in high spirits. I was touched, impressed with the vivacious and scholarly future colleagues at the College of Business, and with the incredibly attractive campus setting. Today, in 2008, only Eli Schwartz and Nicholas W. Balabkins remain at the College of Business. I started teaching at Lehigh in mid-September of 1957. My teaching load was 12 hours a week and classes were held on Saturdays as well. I taught Intermediate Price Theory, Money and Banking, and Principles of Economics. I shared an office with my friend Donald Tailby on the third floor of Christmas-Saucon Building, now the home of the mathematics department. I started doing research on my first day. Since I was not interested in economic theory, and since mathematics was not my forte, I decided to use my linguistic skills and read the Russian and German economic literature. In 1957 the Cold War was hot, and the Soviets had successfully launched Sputnik. Many American academics wondered what made the Soviets “tick.” In such a milieu, I decided to read the Great Soviet Encyclopedia, just published in Moscow. I surveyed the published Soviet versions of well-known Western economists, dead and alive. They were extremely hostile, and virtually all Western economists were called the “boot and saliva lickers of capitalists.” The write-up on Schumpeter was not as harsh. It appears in Volume 8 of the Encyclopedia as follows: A German bourgeois economist at Bonn University, and from 1932 on, at Harvard. In economic analysis, Schumpeter differentiates between Statics and Dynamics. The former studies equilibrium conditions, whereas the latter deals with conditions of transition from one equilibrium to another. Schumpeter takes the capitalist entrepreneur as the central figure of dynamic development; he sees the origin of profit in the new methods of production and in the lowering of production costs. On the basis of such reasoning, Schumpeter concludes, in a bourgeois-apologetic way, that without economic development there is no entrepreneurial profit, and that with no profit, no development. In addition to entrepreneurs, Schumpeter also praises bankers who, as creators of new capital, bring about favorable conditions for economic development.
In the concluding section of this account, Schumpeter is called “one of the most sincere apologists of capitalist monopolies in the bourgeois literature” (Bolshaya Sovetskaya Entsiklopedia, p. 235). By Communist standards, this was a remarkably restrained account of Schumpeter’s work. What obliged the Soviet ideologues, at the height of the Cold War, to be so civil to Schumpeter’s economic legacy? Throughout his life, Schumpeter had insisted that with the evolution of the capitalist system into “trustified” capitalism, innovations gradually become more and more mechanized and routinized. His innovative entrepreneur was more a representative of the “laissez-faire” type of capitalism of the nineteenth century than that of the mid-twentieth century (Schumpeter, 1947, p. 132). In contemporary big-corporation-dominated capitalism, innovations originate not with the individual
Joseph A. Schumpeter: Not Guilty of Plagiarism
117
innovator, but in corporate research labs and drawing boards. The disappearance of the innovator, Schumpeter predicted, would open the way to the eventual emergence of a centrist socialist system, in which all decisions would be made by political authority. The very success of capitalism, not its failure, Schumpeter warned, thus contained the seeds of its undermining and undoing. Furthermore, the growing hostility of intellectuals towards capitalism would inevitably destroy its economic and social base. Schumpeter was particularly concerned about the intellectual hostility to capitalist systems in democratic countries by radical students, politically correct faculties at universities and colleges, and media (Bottomore, 1981, p. 36). In “The March into Socialism,” an address delivered in December of 1949 at the annual meeting of the American Economic Association, Schumpeter was given a standing ovation at the end of his presentation (Schumpeter, 1950, pp. 415–25). A year after my arrival at Lehigh University, I published a survey article about the Western economists that was included in the Great Soviet Encyclopedia. Eli Schwartz has told me frequently since then how much he liked it. Over the next few decades, I read all I could about Joseph A. Schumpeter. The emergence of the innovator concept fascinated me and still does.
The Intellectual Seeds of Innovator in Czernowitz, 1909–1911 It was at the University of Czernowitz, from 1909–1911, where Schumpeter wrote the main part of his second book, Theorie der wirtschaftlichen Entwicklung, published in 1912. Richard Swedberg has called it Schumpeter’s “greatest work in economics” (Swedberg, 1991, p. 18). This volume introduced the concept of the “innovating entrepreneur.” An innovator is like today’s Bill Gates, who launched the now-ubiquitous PC and almost single-handedly spurred the economic growth that America enjoyed in the 1980s and 1990s until the dot-com bubble burst in the late 1990s. Since 1912 economists and journalists have assumed that the idea of the “innovating entrepreneur” popped from Schumpeter’s head like Athena from the head of Zeus. To this day, this idea still prevails; it is described in great detail in Thomas K. McCraw’s Prophet of Innovation, published in 2007. And Schumpeter himself, from 1912 on, claimed to be the sole progenitor of the innovator concept. It was his “baby,” and no one else’s. The purpose of the rest of this essay is to offer evidence to the contrary. It seems that Schumpeter did not acknowledge a work on the innovating entrepreneur published in 1867 by a fellow German economist. Who was that German economist? Do we have a case of “infelicity of attribution” on our hands? Before we answer these questions, it will be instructive to look more closely at Schumpeter’s years in Czernowitz and their impact on his career. Czernowitz was not Vienna, but it was by no means an intellectual desert in Nowheresville of the Austro-Hungarian Empire either. It has recently been called a multi-cultural pearl of Bukowina on the most eastern part of the empire (Braun, 2005). At one time, Czernowitz had newspapers published in six languages— German, Ukranian, Rumanian, Polish, Yiddish, and Hebrew—and printed in three
118
N.W. Balabkins
scripts—Latin, Cyrillic and Hebrew. German was the official lingua franca, but the various nationalities co-existed peacefully. The educated, German-speaking Jewish population exerted a dominant influence on life in Czernowitz and left an indelible mark on Joseph A. Schumpeter, consciously and subconsciously. We know from many sources that in Czernowitz Schumpeter was always busy, reading, taking careful notes, writing, teaching, and lecturing. We also know that he was bored to tears with social scientists who wrote and talked “unendliche Geschichte”—in English, endless history. As Schumpeter’s biographer, Robert L. Allen, writes, Schumpeter lacked intellectual stimulation in Czernowitz and placed all his chips on abstract theory and on the indispensable use of math in scientific economics. His role models included Walras and Pareto, two mathematically trained engineers who had transformed the body of classical economics into an algebra- and calculus-studded economics, or, more precisely, into mathematical models and systems of equations (Walker, 2006). Over time, Walrasian and Paretotype economics became known as “general equilibrium theory.” Today, in 2008, this form of economic theory is still the “bread and butter” of all graduate students in economics. For them, as the adage goes, “No math model, no economics for me.” From his student days on, Schumpeter strove to apply mathematics in economics, even though he was not trained as a mathematician himself. He was much taken by Walras, and, while in Czernowitz, he annoyed his colleagues by spreading the new mathematical gospel. However, Schumpeter was also a broad-based young economist who was fluent in four languages and who knew sociology, economic history, and statistics. But from the early days of his academic life, he disliked or even detested policy-making economics. He intended to become a scientist, not a politician. Even though much has been written about Schumpeter’s precocious academic years, his early life is still “a bit of a myth,” as Swedberg calls it (Swedberg, 1991, p. 22). Swedberg’s reference can be de-mystified a bit by examining Erich Schneider’s book, Joseph A. Schumpeter. Leben und Werk eines grossen Sozialökonomen (Scheider, 1970). Schneider was a student of Schumpeter in Bonn, where he earned what in German is called the Habilitation, or second doctorate, which gave him the right to teach at the university level (Bombach and Tacke, 1980, p. 47). In the fall of 1970, Schneider was a visiting professor of economics at the University of Pennsylvania. When I heard about his appointment there, I immediately invited him to speak at Lehigh’s seminar on economic growth. Prior to coming to Philadelphia, he had been lecturing in Brazil on economic development and on Schumpeter’s intellectual legacy. He accepted my invitation. His seminar was magnificent. Schneider was a great communicator, erudite, at home in mathematics, fluent in five languages, and a real charmer. My wife, Gladys, invited him to dinner, along with a few of my colleagues and quite a few graduate and undergraduate students of economics. Once at our home, my wife, who happens to be an internist and cardiologist, looked at Professor Schneider and noticed that he was very flushed. She directed the professor and his wife to the guest room, listened to his heart, and took his blood pressure. It was over 200! She asked Schneider
Joseph A. Schumpeter: Not Guilty of Plagiarism
119
to rest for a while, gave him some medication to bring down the blood pressure, and left him in the room with his wife. Once he had rested and his blood pressure had come down, he joined us for dinner. It was a great experience for us all. The dinner party broke up after midnight, and my students drove him back to Philadelphia. For a long time thereafter, they spoke of Professor Schneider’s visit at Lehigh. Alas, his heart gave him more trouble after his return to West Germany, and he died at the podium delivering a lecture a few weeks later, days before his 70th birthday. During his memorable visit to Lehigh, Schneider told me that he had recently published a book on Schumpeter. Apart from having been a professor of economics at the University of Kiel until his retirement in 1969, Schneider also served as the much respected Director of the Institut für Weltwirtschaft in Kiel. The Institut has for decades maintained a magnificent research library, and its economics and bibliographical collections have no peers in the world. For me, the Kiel library has since 1960 been the eighth wonder of the world. The library staff have prepared a unique and useful guide to what Schumpeter published over the years, giving us the titles of books, articles and book reviews, and the dates and names of publications. We learn that while he was in Czernowitz, Schumpeter in 1909 published two articles, one in German and the other in English, along with seven book reviews in German, French, and English. In 1910, he published three articles, all in German, of which one dealt with Leon Walras’s work. In 1910, Schumpeter also published nine book reviews, one of which brings us to the question of “infelicitious attribution” (Schneider, 1970, p. 90). I am referring to Schumpeter’s review of E. Fabian-Sagal’s book, Albert Schäffle und seine theoretisch-nationalökonomischen Lehren, which he published in Archiv für Sozialwissenschaft und Socialpolitik, vol. 31, 1910, pp. 271–72. In this review, Schumpeter wonders about the permanent legacy of Schäffle’s lifetime work, asking whether Schäffle left insights for the future generations of economists or merely dealt with the economic and social problems of his day. Schumpeter concludes the review by citing Fabian-Sagal’s belief that Schäffle was no genius and that his work would radiate no light over the centuries, but that he was nevertheless a great man (Schumpeter, 1910, p. 271). Thus did Schumpeter denigrate Schäffle with faint praise, but as we shall see, he would do much worse than this.
The Genesis of the Innovator Concept The genesis of the innovating entrepreneur has been discussed by many economists. Eduard März wrote that the innovator idea was not a “bolt from the blue,” and cited, as influences, Marx, Wieser, Sombart, Wirth, and Tarde among many others (März, 1991, p. 57). Robert L. Allen referred to Schumpeter’s English preface to a Japanese translation of the Theory of Economic Development, in which Schumpeter acknowledged his dependence on Walras and Marx (Allen, 1991, p. 114). In 1995, Nicolo De Vecchi discussed Schumpeter’s admission that Eugene von Philippovich,
120
N.W. Balabkins
Albert Schäffle, and Schmoller had influenced him (De Vecchi, 1995, p. 147). Indeed, said Allen, all the forerunners of the innovator idea “had been forming in Schumpeter’s mind for several years: they finally solidified in words in Czernowitz” (Allen, p. 100). Thomas K. McCraw agreed, writing in 2007 that it was in Czernowitz that Schumpeter had listed the five types of innovation defining the entrepreneurial act (McCraw, 2007, p. 73). To McCraw, Schumpeter’s Theory of Economic Development was an exemplary book, “an obvious tour de force, all the more noteworthy for having come from a 28-year-old reporting from Czernowitz, which seemed to be the middle of nowhere” (McGraw, p. 76).
Schäffle’s Unknown Legacy to Schumpeter For the young Schumpeter at the University of Czernowitz, Schäffle’s writings were next to useless. Albert Schäffle had been born in 1831 and died in 1903. He was a government official, journalist, professor of social sciences at the universities of Tübingen and Vienna, minister of commerce in the Austro-Hungarian Empire of 1871, author of 33 books, and decade-long editor of the Zeitschrift für die gesamte Staatswissenschaft (Sauerman, 1978, pp. 1–11). Except for three books, Schäffle’s vast literary output remains even today untranslated into English. Unlike the German-educated economists and social scientists before World War I, today’s American economists are not truly bilingual; they speak only English, and they write and think in mathematical terms. Schäffle was a broad-based social scientist, a jack of many trades, but he nevertheless produced a theoretical volume, virtually unknown anywhere today, called Die national-ökonmische Theorie der ausschliessenden Absatzverhältnisse, insbesondere des literarisch-artistischen Urheberrechts, des Patent-, Muster-, u. Firmen-schutzes nebst Beiträgen zur Grundrentenlehre. In English, the title would be: The Economic Theory of Restrictive Sales Conditions in Relation to LiteraryArtistic Copyrights, Patents, Trademarks and the Firm’s Goodwill, Together with Contributions to the Rent Theory. In this difficult and convoluted book, we find Schäffle’s ideas on the innovator scattered all over. Since the focus of this essay is on the emergence of Schumpeter’s innovator, the similarity of the concept in the writings of both authors requires some specific references to Schäffle’s weighty tome. For instance, on page 268 Schäffle differentiates between two kinds of innovations. “Commercial innovations” (kommerzielle Neuerung) consist of the discovery of new sources of inputs, that is, raw materials, and the development of new markets, that is, marketing outlets (neue Bezugs-oder Absatzquelle). “Industrial innovations” (industrielle Neuerung) refer to the introduction of new products and/or new production methods for an existing commodity. An entrepreneur introducing one of these innovations, Schäffle wrote, either singly or in combination with others earns “above normal rent” (i.e., above normal profit or what is referred to in today’s microeconomic textbooks as pure profits or supernormal profits). Schäffle emphasizes that these “above-normal” rents are
Joseph A. Schumpeter: Not Guilty of Plagiarism
121
of a temporary nature. The incentive to create something new, says Schäffle, leads to the emergence of monopoly rents (pure profits) and thus directs economic progress. On the other hand, the incentive to copy successfully introduced commercial and industrial innovations creates spread effects. This emulation process “democratizes” innovation-induced rents (i.e., pure profits) causing the former innovators to resume being ordinary business people earning normal rents or, in some cases, even to go bankrupt. The emulation, Schäffle writes, is marked by diligence, industriousness, and average economic ability. He characterizes the entrepreneur who successfully introduces industrial and commercial innovations as the “aristocratic” part of the process. Such men, he writes, possess sagacity, special talent, a bit of luck, and maybe even a touch of economic genius. It is in these passages where one finds striking similarities between Schäffle’s process of “creative destruction” and that of Schumpeter. Schäffle’s different forms of innovations—industrial and commercial—both result in “monopoly rents” (pure profits) that give economic progress its direction. But these profits, he warns, are of a transitory nature because they inspire emulation. Eventually all “monopoly rents” are competed away, and the former innovator is left with only what today is called a normal profit. Schäffle’s two types of innovations and their four manifestations reappear in Schumpeter’s innovator, accompanied by a new innovator who breaks up existing monopolies or creates new ones. Schumpeter sets his innovator in the framework of the equilibrium, disequilibrium, and back-to-equilibrium sequence. Schumpeter’s initial equilibrium is always referred to as a stationary economy. Disequilibrium is known as a period of innovation and growth. The return to new equilibrium is again referred to as a stationary economy. Schäffle does not employ such an attractive analytical framework. In Schumpeter’s stationary economy, there are no innovations and no pure profits; only normal profits prevail. An exogenous shock disrupts this comfortable routine, and a new production function appears in the form of new commodities, technological changes in the production of existing commodities, the opening of new markets, and the discovery of new sources of raw materials. In this new situation pure profits emerge, but they are not permanent. Schäffle’s emulators, in Schumpeter’s book, become “a swarm of imitators” who reduce the innovation to a new routine. For Schumpeter, as for Schäffle, routine is not entrepreneurship. Schumpeter’s innovator earns pure profits, while Schäffle’s earns monopoly rents; but both become victims of what Schumpeter calls the process of “creative destruction,” as the economy returns, once again, to a static status quo. It is this dynamic framework of John Bates Clark (1847–1938) of Columbia University that makes Schumpeterian analysis of the innovator so appealing and unforgettable.
How Did I Find Schäffle’s 1867 Book? In the summer of 1981, when I was a visiting professor at the University of Frankfurt, West Germany, I had a chance to visit with Erich Egner, my former
122
N.W. Balabkins
professor at my alma mater in Göttingen. We spoke about his numerous books and reminisced about the immediate postwar years. I had told him how he had fainted once in 1947, while lecturing, because of hunger. The food ration at that time hovered around 1,200 calories per day, and Egner gave his portion to his growing children. I also told him that I was writing an essay for a centennial lecture series on Schumpeter, to be held at Lehigh in 1983. He suggested that I read an essay on Schäffle published in 1961 by Professor Knut Borchardt at the University of Manheim. I had known Borchardt for a number of years; in fact, in the mid-1960s he had invited me to give a seminar at the University of Manheim. In his 1961 essay, Borchardt showed how Schäffle’s work in economic theory of 1867 portended the concept of the innovator that Schumpeter came up with in Czernowitz and published in his Theorie der wirtschaftlichen Entwicklung in 1912 (Borchardt, 1961, pp. 610–634). Almost 50 years have passed since Professor Borchardt uncovered Schäffle’s seminal ideas on the innovator; but in 2008, the entire economics profession still asks, “Who was Schäffle?”
Infelicity of Attribution on Schumpeter’s Part There are some striking similarities between Schäffle’s and Schumpeter’s prototypes of innovators. Schäffle wrote his volume in 1867, 16 years before the birth of Schumpeter. However, one is puzzled why Schumpeter did not acknowledge his intellectual debt to Schäffle. In fact, in 1914 in Schumpeter’s Epochen der Dogmenund Methodengeschichte, published very soon after the appearance of his Theory of Economic Development, he gave a highly unflattering assessment of Schäffle. This work, translated into English in 1954, as Economic Doctrine and Method, reads as follows: In fact it is difficult to place this powerful personality [Schäffle] correctly in a history of economic doctrine. He absorbed most of the trends of his age in the field of Sozialpolitik, history and sociology, and he also was a theoretical economist. In everything he did, he was successful in his presentation, original in his formulations and systematic in his treatment, but he was not really a profound scholar . . . [His books] “have had an extremely stimulating effect, but it would be difficult to quote from them even a single permanent result, even a single interpretation that was at once original and fruitful” (Schumpeter, 1967, p. 187).
Obviously, Schumpeter had read Schäffle’s published work before 1914, when his Epochen des Dogmen-Methodengeschichte was published. The similarity of Schäffle’s and Schumpeter’s innovators is startling, so why such a harsh assessment by Schumpeter of his intellectual predecessor? What is undeniably Schumpeter’s own is the attractive framework for the appearance and disappearance of his innovator. Alas, Schumpeter seemed intent on forgetting what he had learned from Schäffle. In 1927, he wrote once again that it was difficult to identify a trace of
Joseph A. Schumpeter: Not Guilty of Plagiarism
123
Schäffle’s influence on the works of contemporary economists (Borchardt, 1961, p. 611). Was Schumpeter’s negative attitude due to his opinion that the progress of economics as a science depended on vision and technique? He was all agog about Walras and Pareto-type economic theory. Was it possible he dismissed Schäffle for his lack of enthusiasm for mathematical techniques? Or was Schumpeter, by his sharp treatment of Schäffle’s work, intentionally covering up his intellectual tracks, so to speak? Professor Fritz Karl Mann of the University of Cologne wholeheartedly rejected Schumpeter’s condemnation of Schäffle’s intellectual legacy (Mann, 1932, p. 76). Schäffle wrote across the entire field of social sciences, whereas Schumpeter forever aspired to be an economic theorist, first and foremost. In an article titled “Unternehmer” published in 1928 in the Handwörterbuch der Staatswissenschaften, vol. 8, Schumpeter re-stated the salient features of his innovator model. At the end of the article, Schumpeter listed a number of books on innovations, including Schäffle’s Kapitalismus und Sozialismus of 1879, but not his 1867 theoretical treatise. In 2003 the 1928 article was translated from Schumpeter’s famously elaborate German prose into English by Markus C. Becker and Thorbjörn Knudsen. This new English-language text provides more grist for the renaissance mill of Schumpeter studies after his death in 1950. It seems strange that Schumpeter persisted once more in omitting Schäffle’s 1867 book, as he had in 1912. We do not know the reason for this repeated omission, but Professor Geoffrey M. Hodgson of the University of Hertfordshire felt compelled to comment that Schumpeter remained, as always, a magnificent enigma (Hodgson, 2003, p. 270). If one is a bit shocked by the way Schumpeter gave the back of his hand to Schäffle’s theoretical legacy of 1867, Schumpeter’s innovator nonetheless endures today as the standard explanation for the vitality of the entire capitalist system. The economics profession continues to neglect the powerful challenge to Schumpeter by Professor Israel M. Kirzner in his closely-reasoned volume, Competition and Entrepreneurship, of 1973. But that creates a vexing puzzle: Any freshman who took the same liberties in Economics 101 would, at the very least, raise the eyebrows of his professors. Maybe the time has come to give Schäffle a footnote in the growing literature about Schumpeter’s innovator. Another footnote goes to Professor Knut Borchardt of the Bavarian Academy of Sciences of Munich, Germany, for his resurrection of Schäffle’s 1867 book. Schumpeter’s one-time student, Wolfgang Stolper, had a unique take on his teacher’s reluctance to pay homage to his elders. Nothing really is new under the sun, Stolper suggested, adding, “There is a fundamental difference between creative and adaptive acts with fundamentally different consequences, even when paradoxically adaptive acts involve creativity. It remains true that adaption is also “creative” but the meaning of the words is subtly changed” (Stolper, 1994, p. 373). For Professor Eli Schwartz the time has come to accept what his longtime colleague has written for him and to decide whether Schumpeter is guilty of plagiarism, of “infelicities of attribution,” or simply “creative and adaptive acts.”
124
N.W. Balabkins
The title page of Schäffle’s 1867 book is provided for reference purposes.
References Allen, R.L. (1991). Opening Doors. The Life and Work of Joseph Schumpeter, vol. 1. New Brunswick, NJ: Transaction Publishers. Bolshaya Sovetskaya Entsiklopedia (Great Soviet Encyclopedia), vol. 48, p. 235. (Author’s translation). Bombach, G. and Tacke, M. (Eds.) (1980). Erich Schneider, 1900–1970, Gedenkband und Bibliographie. Kiel, Bibliothek des Instituts für Weltwirtschaft, 1980. Borchardt, K. (1961). “Albert Schäffle als Wirtschaftstheoretiker,” Zeitschrift für die gesamte Staatswissenschaft, vol. 117, pp. 610–34. Bottomore, T. (1981). “The Decline of Capitalism Sociologically Considered.” In Heertje, A., (Ed.), Schumpeter’s Vision. Capitalism, Socialism and Democracy after 40 Years. New York: Praeger. Braun, H. (Ed.) (2005). Czernowitz. Vienna, Steindl Verlag. De Vecchi, N. (1995). Entrepreneurs, Institutions, and Economic Change, The Economic Thought of J.A. Schumpeter (1905–1925). Hants, United Kingdom: Edward Elgar Publishing Ltd. Hodgson, G.M. (2003). “Schumpeter’s ‘Entrepreneur’ in Historical Context.” In Koppl, R. (Ed.), Austrian Economics and Entrepreneurial Studies. New York: JAI Press. Mann, F.K. (1932). “Albert Schäffle als Wirtschafts-und Finanzsoziologe.” In F. K. Mann, et al. (Eds.), Gründer der Soziologie, Jena, Verlag von Gustav Fischer. März, E. (1991). Joseph Schumpeter. Scholar, Teacher and Politician. New Haven, CT: Yale University Press.
Joseph A. Schumpeter: Not Guilty of Plagiarism
125
McCraw, T.K. (2007). Prophet of Innovation. Joseph Schumpeter and Creative Destruction. Cambridge, MA: The Belknap Press of Harvard University Press. Sauerman, H. (1978). “Die gesamte Staatswissenschaft im Spiegel dieser Zeitschrift.” Zeitschrift für die gesamte Staatswissenschaft. vol. 134, pp. 1–11. Schneider, E. (1970). Joseph A. Schumpeter. Leben und Werk eines grossen Sozialökonomen. Tübigen, J.C.B. Mohr (Paul Siebeck). Schumpeter, J.A. (1967). Economic Doctrine and Method. An Historical Sketch (translated by R. Aris), New York: Galaxy Books. Schumpeter, J.A. (1950). Capitalism, Socialism and Democracy, Third Edition. New York: Harper. Schumpeter, J.A. (1947). Capitalism, Socialism and Democracy. New York: Harper. Schumpeter, J.A. (1910). Review of Fabian-Sagal, Dr. Eugenie, Albert Schäffle und seine theoretisch-nationalökonomischen Lehren. Eine nationalökonomische Studie. In Archiv für Sozialwissenschaften und Sozialpolitik, vol, 31, p. 271. Stolper, W.F. (1994). Joseph Alois Schumpeter: The Public Life of a Private Man. Princeton NJ: Princeton University Press. Swedberg, R. (1991). Schumpeter: A Biography. Princeton, NJ: Princeton University Press. Walker, D. (2006). Walrasian Economics. Cambridge: Cambridge University Press.
Economics and the Tanakh—the Hebrew Bible Harriet L. Parmet
Introduction The Hebrew Bible or Tanakh offers economic utopias for building a theocratic society. This paper will present various economic situations and concepts found therein. These themes include social policies, the Sabbath, years of Sabbath and Jubilee, the agrarian society and the ethics of social justice. Quotes are offered from the Torah (also known as the Five Books of Moses) as well as the larger corpus of biblical texts, the Tanakh. Economics, as defined by the economist Lionel Robbins, is the social science which studies human behavior as a relationship between ends and scarce means which have alternative uses. As such, the relationship is a balance of choices and is a constant preoccupation of twenty-first century people in western societies. Economic questions overshadow our political and domestic lives. The modern economic system—consisting of free markets, transportation facilities, commercial institutions, banks, monetary systems and policies, industrial and corporate organizations, gross national products, salaries and wages—is a social institution all unto itself, manifesting dominance over other social institutions. The studies of economic histories and comparative economists within the last 50 years have demonstrated that ancient economics was a different phenomenon from what moderns think of as “economics.” Therefore one must acquire a set of conceptual and particular lenses, when reading the Bible, in order to perceive appropriately the nature and character of ancient economics. One dominant form of economic exchange in antiquity was the redistribution observed in political economics. Redistribution, as seen in the institutions of state and religious taxation, concerns the politically or religiously induced extraction of a percentage of local production, the store-housing of that product, and its eventual redistribution for some political end. An example of this is the Jewish tithe (Deuteronomy 14: 22–23), and the Jewish Temple can be seen as a redistributive institution primitive to the modern but nonetheless effective: Thou shalt surely tithe all the increase of thy seed, that which is brought forth in the field year by year. And thou shall eat before the Lord, thy God, in the place which He shall choose to cause His name to dwell there, the tithe of thy corn, of thy wine, and of thine oil,
J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_12, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
127
128
H.L. Parmet
and the firstlings of thy herd and of thy flock; that thou mayest learn to fear the Lord thy God always.
The intrinsic notion of redistribution is the collection of economic surplus to a central point and redistribution at another time.
Land and Money in Ancient Times The dominant industry in antiquity was agriculture. In all probability 90% of the populace was engaged in diverse ways on the land. Since land was the major factor of production, control of land was the chief political and religious question of ancient times. One must consider the political direction of the Jubilee laws in Leviticus 25: 23, 25–28: And the land shall not be sold in perpetuity; for the land is Mine; for ye are strangers and settlers with Me. . . . If thy brother be waxen poor, and sell some of his possession, then shall his kinsman that is next unto him come, and shall redeem that which his brother hath sold. And if a man have no one to redeem it, and he be waxen rich and find sufficient means to redeem it; then let him count the years of the sale thereof, and restore the overplus unto the man to whom he sold it; and he shall return unto his possession. But if he have not sufficient means to get it back for himself, then that which he hath sold shall remain in the hand of him that hath bought it until the year of Jubilee; and in the Jubilee it shall go out, and he shall return unto his possession. . . . and then contrast them with First Samuel 8:14: “And he will take your fields, and your vineyards, and your olive yards, even the best of them, and give them to his servants.”
and First Kings 21:1–29: The Naboth vineyard and King Ahab story, pp. 485–87. Labor was imbedded in other institutions, preeminently in kinship and household contexts. Industry was small scale conducted by families and their slaves.
Money and Antiquity Money did not appear in antiquity until after the eighth century B.C.E. and then without the same significance of today. Ancient money was generally used as bullion to store real values: to be used to pay mercenaries or taxes, to acquire land if legally possible (from other landowners who had it), or to create more dependencies through loans. It possessed a rudimentary form of abstract exchange; it did not eliminate the need for personal exchanges between trading parties. Money was mostly available to and in the hands of the political elite. Such capital was obtained through slave-based mining of raw materials and wars of conquest. Those who worked for a wage were regarded as dependent or degraded—almost as a form of slavery. Money facilitated long-distance commerce and trade. Because of its potentially disruptive social effects, long-distance trade was insulated from local commerce. For the sake of economic security, local oligarchs prevented the incorporation of local markets into long-distance trade networks (Polanyi, 1944).
Economics and the Tanakh
129
Land, as noted, was the most precious commodity for the ancient elite; control or ownership of land implied honorable lineage and was the material basis for household (economic) security. Therefore people in antiquity who acquired wealth through commerce or other means attempted to achieve respectability by investing in land. Ancient societies resisted placing a monetary value upon land to protect the position of long-standing elite groups and to discourage newcomers from obtaining respectability. The narrow margin between subsistence and starvation is attested everywhere in biblical literature (Deuteronomy 15:11): “For the poor shall never cease out of the land; therefore I command thee, saying: ‘thou shalt surely open thy hand unto thy poor and needy brother in thy land.’”
Biblical laws and moral injunctions are phrased to preserve the status of poor Israelites within the total covenant community (Exodus 20: 1–17, p. 98; 21–23, p. 99). The traditions of the Tanakh and the prophets demonstrate a profound concern for the economic conditions of the Israelites, but within the covenant. Economic concerns are always juxtaposed upon a larger social framework. Additionally the biblical concern with economics is always couched in agrarian terms. The Jubilee Year mentioned in Leviticus acknowledges the preeminence of land and family (Leviticus: 10–16, p. 167). The degraded status of labor and the prevalence of slavery in ancient Israel should also be noted (Deuteronomy 15:12, 16–17, p. 257). Moreover, one of the typical paths to slavery in the absence of war is discussed: increasingly asymmetrical power relations led to loss of ability to maintain one’s status in the community (Leviticus 25:39, p. 168). Biblical laws attempt to counteract the corrosive social effects of debts in covenant relationships (Deuteronomy15:1–2, p. 257). Interpersonal loans must never be made by charging interest (Exodus 22:25–27, p. 102). This zerointerest rule is not aimed to lead to a situation where lending is not practiced. The Torah gives strong orders to support lending money for the poor even without interest (Mangeloja, 2004). Generally, debts should never get so large that they endanger the community fabric. This manifest was an early lesson in economics and social reform. The political and redistributive economic machinery of the Israelite monarchy is illustrated by the Deuteronomic account of Solomon’s reign (First Kings 4:7, 22–23, p. 453). The report of First Kings makes clear that this economy was political, but with a domestic format: Solomon considered his kingdom his own household to do with as he pleased. He monopolized the lines of trade in such a manner that the notion of a free market applied to Solomonic commerce would be inappropriate (First Kings 10:2 p. 465, 11, 14–25, pp. 14–25). Solomon, as did most ancient monarchs, used commercial ties for his own aggrandizement. First Kings (9:15–22, p. 464) indicates the un-free quality accorded Israelites and non-Israelites under Solomon. Forced labor was reserved for the non-Israelites. However, the main cause for the division of the monarchy was the displeasure of the Israelites with the Solomonic order (First Kings 12:1–16, p. 469; Oakman, 1991).
130
H.L. Parmet
Vulnerability of Wage Earners Wage earners were one of the most vulnerable groups in biblical societies because they were not able to support themselves from the land. They were dependent upon those who did have power. Mosaic Law, however, assured laborers that their wages, which represented their livelihood, would be paid daily (Deuteronomy 24:14–15, p. 268). Job 31: 38–40, p. 1064 is a powerful expression of biblical justice. Verses 38 to 39 read: “If my land has cried out against me, and its furrows have wept together; if I have eaten its yield without payment. And brought those who labored on it to grief. . . .” The curse that Job brings on himself if he has thus sinned is stated in verse 40: “Let thorns grow instead of wheat and foul weeds instead of barley.” It is the laborer who is the victim of which Job declares himself innocent. For this sin, the land cries out because the owner has enjoyed its produce without justly compensating the laborers. This cry associated with justice is not a cry for help, but an appeal by someone who has suffered unjust treatment. It is interesting to note how the Bible makes perfect use of what the modern reader recognizes as personification. Such a cry of protest is connected to the mistreated wage earner elsewhere in the Bible. In Deuteronomy 14:15, the laborers who have not received the wage for that day “cry to the Lord.” Job 31 is similar to the prophet (Habakkuk 2:11). The stories of the building erected with exploited labor cry out. In Job 31, it is the land tilled by the laborers that cries out in protest. The imagery of Job (31:38–40) significantly contributes to this idea. The image of the land of the crier shows the sin to be so penetrating that the cultivated land has become saturated with it. Even if one gets rid of the laborers, the land itself stands in testimony against their abuse (Mott, 1993).
Eli Ginsberg’s Studies on the Economics of the Bible Eli Ginsberg’s Studies in the Economics of the Bible focuses on 1. 2. 3. 4.
Slavery The Sabbatical year The Jubilee year and Land proprietorship in early Hebraic society.
The various economic matters considered by the book are included in the Torah (Pentateuch) which constitute the first division or five books of the Tanakh, or Hebrew Bible, and which according to the Orthodox view were given to Israel by divine revelation in the desert of Sinai, prior to the settlement in Canaan. Accordingly, the author refers to the “laws” of slavery, of the Sabbatical year, of the Jubilee, etc., as nominal legislation. Yet he offers proof that the institution of the Sabbatical year was not observed in pre-exilic times—i.e. from the settlement
Economics and the Tanakh
131
in Canaan to the Babylonian Exile, 586 B.C.E. In writing of the law of Jubilee, he says: “There is not an iota of evidence that the Jubilee was ever observed . . . ,” leading the reader to assume the Jubilee never became a parcel of Jewish heritage. The Jubilee, as stated by Ginsberg, was impractical because “the terrific disparity between the patrician land-owning class had become too wide to bridge” (Ginsberg, 1932).
Significance of the Jubilee What is this Jubilee Year then, and what is its significance? The Jubilee (named after the jovel, a ram’s horn that sounded to herald the comprehensive remission) aimed to dismantle structures of social-economic inequality by releasing each community member from debt. In the 50th year (seven times seven plus one), the land should lie restful for yet another year, while every family returned to the equal share of the land it had been assigned when the people of Israel first came to the Promised Land. In this manner the rich were to be released from the extra land they had acquired, and the poor were to be released from their landless status. Even indentured servants, no matter where they stood in their own 7 year of service or life-long obligation, were to be released to return to their original family holding. All this suggests that Sabbath Economics applied at each harvest, not just every other generation. One sees that political leadership failed to implement the Year of Jubilee theory, indicating that high transaction cost is the chief flaw in such a policy. Since people are reluctant to surrender their property without compensation, those who own more than the median amount are certain to resist vigorously. Furthermore, the wealthier individuals are likely to have greater influence with the government than are the poor, and would be in a position to forestall any serious efforts at implementation. Therefore, equalization of landholdings through systematic redistribution is in all likelihood possible only through revolution, and periodic revolution is prohibitively costly in human and financial terms. What we do learn is that the concept of jubilee was to create a more equitable society by drawing rich and poor closer together. However, there were two socioeconomic requirements setting limits to poverty and wealth: everyone was entitled to work, and everyone was both entitled and obligated to rest. From Leviticus (19:9–10), we are commanded: “When you reap the harvest of your land, you shall not complete the harvest in the corners of your field nor shall you gather the gleaning of your harvest . . . or of your vineyard. . . . You shall leave them for the poor and the foreigner: I am YHWH your God.” In the book of Ruth, this command is carried out. Boaz acted with great generosity to Ruth the penniless widow, but he was not free to act ungenerously. It was the law of his society that guaranteed Ruth a place in gleaning his crops. Boaz could not order his regular workers to be economically sufficient. They could not harvest everything: not what grew in the corners of the field, nor what they missed the first time around. Social compassion was more important than efficiency. Through Ruth, the Bible affirms that in a decent society everyone is deserving of decent work for a decent
132
H.L. Parmet
income. Moreover, Ruth as well as Boaz was entitled to the Sabbath: time off for rest, reflection, and celebration. In both places where the Ten Commandments are recited (Exodus 20:8–11, p. 98 and Deuteronomy 5:12–15, p. 143) it is made abundantly clear that the whole family, all of the servants, and the “foreigner within your gates” are all to rest one day of every seven. In the deuteronomic version, the Bible explicitly says that the reason for this Sabbath rest is to remember what it was like to be a slave in Mitzraim (Egypt), where it was never possible to rest (Waskow, 1997).
Ethics and Economics Eli Schwartz, Professor of Economics at Lehigh University, when analyzing a disdain for the study of economics, which emphasizes production for profit and the accumulation of wealth held by some academics and other professionals in opposition to spiritual and social values, offers the following as his response: When examined carefully, the larger part of the rules of ethical behavior promoted by religion and subsumed in the economic discipline are not promulgated solely on a notion of self-sacrifice. They are subtly founded on a rational base of enlightened self-interest. So the ordinance in Leviticus, 19:35–36, “Ye shall do no unrighteousness in judgment, in weight or in measure. Just balances, just weights, a just ephad, and a just hin, shall ye have,” implies the whole corpus of business and industrial law. On a very precise and important level, the rule simply means that the measure of commodities sold in the market shall not be falsified. Furthermore, it just as surely implies a whole multitude of derived laws of behavior, such as that the numbers on financial statements be truthful, that the quality of goods not be falsified, that the laborer not charge for work not done, and that full disclosure be given on the sale of a security. There is grave harm wreaked on the economy when the precepts of just weights and measures are extensively violated. An inordinate amount of time is consumed when each transaction has to be validated: both buyer and seller waste time and effort. The force of the doctrine holds for both commodity and financial markets. A prevalence of false financial reporting would make impossible the existence of a modern financial market. There could be no banks and no security markets, and it would not be possible to accumulate the amount of capital necessary for the financing of large scale efficient enterprises. A simple, short, pragmatic ordinance, carefully placed in a book devoted to rules of purity and sanctification, the precept of just weights and measures, is a basic foundation for a functioning efficient economy (Schwartz, 2005).
Professor Schwartz further writes: I do not think that anyone can point to a developed economy where falsification, dishonesty, cheating and breach of agreements are at the core of economic behavior and business transactions. In fact the prevalence of immoral behavior may be coincidental with the lack of economic development in many countries and regions. If the society as a whole has no learning of or understanding of moral strictures and guidance, no amount of laws or regulations can save it. We might take in a deeper figurative sense the instructions to Israel, 11, 13–17. If the commandments or guides that have been given are followed, the rains will come in due course, the land will be fertile and the people and their offspring will be fed and nourished. But if the basic moral strictures are breached, the land becomes parched and desolate and the sources of parnasa (livelihood) dry up (Schwartz, 1996).
Economics and the Tanakh
133
The Economics of the Sabbath The biblical standard of social and economic justice is grounded in the call to “keep the Sabbath,” a Hebrew word meaning to rest or stop working. Sabbath captures the double theme of the Creation story in Genesis: abundance and limits. And the archetypal story of hunger and bread in the wilderness (Exodus 16), sandwiched between two stories of thirst and water (Exodus 15:22–27, p. 94 and 17:1–7) plus the Manna story all illustrate YHWH’s alternative to the Egyptian economy (Exodus 16:6, p. 96). God “raining bread from Heaven” symbolizes cultivation as a divine gift, a process that begins with rain and ends with bread. The story narrates a test to see if Israel will follow instructions on how to “gather”—a symbolic word in traditional societies for harvesting (Exodus 16:4). The Israelites’ first lesson outside of Egypt concerns economic production. Moses’ instructions present us with three defining characteristics for this alternative economic practice. First, every family is told to gather just enough bread for their needs (Exodus 16:16–18). Contrasted with Israel’s Egyptian condition of oppression and need, here everyone has enough: “Those who gathered more had no surplus, and those who gathered less had no shortage.” In the biblical economy there is no such thing as “too much” and “too little.” Exodus 16’s “theology of enough” is underlined by the version of the Manna story in Numbers 11, in which the people’s persistent “cravings” are punished with a plague of “too much” (Numbers 11:33–34, p. 195). Second, this bread should not be “stored up” (Exodus 16:19–20, p. 94). Wealth and power in Egypt were defined by surplus accumulation. It is no accident that Israel’s forced labor consisted of building “store cities” (Exodus 1:11), into which the empire’s plunder and the tribute of subject peoples were gathered. The Bible understands that dominant civilizations exert centripetal forces, drawing labor, resources and wealth into greater and greater concentrations of idolatrous power (e.g., Tower of Babel; Genesis 11:1–9, p. 14). Therefore, Israel is ordered to keep wealth circulating through strategies of redistribution and not to concentrate on strategies of accumulation. The third instruction introduces Sabbath discipline (Exodus 16:22–30, p. 94). “On the sixth day, when they distribute what they bring in, it will be twice as much. . . . Six days you shall gather; but on the seventh, which is a Sabbath, there will be none” (Exodus 16:5, 26, pp. 93–94). Here we see that Shabbat is instituted before the covenant at Sinai. It is reiterated in ultimate terms at the conclusion of the covenant code: If the people do not practice Sabbath, they will die (Exodus 31:12–17, p. 115). Not only is Shabbat the crowning blessing of creation, it is also the “beginning and ending of the law.” Sabbath regulations are indicative of God’s strategy for teaching His people about its dependence upon the land as a gift to share equitably, not as a possession to exploit. The prescribed periodic rest for the land and for human labor means to disrupt human attempts to control nature and maximize the forces of production. Because the earth belongs to God and its fruits are a gift, the people should justly distribute those fruits, instead of seeking to own and to hoard them.
134
H.L. Parmet
The Manna story, in sum, illustrates human dependence upon the divine “economy of grace.” Sabbath observance means to remember every week this economy’s two principles: the goal of “enough” for everyone and the prohibition against hoarding. This vision is, of course, utterly contrary to economics as we know it. Our incredulity is rather humorously anticipated in the story itself: “Manna.” The Hebrew “Mah Zeh” from which Manna is derived, means “What is this?” (Exodus 16:15, p. 93). The egalitarian code of Exodus 23 extends the Sabbath cycle to a 7th year: “You shall let the land rest and lie fallow, so that the poor of your people may eat: and what they leave the wild animals may eat” (Exodus 23:10–11, p. 102). The Sabbatical year restores equilibrium by restraining the activity of “productive” members of the economy and freeing constraints upon those the economy has marginalized, both the disenfranchised (the poor) and the undomesticated (wild animals). The Sabbath rest commanded for the land and the laborer restores the primacy of the all-encompassing and integrated system of nature and forces humans to readapt to its limits. Rewording Rabbi Arthur Waskow’s union of ideas on this subject: the Shabbat portends the peace agreement ending the primordial war between ourselves and earth, which began as we left Eden—which came from a misdeed of eating and brought us painful toil and turmoil in our eating. The Deuteronomist goes even further, interpreting the Sabbatical year to include debt release (Deuteronomy 15:1–18, pp. 257–58). This interpretation was intended as a hedge against the inevitable tendency of human societies to concentrate power and wealth in the hands of a few, creating hierarchical classes with the poor at the very bottom. In agrarian societies like biblical Israel, the poverty cycle began when a family fell into debt, deepened when it had to sell its land in order to service the debt, and reached its conclusion when landless peasants could only sell their labor, becoming bond-slaves. The larger land owners acted as creditors and foreclosed, thereby adding to their holdings. The prophet Isaiah railed against this process of economic stratification by which wealthy creditors “add house to house and field to field, until there is room for no one but you” (Isaiah 5:8, p. 538). The Sabbath year debt release intends to safeguard both social justice (“there will be no one in need among you”) and sound fiscal policy (“creditor nations will not rule over you” (Deuteronomy 15:4–6). Anticipating the human tendency toward selfishness, the practical Deuteronomist specifically forbids people from tightening credit in the years immediately prior to the Sabbath remission (Deuteronomy 5:7–11, p. 256). The remission applies to debtor slaves as well, not only freeing them, but demanding that they be sent away with sufficient resources to make it on their own (Deuteronomy 15:12–17, p. 256). Whether or not the community will enjoy the blessing of the land is contingent upon its fidelity to this Sabbath discipline which like Deuteronomy, like Exodus and the Hebrew Siddur (Prayerbook), is a constant personal reminder of the Israelite liberation from Egyptian slavery (Deuteronomy 15:15, p. 256).
Economics and the Tanakh
135
Slavery in Israel The Torah gives Jewish men the right to sell themselves as slaves to fellow Jews, but they do not lose their status in the civil or religious community. Rabbinic literature emphasizes that rather than slaves, they are “indentured servants,” who for a specific period of time are not free to resign their employment, but who should be treated with respect (Leviticus 25:39). The sages tell us that “one who buys himself a slave, buys himself a master” (Abaye’, 220, C.E.). It is forbidden to have the Jewish man work merely to keep him busy or to set him to tasks without a specific time limit or useful purpose. Labor without a purpose and degrading work demoralize the human being and are therefore prohibited. Righteous employers take care that their employees are not overworked, but treated respectfully. King Solomon used a large number of workers when building the Temple in Jerusalem; but even in this sacred task, employees were allowed to work in shifts, so as not to keep them away from their families for too long a time. Jews may purchase slaves from surrounding nations and they become the property of their owners (Leviticus 25:44, 46, p. 169): Both thy bondmen, and thy bondmaids, which thou shalt have, [shall be] of the heathen that are round about you; of them shall ye buy bondmen and bondmaids. . . . And ye shall take them as an inheritance for your children after you, to inherit [them for] a possession; they shall be your bondmen forever: but over your brethren the children of Israel, ye shall not rule over another with rigour.
The Torah states clearly that the non-Jewish slave is his master’s property and is a heritage (but not so with Jews, because their true owner can only be God). The non-Jewish slave’s owner does not have the same restrictions as Jewish slave owners. The Talmud observes that the inheritance of non-Jewish slaves is a positive commandment, prohibiting a Jewish owner from freeing his non-Jewish slaves. If necessary, a slave may be freed as a performance of a mitzvah (a good deed). A slave may not be freed merely to please him; but if the owner will derive some benefit from freeing a slave, he may do so. If a Jew is sold as a slave to a non-Jewish resident of “Eretz Yisrael,” his kinsmen have the responsibility to redeem him, but without depriving the owner of his legitimate property rights (Leviticus 25:47–55, p. 169). One of the provisions of the Jubilee ordinance was the releasing of slaves. The slaves were free to return (with their family, if any) to their land of inheritance. This is a difference compared to slave release after 6 years of service, when families were not necessarily permitted to follow the male slave. Additionally, the slave in the Jubilee Year was allowed to retrieve his inheritance soil, but no other financial help was provided. All Jewish slaves must be freed, even if they have not worked the minimum of 6 years or if they have elected to remain with their masters after the 6 years. In Exodus and Deuteronomy, the Hebrew’s service as a slave is unqualified. He serves as a slave for 6 years. Israelites, as we have commented, enslaved both fellow-Israelites and non-Israelites. The laws for treatment differed for each group.
136
H.L. Parmet
Ordinances for treating Israelites are especially strict, for the exodus from Egypt was intended as the liberation of the Israelites from oppression. Thus, an Israelite could never be more than nominally a slave to other humans (Leviticus 25, pp. 167–69). To be a slave to a private citizen, an Israelite would have to act affirmatively, e.g. by stealing or borrowing and defaulting on a debt. His bondage would be for a limited time. Despite the justification, slavery was highly unpopular among the liberty-loving Israelites. After Solomon’s death, slavery of both Israelite and non-Israelites—those who were purchased in Phoenician slave markets or captured in war—diminished, apparently for economic reasons. Slavery was efficient when the free wage exceeded the slave rate (which would include subsistence and the costs of special provisions guaranteed to Hebrew slaves). The free wage rate tended to be higher when work was more agricultural and the ratio of land to labor was high. Less victorious in war, Israel did not capture many more heathen slaves after Solomon. In accord with the cycle of seven represented by the Sabbath, an owner could hold a Hebrew slave for only 6 years as stated. In the seventh he was free. When his ownership expired the master had to pay the Hebrew slave some of the fruits of his labor because he was entitled to them and because he would need some material goods to establish himself as a free man: “Thou shalt furnish him literally out of thy flock, and out of thy threshing floor, and out of thy winepress. . . .” (Deuteronomy 5:12–15,18, p. 257). As Hebrew slavery was somewhat benign, slaves might have opted to remain in the master’s household. They had a choice. If a slave elected to remain, the master would ceremoniously pierce the slave’s ear before a judge. At a slave’s choice, then, the slave would serve until his death (Deuteronomy 15:16, 17, p. 257; Exodus 21:6, p. 99). The non-Israelite had no choice. He could be kept forever, as long as the master did not abuse him (Leviticus 25:44–46, p. 169; Buchholz, 1988). In this manner, biblical law provided legal rights to the slave. Very simply, if an owner maltreated his slave, the slave was freed. One of the frequently quoted biblical sayings, paraphrased “An eye for an eye, a tooth for a tooth,” refers not only to direct retributive punishment, but also to a condition for freeing abused slaves: “And if a man smite the eye of his bondman, or the eye of his bondwoman, and destroy it, he shall let him go free for his eye’s sake. And if he smite out his bondman’s tooth, or his bondwoman’s tooth, he shall let him go free for his tooth’s sake” (Exodus 21:26, 27, p. 100). Israelite society thereby punishes the slave abuser by taking away his slave. Biblical law commands one not to deliver a fleeing slave. “Thou shalt not deliver unto his master a bondman that is escaped . . . thou shalt not wrong him” (Deuteronomy 23:16,17, p. 267). Capital punishment was ordered for those who kidnapped and then sold slaves or sold free men into slavery. Legitimate buyers of Hebrew slaves were required to deal in private, so that the social stigma of a public market would not debase the slave (Leviticus 25, p. 167). Hebrew slaves especially had to be treated as kin. A slave could attain a high position and could be adopted by the master. He could own property, buy his freedom, and marry a free woman. Biblical laws required the master to be responsible
Economics and the Tanakh
137
for the slave’s family. Should the slave be a father, the master must take the children into his care and maintain them (Leviticus 25, p. 167). Being a slave was not an attractive position, but the legal guarantees referenced above may have assuaged the pain. The concept of imago dei offers the source of beneficence demanded of owners and employers. “Did not He that made me in the womb make him?” (Job 31:15).
Conclusion This essay has explored biblical precepts relative to economics—notably slavery, money in antiquity, redistribution as political economy, the Jubilee Year, the Sabbath, the Sabbatical Year, land ownership, and the ethics of social justice pertaining thereto. What has been stated but not emphasized is the theological principle governing the study as enunciated by many scholars. The economics of the Bible, a religious guide as well as an historical and cultural text, tells us that the whole land system is not ours at all but belongs to the Supreme Being, God. This land was to be understood as both a divine gift and under divine ownership. Divine ownership is asserted for the sake of protecting both the family and the land with the family as the basic unit of Israel’s society and the basic unit of Israel’s land tenure. The family is also the inextricable locus of Israel’s relationship with God where teaching, catechism, and consecration occur. In this view God entered into a contract with the people of Israel. If they performed, He would perform. They were enjoined to “know [His] ordinances and do them” (Leviticus 25:18–19, pp. 167–68). In return, He would ensure that they “dwell in the land in safety.” If both parties performed, far from suffering because they decreased potential labor supply, land supply, and capital formation, the Israelites would dwell peacefully, secure from drought and sorrow. If the society and economic practice are based on that theological kind of very strict and absolute moral base, the gross utility (including material and spiritual) of consumers is maximized and poverty is minimized. In Studies of Economics of the Bible, as articulated by Ginsberg, there is no scarcity of material resources (if controlled by biblical institutions), and the needs of consumers can be met (if consumers follow the Torah’s commandments, and the covenant concept sublimates their personal desires). Contrary to modern economic theory, based on the assumption of the scarcity of resources and unlimited consumer needs, the economics of the Bible is based on God’s resources which are enough for all the true needs of His people (Wright, 1990). Thus economics has application from earliest times, and the Bible is still the chronicle that has validity today.
References Abaye’. Talmud, Babylonian Amora: Kiddushin, 20 A, circa 220 C. E. Buchholz, T.G. (1988). “Biblical Laws and the Economic Growth of Ancient Israel,” Journal of Law and Religion, 6(2), 406–408.
138
H.L. Parmet
Ginsberg, E. (1932). Studies in Economics of the Bible. Jewish Publication Society, Philadelphia. Mangeloja, E. (2004). “Economic Utopia of the Torah,” University of Jyvaskyla, Finland, p.14. Mott, S.C. (1993). “The Land Cries Out against Unpaid Wages,” Ariel, 16, 37. Oakman, D.E. (1991). “The Ancient Economy in the Bible,” Ariel, 21, 34–39. Polanyi, K. (1944). The Great Transformation. Boston, Beacon. Schwartz, E. (2005). “Leviticus and the Law of Markets,” Center for Business Ethics, Jerusalem, Israel. Schwartz, E. (1996). “Immorality Costs,” Center for Business Ethics, Jerusalem, Israel. The Holy Scriptures (Tanakh). Jewish Publication Society, Philadelphia, 1955. (All citations and pagination appearing within the parentheses refer to this English translation.) Waskow, A. (1997). “Holy Economics,” Soujourners, pp. 34–35. Wright, C.J.H. (1990).God’s People in God’s Land: Family, Land and Property in the Old Testament. William B. Eerdmans Publishing Co., Grand Rapids, Mich.
Muted Signals in Academe: Letters of Recommendation and Grade Inflation Robert J. Thornton
Introduction One morning at coffee several years ago (ok, so it’s more like twenty), Eli told a funny story about his writing of a letter of recommendation for a top student. After having written several very favorable paragraphs, Eli concluded the letter by saying that “He simply couldn’t recommend the student too highly.” When he looked at what he had written, Eli said, “I realized that one could interpret the concluding sentence as either high praise or damning! The sentence was completely ambiguous!” To Eli and his inadvertent stumbling upon this ambiguous recommendation, I owe a debt of gratitude. For in thinking later about other possible ambiguous recommendations, I was motivated to write a book, The Lexicon of Intentionally Ambiguous Recommendations (2004), that went on to receive widespread publicity (or maybe notoriety). The Lexicon contained hundreds of phrases like Eli’s original one that could be used (tongue-in-cheek, I hope) to write letters of recommendation for someone who is hopelessly incompetent, or lazy, or dishonest, or disagreeable, etc., without the writer having to worry about being untruthful or later being sued. A few examples: • • • • • •
I would waste no time in making this person an offer of employment. I’m sure that you’ll find her fired with enthusiasm. I recommend this person with absolutely no qualifications whatsoever. It’s been a good two years since she left our firm. You’ll be lucky to get him to work for you. I would not hesitate to give him sweeping responsibility.
There is a serious side to the subject of letters of recommendation, though, and it’s one of increasing concern. Letters of recommendation today are generally less frank, less honest, and less reliable than they were a few decades ago. A letter from a college professor to a graduate/professional school admissions committee or to a prospective employer is very unlikely to reveal any negative attributes of a student or to mention poor performance in class. Of course, this problem is by no means restricted to colleges and universities. In the U.S. this same reluctance to J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7_13, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
139
140
R.J. Thornton
reveal negative information in letters of recommendation is generally characteristic of employers when writing about former employees. In fact, many employers today refuse to provide any detailed references at all, instead simply giving information about dates of employment and job title. Certainly one reason for this is that such letters are no longer strictly confidential. Under the Family Educational Rights and Privacy Act (FERPA), the person about whom the letter of recommendation is written has the right to view the letter and take further action if the contents are not to his or her liking.1 Put simply, letters of recommendation—both in and outside of academe—have lost much of their value as a signal for conveying useful information to employers and graduate/professional schools. But there is another signal that, just as with letters of recommendation, has also become increasingly unreliable in recent decades: grade point averages (GPAs). This is a problem that is unique to academe, and it reflects what many refer to as “grade inflation” that characterizes most colleges and universities today. A few anecdotal examples (more evidence will be presented later): • About half of all undergrad grades given at Harvard are A or A–. And at commencement in June 2004, 91% of Harvard graduates received “Latin honors.” • In 1997 about 42% of grades at Princeton were As, compared to only about 31% in 1973 (Rosovsky & Hartley, p. 6). • At Lehigh University, the median GPA of graduating seniors in 1982 was 2.85. For the graduating class of 2004 it was 3.14. Just as with letters of recommendation, grade inflation means that employers and graduate/professional schools are no longer able to use GPAs as effectively to distinguish outstanding students from average or even mediocre students. Although the causes of grade inflation are many and differ from the cause(s) of bloated recommendations, the net result is the same. Grades have become an unreliable signal of student achievement in college. In this festschrift essay I first analyze the extent of these twin problems of unreliable letters of recommendation and grade inflation today. I then attempt to explain why these two phenomena have arisen. Using the signaling theory framework of Michael Spence, I then analyze the consequences of these two problems, not just on employers in their attempt to seek out reliable information, but also on students with respect to effort, incentives, and their desire to distinguish themselves from others. I conclude with some modest recommendations.
How Widespread Is the Problem? Letters of Recommendation In a recent article in the Chronicle of Higher Education(2000), Alison Schneider offered the following “advice” for academics:
Muted Signals in Academe
141
Need to write a recommendation for someone going on the academic job market? Trying to decipher an outside review in a tenure case? Here are a few helpful hints. “Good” does not mean good. It means hopelessly mediocre. “Solid” is shorthand for plodding and unimaginative. And “flashes of brilliance” is a nice way of saying that the scholar in question suffers from long languors of incoherence punctuated by random insights.
Ms. Schneider goes on to argue that in academe, some letters really are “too good to be true” and “puffery is rampant while evasion abounds.” And what is the reason that she offers? What would you expect, scholars ask, when a mild criticism or an off-the-cuff adjective can crush a career — the letter writer’s included. Lawsuits, reprisals, frayed relations with colleagues. There are good reasons, professors insist, why grades are not the only thing inflated in academe. (p. A-14)
While Schneider’s focus is on letters written by academics largely for academics, her observations reflect the situation outside of academe as well. A brief glimpse at the origin of letters of recommendation in the U.S. can give some insight on the problems that such letters pose today.2 Back in the nineteenth century, employers in many industries (like textile mills and railroads) made wide use of blacklists. Using this method, employers would send out names of disfavored workers (and sometimes names of approved workers—“whitelists”?) to other companies. But because of heavy worker protest, many states at the close of the century began to pass anti-blacklist legislation. In response, employers instead began to require letters of reference from previous employers, a practice that had formerly been used mainly by employers of domestic servants. However, the United States followed English common law here, whereby employers were not required by law to provide references. Nonetheless, the practice soon became very common, so much so that in a 1963 survey poor references were found to be among the four most common reasons why employers refused to hire a worker. And as late as 1980 another survey found that about 96% of employers checked references of some kind before making a decision to hire (Finkin, pp. 420–21). Paetzold and Willborn (1992) point out, however, that in recent years employers have become increasingly reluctant to provide references. They claim that modern human resources management thinking (they call it “folklore” (p. 123)) recommends that the only information that should be given about a former employee is that the person worked for the company at a certain position for a certain period of time. The rationale is that the less information provided, the lower the risk of exposure to workplace defamation suits and the attendant legal costs of fighting such lawsuits, even those that lack merit. In effect, those who think this way seem to be applying sound economic reasoning. An employer who provides a reference gains nothing, except for the appreciation of the party receiving the reference. The marginal gain is zero. But the marginal cost to the employer of providing such a reference, and in the process devoting the time and resources to make sure that the information provided in the reference is accurate while possibly facing a future lawsuit, can be considerable. Elementary economics tells us that the “production” of a good or service with
142
R.J. Thornton
positive marginal costs and small (or zero) marginal returns will be just that: namely, small (or zero.) On the other hand, the situation is the reverse for the employer who is the recipient of the reference letter. As long as the information provided is useful, this employer receives benefits from the reference letter but bears none of the costs associated with its production (Finkin, p. 422). As Paetzold and Willborn contend, however, the feeling that the costs to employers of providing reference letters have risen in recent decades may be wrong. Their examination of a sample of reported employment defamation lawsuits from 1970–1990 shows that: (1) there was no substantial increase in the relative frequency of such lawsuits over this time period; (2) plaintiffs were actually less likely to win defamation awards from such lawsuits in 1990 than they were earlier; and (3) plaintiff damages were no higher in successful employment defamation cases in 1990 than they were in 1970. Since the time period that Paetzold and Willborn studied corresponds closely to the period when employers became increasingly reluctant to provide employment references, they conclude that employers who are loath to provide references may be reacting to the high level of defamation damages reported by the press in a few “atypical” cases (Paetzold & Willborn, p. 124). To the best of this author’s knowledge, college professors have not shown the same reluctance to provide letters of recommendation for their students that employers have for former employees. Nevertheless, in writing letters that are less than candid, that fail to mention anything negative about the subject of the letter, and that are almost always favorable and often glowing, the end result is pretty much the same as providing no information at all. The return address on the envelope might as well be Lake Wobegone.
Grade Inflation What is grade inflation? A good working definition is that it represents an upward shift in student grade point averages without a corresponding rise in student learning or achievement. Unlike its economics counterpart of price inflation, where there is theoretically no upper limit to the price level, there exists a maximum GPA of 4.0 in most institutions. Some have therefore suggested that a better term therefore would be “grade compression” rather than grade inflation, with more grades becoming concentrated in the A and B range. Readers who are academics probably have some personal experiences concerning the existence and consequences of grade inflation—e.g., students unhappy about a test grade of B, dropping a course because their course average is “only” a C, everhigher proportions of students graduating with honors, etc. But what is the hard evidence about grade inflation? The following table and figure tell the story. Table 1 is based on the results of three surveys (Undergraduate Surveys) undertaken in the years 1969, 1976, and 1993. The first two surveys were sponsored by the Carnegie Foundation and examined student life on campuses in the U.S. The opinions and experiences of about 70,000 undergraduates in the 1969 survey and
Muted Signals in Academe
143
Table 1 Cumulative grade point average reported by undergraduates: 1969, 1976, 1993 Percentage Reporting Cumulative GPA
1969
1976
1993
2 5 11 17 19 23 18 7
8 11 18 22 15 15 10 3
13 13 18 21 13 12 7 2
7 25
19 13
26 9
Disaggregated A+ or A A– B+ B B– C+ C C– or below Aggregated A– or higher C or below
Source:Undergraduate Surveys (1969, 1976, 1993) from When Hope and Fear Collide by A. Levine and J. Cureton, Jossey-Bass Publishers, 1998.
25,000 in the 1976 survey were polled. The 1993 survey was undertaken by Levine and Cureton (1998) with the support of the Lilly Endowment and looked at a substantially smaller group of undergraduates (9,100). The numbers in the table speak for themselves. Over the 25 year period studied, the proportion of all grades given that were A– or above has nearly quadrupled while the proportion of grades of C or below has fallen by nearly two-thirds. Figure 1 has been taken from a website titled “gradeinflation.com” set up by Stuart Rojstaczer, formerly a professor of geology at Duke University.3 Rojstaczer’s data pertain to grade levels and trends at several dozen universities which he has tracked over time. Once again, the data in the figure are self-explanatory. Moreover, anecdotal information related to grade inflation abounds. In addition to those examples given earlier: • At a certain college in the South, 60% of freshman grades and 50% of sophomore grades are to be based on “effort.” What this means is that a student can still pass a course even though his or her academic performance in the course would have resulted in a failing grade. • Grade inflation is not restricted to the U.S. Observers report that Canada, the UK, Australia, India, and France are all experiencing grade inflation as well. • The incidence of grade inflation differs across disciplines. As Rosovsky and Hartley (2002) have put it, “Grades tend to be higher in the humanities than in the natural sciences” (p. 5). What are the various reasons offered for the recent rise in GPAs? Unlike the case of reference letters, where most observers agree that fear of litigation has been the
144
R.J. Thornton
Figure 1 Recent GPA trends nationwide∗ Source: http://www.gradeinflation.com/ ∗ Average undergraduate GPA for Alabama, California-Irvine, Carleton, Duke, Florida, Georgia Tech, Hampden-Sydney, Harvard, Harvey Mudd, Nebraska-Kearney, North Carolina-Chapel Hill, North Carolina-Greensboro, Northern Michigan, Pomona, Princeton, Purdue, Texas, University of Washington, Utah, Wheaton (Illinois), Winthrop, and Wisconsin-La Crosse. Note that inclusion in the average does not imply that an institution has significant inflation. Data on GPAs for each institution can be found at the bottom of this web page. Institutions comprising this average were chosen strictly because they have either published their data or have sent their data to the author on GPA trends over the last 11 years. Last update, March 17, 2003.
driving (if not the sole) force, a host of explanations have been offered for grade inflation. Table 2 lists the major ones. Notice that all of the reasons offered in Table 2 might explain the trend toward higher grades; however, not all of them are necessarily causes of grade inflation. When economists measure price inflation, they attempt to measure the increase in prices of a fixed bundle of goods and services. If the quality or quantity of any particular good or service in that bundle has risen, the price of that good or service may also rise. But in this case the higher price would not be said to reflect inflation. Similarly, if students are actually learning more or working harder in college, then factors 2 and 3 in Table 2, and perhaps 4 and 5 as well, could not be said to be responsible for grade inflation. In this case the “quality” or the “quantity” of what has been learned has risen. The case for higher grades reflecting higher levels of student achievement is not strong, however. Over the period of rising GPAs, SAT scores have not risen. Furthermore, more college students have had to take remedial courses, as Rosovsky and Hartley point out (2002, p. 6). Which of the possible causes of grade inflation outlined in Table 2 are considered to be most important? Much suspicion has fallen on cause 13, the now nearly-universal system of student evaluation of instructors. Not only are most college instructors regularly evaluated by their students, but most colleges also use the
Muted Signals in Academe
145
Table 2 Alleged causes of grade inflation 1. 2. 3. 4. 5. 6. 7. 8.
9. 10. 11. 12. 13.
Began in 1960s when faculty gave higher grades to keep male students from being drafted. Students today are smarter and work harder. Faculty have developed more effective teaching methods. More liberal course-withdrawal options exist, including the option to repeat courses. Colleges have tried to reduce student attrition, and such attrition is reduced with higher grades. Colleges have more non-traditional students (adults and life-long learners work harder). Traditional and tougher core requirements (e.g., math electives and foreign languages) have been reduced or eliminated. Greater use is made of adjunct and part-time faculty (higher grades → higher student satisfaction → higher probability that contracts will be renewed). The student-as-customer philosophy. “Wimpy” professors (fearful of angry students and parents). Greater emphasis on faculty research → less time for careful and objective grading. Changing philosophy of grading (use of grades as motivators rather than measures of achievement levels). Spread of student evaluations of professors.
results of student evaluations to some degree for contract renewal, salary raises, as well as promotion and tenure decisions. In a recent controversial book, Valen Johnson (2003) argues that there is a growing body of evidence that student evaluations are indeed the primary cause of grade inflation. He points to the fact that grades given and the ratings that students give to instructors are positively correlated. The “naïve” (and probably false) explanation for this correlation is that easy graders are more likely to gain the favorable approval of students. The “teacher effectiveness” view (also probably false, says Johnson) reasons that students learn more, and therefore receive higher grades, in courses that are taught by more effective instructors. The “attribution theory,” that Johnson seems to favor, holds that people commonly tend to take personal credit for success while they blame failure on others. In the classroom, this means that students who receive high grades are not more likely to give instructors higher evaluations. But students who receive low grades tend to blame instructors rather than themselves, and are thus more likely to give lower evaluations. In any case, whatever the reasons for grade inflation and its not-just-in-academe twin, bloated letters of recommendation, the net effect of both is to distort their value as signals of academic achievement, as predictors of future success in graduate and professional study, and as indicators of potential productivity as employees in the workplace. But so what, some may ask. What’s the problem if employers choose to say only nice things (or maybe nothing at all) about their former employees? And what’s the harm if professors elect to give most of their students high grades? Is there a downside? In the next section I discuss the economic costs associated
146
R.J. Thornton
with such practices. Using the signaling model/ theory of Nobel Laureate Michael Spence, I also discuss what many employers and students who have been adversely affected by the failure of these signals are doing as a result.
The Benefits and Costs of Reference Letters and Grades as Signals What are the advantages of frank and honest letters of recommendation over either their bloated counterparts or the situation where employers choose not to write any letters at all? Paetzold and Willborn provide a litany of the various benefits (pp. 124–28): • Letters of recommendation can be used to verify information provided by the candidate or applicant; • They may provide information not disclosed by other sources; • They can distinguish candidates who are qualified from those who are not; • Because former employers are the ones best able to provide accurate information about a person’s previous work experience, honest letters of recommendation can usually serve as the least expensive way of gathering such information. In the absence of such letters, employers must resort to other more expensive ways of procuring it—such as testing, more intensive and extensive screening and interviewing, trial employment periods, etc. • Honest letters of recommendation can also serve to motivate workers. Workers will be less likely to shirk if they know that their employment record will be accurately conveyed to other employers should they leave the company. This can in turn foster a sense of fairness among employees in the workplace, bolster worker morale, and raise worker productivity; • Finally, honest letters of recommendation can serve as a productivity-enhancing selection mechanism in the hiring process. This will be so as long as productive workers prefer to work for employers who evaluate and reward work fairly and communicate such information to other employers via letters of recommendation. It is also important to note that the costs of inaccurate letters of recommendation (as well as the refusal of employers to write such letters) are not incurred only by employers. There are also externalities—costs imposed upon society as a whole. As Paetzold and Willborn point out further: [The] trend toward not issuing references may serve to distribute income away from more productive employees and toward those who are less productive, thereby imposing costs on both employees themselves and the public. Additionally, reduction in the issuance of reference information would be expected to result in a higher job turnover rate (p. 127).
This last outcome arises because the matching of employees with employers would be more haphazard and costly.
Muted Signals in Academe
147
What about the costs of grade inflation? Grades, of course, serve several functions. They serve to measure the degree to which a student has mastered the subject matter taught. Grades communicate this information both to students and to external parties, such as employers, graduate and professional schools, as well as to parents. Grades can also serve as a motivator for students to put in greater levels of effort. Does grade inflation interfere with these functions? Although many (including this author) would answer “Of course it does,” not everyone agrees. Some observers contend that low grades serve to discourage students and can also destroy student self-esteem. Some would also argue that grade inflation does not necessarily distort the signals that grades provide to external parties, not as long as a relative grade ranking is maintained. More specifically, this line of reasoning holds that grades can still provide useful signals even if only grades of A, A–, B+, B, and B– are given. It’s not my intention to evaluate here the soundness of arguments that grade inflation is not as serious a problem as some claim. But no matter what position one takes, there are certainly economic costs associated with grade inflation. To put these costs into perspective, it is helpful to turn to a simple variant of the signaling model of Michael Spence (1973). Referring to Fig. 2, suppose that there are two types of college students—Group 1 (talented high achievers) and Group 2 (not-so-talented low achievers). Assume that individuals in Group 1 would have a marginal product (MP) in the labor market after graduation equal to W1 (the wage that they could expect to be paid), while individuals in Group 2 would have a marginal product (and wage) equal to W0 .
Figure 2 A job market signaling model
148
R.J. Thornton
Let us assume for simplicity that W1 is twice as high as W0 . Finally, let us also assume for the moment that there is no problem of grade inflation and that the high achieving students work for and thereby earn their high grades (H)—grades that in turn serve as effective signals to employers of their high-achiever status. Why should high-achieving students do so? After all, isn’t getting high grades “costly” in terms of time and effort? The answer is that it is worthwhile here (in terms of costs vs. returns) for such students to work hard and invest in the high GPA signal. This can easily be seen in Fig. 2. Suppose that the costs to the students in Group 1 of attaining higher GPAs (measured on the horizontal axis) is reflected by line CH . The attainment of the high GPA signal at point H would therefore “cost” high achieving students amount AD in terms of time and effort, but it would also return differential amount BE—the difference between the earnings of high achievers and that of low achievers. Clearly, the return from working hard to acquire the high GPA signal exceeds the cost of acquiring the signal. But what then is to prevent students in Group 2 from also working hard to secure high GPA signal H? Students in this group—being less talented—find it more costly to acquire higher GPAs. For them greater time and effort must be expended, measured by line CL . And to achieve high GPA level H would “cost” them amount CD. But since CD is greater than the expected returns to the higher GPA (again the difference in the earnings of high and low achievers), students in Group 2 will not make the effort to attain GPA level H. Now let us change the scenario. Let us assume that grade inflation sets in, with the costs to both groups of attaining a certain GPA falling. In Fig. 2, this would result in lines CL and CH both rotating downward in a clockwise direction. At some point Group 2 will find it worthwhile to invest the now-less-costly time and effort to attain high GPA level H. And since both Group 1 (talented high achievers) and Group 2 (not-so-talented low achievers) are now attaining high grade level H, grade inflation has effectively rendered GPAs useless as a signal to external parties such as employers. Although Fig. 2 and its accompanying explanation refer to grade inflation, the Figure can easily be modified to pertain to letters of recommendation as well. If minimal effort and poor academic performance on the part of members of Group 2 nonetheless result in the same good-letter-of-recommendation signal to employers, a similar outcome occurs. There are still other costs to grade inflation besides the resulting loss of GPA’s signaling power and the weakening of students’ motivation to work hard. With grade inflation, students are more likely to drop, and repeat later on, courses in which their probable grades are lower than they would like. Academics reading this article no doubt have experienced the unhappy student dropping a course at the end of the 12th week of instruction because they know there is no hope for an A. Although grade inflation is not the only reason, the average time it takes students to complete a bachelor’s degree in the U.S. is now close to 5 years.
Muted Signals in Academe
149
Grade inflation and student expectations of higher grades also cause some students to steer clear of demanding courses and majors and instead select courses and majors that are easy and present a greater likelihood of a high grade. This in turn puts pressure on instructors and departments with stricter grading standards to ease up in their own grading practices or else risk losing students and resources from the university. The fact (mentioned earlier) that grade inflation has not manifested itself to the same degree across disciplines means that it has the potential to create a sense of injustice and become demoralizing to many students. Surely some students in math or physics (two challenging majors that have not been as prone to grade inflation) who graduate without honors must feel some resentment when they see music majors (one of the majors most subject to grade inflation) graduating with high honors. There is yet another category of costs that must be mentioned. Because letters of recommendation and GPAs are no longer as reliable, employers must now use other probably more costly signals4 and channels of information to distinguish good job candidates from not-so-good candidates. It means, for example, that they are likely to spend more time interviewing job candidates, requiring extended or multiple “plant visits” (off-campus job interviews), and digging more deeply into candidates’ background and experience than they otherwise would have.5 It is no coincidence that the number of employment screening and reference checking agencies has risen in recent years. Finally, because Group 1 students (to return to our Spence signaling model in Figure 2) can no longer rely on letters of recommendation and higher GPAs to distinguish them from Group 2 students, Group 1 students must use other (also usually more costly) ways to set them apart. To mention just a few examples, more college students today pursue career-related internships (even though many are not paid) during the summer. (One side effect is a shortage of swimming pool lifeguards!) From my experience, more students also now pursue double (or triple) majors, seek entrance into special programs, elect to pursue a master’s degree after their bachelor’s degree, etc.
Conclusion Those who decry the lost value of letters of recommendation and GPAs as accurate signals are not without suggestions for change. In the space allotted here, I can mention just a few of these suggestions—and only briefly at that. With respect to letters of recommendation, Kenneth Dau-Schmidt (2001) has presented an interesting proposal. Why should an employer give a reference and take any risk of defamation liability when he cannot compel other employers to give him references, and thus gets nothing in return? The solution is for states to encourage employers to use the new information technology to create reference pools which all participating employers can access with the permission
150
R.J. Thornton
of job applicants. Once established, employers would see benefits from participating and giving job references because they would be sure of receiving useful employee references in return (p. 17).
Dau-Schmidt’s proposal would also restrict employer liability under defamation law while allowing the employees the right to view references on file and to correct erroneous information, just like under the Fair Credit Reporting Act. With respect to eliminating grade inflation, there have been numerous proposals put forward—not surprising in light of the multitude of reasons that have been offered for the existence of grade inflation. (See Table 2 again.) The proposals range from the complete elimination of student evaluation of professors (not likely to happen) to punishing professors for being “wimps” in their grading practices. (One university a few years ago cut the merit raises of several professors for giving out too many A grades.) Some very interesting policies have also been put into effect at a few institutions. Princeton now prohibits departments from giving more than 35% of their grades as A or A–. Harvard has begun to limit the number of students graduating with honors to 60% of the graduating class. One Harvard professor (Harvey Mansfield) has given students in his political philosophy class two grades—one an “official” grade that goes on the student’s transcript and the other a “private” grade giving a realistic and honest assessment of how well the student has done (2001, p. B24). Less extreme proposals include the encouragement of more “dialogue” among faculty to increase the degree of comparability in grading practices, expanding college transcripts to show the grade distributions in the courses that a student has taken, and the setting of a “standard” grade distribution for professors to follow (Rosovsky & Hartley). One of my favorite proposals was put forward by a colleague of mine at Lehigh several years ago. Professor Perry Zirkel of Lehigh’s College of Education proposed constructing an index that would be comprised of the grades given by a faculty member and his/her student evaluation scores (Zirkel, 1995). Zirkel offered to endow an annual cash award to the professor who scored the highest on his index. His proposal was voted down by the faculty.
Notes 1. Of course, sometimes this right is waived, as many of us have seen on graduate school recommendation forms with the boxes that indicate whether the candidate has done so. 2. The information in this section relies heavily on Matthew Finkin, “From Anonymity to Transparence: Screening the Workforce in the Information Age,” Columbia Business Law Review, volume 2000, number 3, pp. 403–52. 3. The information from gradeinflation.com had not been updated since 2003. In a personal communication with the author, however, Professor Rojstaczer has indicated that after a 6 year hiatus he has now updated and expanded the information in his grade inflation. com website. The updated information was too late to be included here, but as he notes “the grade inflation that began in the 1980s has yet to end.” 4. I use the word “probably” more costly here because these alternative signals would presumably have been used before had they not been more costly. 5. Some employers will only give recommendations over the telephone.
Muted Signals in Academe
151
References Dau-Schmidt, K. (2001). “Employment in the New Age of Trade and Technology: Implications for Labor and Employment Law,” Indiana Law Journal, 76(1), 1–28. Finkin, M. (2000). “From Anonymity to Transparence: Screening the Workforce in the Information Age,” Columbia Business Law Review, 2000(3), 403–52. Johnson, V. (2003). Grade Inflation: A Crisis in College Education. New York: Springer-Verlag. Levine, A. & Cureton, J.S. (1998). When Hope and Fear Collide, Portrait of Today’s College Student. San Francisco, CA: Jossey-Bass Publishers. Mansfield, H.C. (2001). “Grade Inflation: It’s Time to Face the Facts,” Chronicle of Higher Education, (April 6) 47(30), B24. Paetzold, R. & Willborn, S. (1992). “Employer (Ir)rationality and the Demise of Employment References,” American Business Law Journal, 30, 123–42. Rosovsky, H. & Hartley, M. (2002). Evaluation and the Academy: Are We Doing the Right Thing? Cambridge, MA: American Academy of Arts and Sciences. Schneider, A. (2000). “Why You Can’t Trust Letters of Recommendation,” Chronicle of Higher Education, 46(43), A14. Spence, M. (1973). “Job Market Signalling,” Quarterly Journal of Economics, 87, (August), 205–21. Thornton, R.J. (2004). The Lexicon of Intentionally Ambiguous Recommendations. Chicago: Sourcebooks. Zirkel, P.A. (1995). “Grade Inflation: A Problem and a Proposal,” Education Week (March 8), 28–30.
A Biography of Eli Schwartz
We recap Eli’s professional career briefly below and then dedicate the rest of the space for personal comments about Eli offered by several contributors to the volume. After serving for 3 years in the U.S. army during World War II (he still tells captivating stories about his days as a tank commander!), Eli went on to study for his PhD at Brown University. He joined the Lehigh faculty in 1954, where he taught for 37 years until his retirement in 1991. He was awarded the McFarlane Chair in Theoretical Economics in 1978. The courses that he has taught include Money and Banking, Public Finance, Interest and Capital Theory, and the History of Economic Thought. Over this period of time he has written or edited seven books, including Theory and Application of the Interest Rate (Praeger 1993), Trouble in Eden: A Comparison of the British and Swedish Economies (Praeger 1980), Corporation Finance (St. Martin’s Press, 1962) which he has recently updated with John Guerard (Quantitative Corporate Finance, Springer Verlag, 2007), as well as five editions of Management Policies in Local Government Finance (International City/County Managers Association) with Richard Aronson. In addition, he has written nearly 50 articles on a variety of topics in economics and finance, many of which have appeared in the very top journals (e.g., Journal of Finance, American Economic Review, National Tax Journal, Journal of Business, and Revue Economique). Since his retirement—his formal retirement, that is, for he has never retired in the true sense of the word—Eli has remained professionally active. As a charter member and vice president of the National Association of Forensic Economists, he has written a number of articles that have helped lay the foundations of this new field of economics that deals with proper methods for estimating tort damages. These articles have appeared in the Journal of Forensic Economics, Pennsylvania Bar Association Quarterly, and the American Bar Association’s Forum. Finally, he has been a regular contributor of articles—a featured columnist, we should call him— for the Allentown Morning Call. His pieces include “European Response to Uncle Sam,” “Today the Shah of Iran Doesn’t Look so Bad,” and “Isms and Theories that Grate on an Economist’s Mind.” Let us turn now to some personal reflections about Eli by some of the contributors. J.R. Aronson et al. (eds.), Variations in Economic Analysis, DOI 10.1007/978-1-4419-1182-7, C Martindale Center for the Study of Private Enterprise, Lehigh University, 2010
153
154
A Biography of Eli Schwartz
J. Richard Aronson Anyone who knows Eli knows that he is not capable of pronouncing a person’s name correctly. Pechman becomes Peshman, Gruber becomes Grubber, Margolis becomes Margilous, and so forth. So several years ago there we were, Eli and I making our way to the University of Hartford to meet with President Archibald Woodruff, who was associated with the Lincoln Institute of Land Policy. We were in search of research funds. For days in anticipation of our visit Eli had been referring to President Woodruff as President “Woodard” and each time I corrected him—“His name is Woodruff” I would say. I must have corrected him a hundred times. “Eli, his name is Woodruff.” Well, the day comes and we are at the University of Hartford and Eli says to the receptionist, “We are here to see Dr. Woodard.” Of course, I correct him, saying, “His name is Woodruff, Eli.” Then we meet his secretary and the scene is the same. By this time I have completely lost my patience. Finally out comes the President, and now wanting to take matters into my own hands I say “Nice to meet you Dr. Woodard,” and Eli says “Rich, his name is Woodruff, not Woodard.” George Bortz Eli Schwartz – A Fish Story I first met Eli Schwartz in 1950 when I went to Brown as a new assistant professor. He was completing his doctoral thesis in public finance, working for the economics department as a teaching assistant. Considering the small difference in age it is not unusual for advanced graduate students and new faculty to treat each other as colleagues. Eli stood out from the others for his fanatical devotion to fishing. He was fascinated by fish farming and the prospect of using science to create artificially a never-ending supply of fish. In Eli’s case this interest was accompanied by an intense desire to prevent the overpopulation of fish species and the determination to catch and eat as many as possible. Eli decided to recruit me in this enterprise. He offered to teach me how to fish and he provided the necessary equipment for casting a baited rod into the ocean waters that abut the shores of Rhode Island. Our wives accompanied us on this excursion, but I don’t recall that they shared any enthusiasm for the haul of flatfish we were expecting to bring home. My expectations of an easy capture started out well enough but ended in disaster. Casting a baited rod with a weighted line seemed easy. There is a sense of power standing on a dock with a large number of creatures looking up at you through the water just begging to get a taste of a bloody wriggling worm. What spoiled the fun, however, was intervention from hordes of sea gulls with ideas similar to ours. They would circle over the water looking for lunch, swooping down on unsuspecting fish. In my enthusiasm to catch my own lunch, I swept the rod over my head. The baited hook, the weights, and the line formed a magnificent arc, appearing to land far into the water. Then came trouble. The hook never got to the water. I had snagged a seagull by the wing. I pulled the line in, but what to do? There was this seagull splashing, croaking, screaming, and flapping its wings, and there was I trying to get the hook out of its
A Biography of Eli Schwartz
155
wing. No easy task, considering the shape of a fishhook. Eli was cool, consoling and full of sage advice. The ladies held the bird in place while the hook was removed. The bird was launched back into space, somewhat the worse for wear, but it flew away banking as it went. My own ego was as wet as my clothing. The rest is history. I gave up the sport of fishing, and went back to lunching on cans of tuna fish. Eli became a fishing master. Harriet Parmet I first met Professor Eli Schwartz in 1976 upon my arrival at Lehigh University to teach Hebrew and related Jewish subjects. I found him to be most helpful as well as an advocate of my endeavors. In those days in the absence of a Jewish Studies Center he attended and offered wonderful topical suggestions concerning Judaic colloquia. Eli read with critical eye my articles prior to submission for publication and I valued his wise and discerning counsel. He took genuine pleasure in seeing my materials in print. Additionally we share a love of the Yiddish language. Our relationship includes my enjoying many familial and social events together with him and his wife Renee. Eli takes pleasure in fishing, in playing bridge, tennis, and golf. An avid reader, he prodigiously absorbs all kinds of literature. Eli is also an avowed Zionist, championing the religion, history, and culture of the State of Israel. He feels that negative criticism should be relegated to those who live there. He has been a Visiting Professor at Tel Aviv University and has received travel grants resulting in numerous articles on Israel. Finally, Eli is also an accomplished poet. His aesthetic poetry runs the gamut of modern life, politics, human relationships, nature, HAIKU, the Hebrew Bible, Yiddish, and personal reminiscences. Among his poems should be mentioned “Demons Walked the Fields,” and “Ardennes, 1945.” It is a joy to honor this man for all seasons with this festschrift so richly deserved. Robert Thornton Back in the early 1970s, Eli wrote a short paper in which he laid out a “new theory” of the business cycle. His thesis was that fluctuations in the age at which toilet training is introduced across generations could affect the business cycle, as well as produce swings in the socio-political environment. In Eli’s words: “The individual who suffers a repressed upbringing and early toilet training. . . will tend to be anally retentive, savings oriented, socially conservative, and sparing in word and thought. In management-psychological nomenclature, we would say that he exhibits the accounting or credit-manager syndrome. On the other hand, the character profile of the adult who has been reared with relaxed toilet training will tend to be imaginative, non-conformist, more prone to risk-taking, and lavish of time, money and credit. His trait profile is more likely to be that of a sales manager, politician, or speculator.” Depending on which character trait tended to dominate the adults of a particular generation, the business cycle would be accordingly affected. The paper that he wrote was completely tongue-in-cheek, but written in a mock serious style. Hoping that the economics profession had a sense of humor, he sent it out to perhaps a half dozen journals but was met with a flat rejection each time. The
156
A Biography of Eli Schwartz
hilarious thing about it was that just about every editor and reviewer took his piece quite seriously. Some sample lines from various rejection letters: • “I regret it is not acceptable for publication. It is inappropriate for the Journal of Political Economy. • “Very neat, but not for us this time. By the way, is there a built in explosion developing in the economy with a secular increase in the frequency of “turesta” which breaks down all the natural restraints?” (from the Journal of Business). • “I recommend that Professor Schwartz be encouraged to present a much tighter formulation of his model. . .that would include empirical evidence” (from the Review of Social Economy). What finally happened to the article? Eli asked me if I had any ideas. I did—I discovered that a toilet-training “cycle” really does exist. (Who would have believed it?) Together we rewrote the paper and sent it to the Journal of Irreproducible Results where it was published in 1985. Let us conclude by giving special thanks to Harriet Parmet, who first suggested the idea of a festschrift for Eli, and to Lehigh’s Martindale Center for the Study of Private Enterprise for its support in putting together the volume. The Editors
Contributors
J. Richard Aronson is the William L. Clayton Professor of Business and Economics at Lehigh University and the Honorary Professor of Economics at the University of York (UK). He served as a Fulbright Research Scholar at the University of York in both 1978 and 1996. Professor Aronson is also the Director of Lehigh’s Martindale Center for the Study of Private Enterprise and a member of the Employee Retirement Study Commission of the Commonwealth of Pennsylvania. Finally, he is co-editor with Eli Schwartz of Management Policies in Local Government Finance for the International City Management Association, now in its fifth edition. Nicholas W. Balabkins is emeritus professor of economics at Lehigh University where he has taught economics, comparative economic systems, and economic development since 1957. He has written a number of books and monographs. His most recent book is Across the Social Sciences: Selected Essays in Political Economy, Frankfurter/M. Germany, 2008. Harold Bierman, Jr. is the Nicholas H. Noyes Professor of Business Administration at the Johnson Graduate School of Management, Cornell University. He is a B.S. graduate of the US Naval Academy, Annapolis; he received his M.B.A. and his Ph.D. from the University of Michigan. A Cornell faculty member since 1956, Professor Bierman formerly taught at Louisiana State University, the University of Michigan, and the University of Chicago. He has also taught for a semester at both INSEAD in Fountainebleau, France, and KUL in Belgium. He was a visiting research scholar at Cambridge University. He was a recipient of the Dow Jones Award from the American Assembly of Collegiate Schools of Business for his outstanding contributions to collegiate business education. He served as a financial consultant at Prudential Bache Securities in New York in 1986. His industrial consulting experience includes consulting for Corning Glass Corporation, Eastman Kodak, Sun Oil Company, Exxon Oil Corporation, IBM, Emerson Electric, and Xerox Corporation. He has written 29 books, including The Capital Budgeting Decision (with Seymour Smidt), Financial Accounting, Managerial Accounting, and Quantitative Analysis for Business Decisions, and more than 170 journal articles.
157
158
Contributors
George Borts has been a member of the Brown University economics department since 1950. He received his undergraduate education at Columbia University and his graduate training in economics at the University of Chicago. His thesis advisors were Milton Friedman and Tjalling Koopmans. He has spent his career teaching and doing research in economics, editing economic journals, consulting in legal proceedings involving economic regulation, and lecturing at universities and research groups in the United States and abroad. He has done research in the areas of transportation, the economics of regulation, regional economic growth, and international economics. For 12 years he was the managing editor of the American Economic Review. Recently, he has been most interested in problems of restructuring partially regulated markets such as health care, electric power generation, and rail freight transportation. His current interest is the immigration controversy in the United States. For recreation, he plays tennis (doubles only), reads historical novels (no bodice rippers), listens to classical music, and hums along at concerts. Ben Gillen is a Ph.D. candidate in economics at the University of California, San Diego. After obtaining his B.A. in Economics from Yale University, Ben moved to San Francisco where he spent 5 years in industry as a financial systems analyst for Symantec Software. In his economic theory research, Ben focuses on social preferences in signaling models and identification of auctions in the presence of nonequilibrium behavior. His research in finance looks at applying Bayesian methods to estimate expected returns and optimal portfolio weights, along with a focus on the role of cross-sectional firm characteristics in portfolio selection. John B. Guerard, Jr., Ph.D. is the Director of Quantitative Research at McKinley Capital Management in Anchorage, Alaska. He earned his B.A. in Economics from Duke University, M.A. in Economics from the University of Virginia, M.S.I.M. from the Georgia Institute of Technology, and Ph.D. in Finance from the University of Texas, Austin. John taught at the McIntire School of Commerce, the University of Virginia, and Lehigh University. He worked with the DAIS Group at Drexel, Burnham, Lambert, Daiwa Securities Trust Company, and Vantage Global Advisors, and served on the Virtual Research team at GlobeFlex Capital. John co-managed a Japanese equity portfolio with Harry Markowitz at Daiwa Securities Trust Company. While serving as the Director of Quantitative Research at Vantage Global Advisors (formerly MPT Associates), Mr. Guerard was awarded the first Moskowitz Prize for research in socially responsible investing. Mr. Guerard has published several monographs, including The Handbook of Financial Modeling (Probus, 1989, with H.T. Vaught), Corporate Financial Policy and R&D Management (Wiley, 2006, second edition), and Quantitative Corporate Finance (Springer, 2007, with Eli Schwartz). John serves as an associate editor of the Journal of Investing and the International Journal of Forecasting. Mr. Guerard has published research in the International Journal of Forecasting, Management Science, the Journal of Forecasting, Journal of Investing, Research in Finance, Research Policy, and the Journal of the Operational Research Society.
Contributors
159
John Hilley received his B.A. and Ph.D. in economics from Princeton University and taught at Lehigh University from 1976 until 1983. He has held several senior positions in the US government including chief of staff for the Senate Budget Committee, chief of staff for Senate Majority Leader George Mitchell, and senior advisor and head of legislation for President Bill Clinton. In the White House he was the principal architect and lead negotiator for the 1997 Balanced Budget Agreement which led to federal budget surpluses until 2001. After leaving government, he was the Executive Vice President of the Nasdaq Stock Market with responsibilities in the areas of international markets and strategic development. Harry Markowitz has applied computer and mathematical techniques to various practical decision-making areas. In finance: In an article in 1952 and a book in 1959 he presented what is now referred to as MPT, “modern portfolio theory.” This has become a standard topic in college courses and texts on investments and is widely used by institutional investors for asset allocation, risk control, and attribution analysis. In other areas: Dr. Markowitz developed “sparse matrix” techniques for solving very large mathematical optimization problems. These techniques are now standard in production software for optimization programs. Dr. Markowitz also designed and supervised the development of the SIMSCRIPT programming language. SIMSCRIPT has been widely used for programing computer simulations of systems like factories, transportation systems, and communication networks. In 1989 Dr. Markowitz received the John von Neumann Award from the Operations Research Society of America for his work in portfolio theory, sparse matrix techniques, and SIMSCRIPT. In 1990 he shared the Nobel Prize in Economics for his work on portfolio theory. Dr. Harriet L. Parmet is professor emerita in the Department of Modern Languages and Literature at Lehigh University where she has taught Hebrew since 1976. She specializes in modern Jewish literature, particularly the work of American Jewish women writers and such personalities as Penina Moise, Rebecca Gratz, and Haviva Reik. She has lectured widely in the Lehigh Valley. Co-founder of Lehigh University’s Jewish Studies Program and founder of the Jewish Colloquia Series, Parmet’s work has appeared in journals such as Midstream, Journal for Feminist Studies in Religion, Shofar, Holocaust and Genocide Studies, Feminist Teacher, Delos, Judaism,and Jewish Spectator. She is the author of The Terror of Our Days: Four American Poets Respond to the Holocaust. Robert M. Solow is the Institute Professor Emeritus at the Massachusetts Institute of Technology, where he has been a professor of economics since 1949. He taught macroeconomics and other subjects to undergraduate and graduate students until January 1996. Professor Solow studied at Harvard and received the Nobel Prize in Economics in 1987 for his theory of growth. For a number of years he served as a member of the Board of Directors of the Federal Reserve Bank of Boston and was Chairman of that Board for 3 years. He is past president of the American Economic Association and the Econometric Society, a member of the National Academy of Sciences, a fellow of the British Academy, and a former member of the National
160
Contributors
Science Board. He received the National Medal of Science in 2000. He has written articles and books on economic growth, macroeconomics, and the theory of unemployment, and occasional reviews in The New York Review of Books and The New Republic. Some of the books for which he is most noted include Capital Theory and the Rate of Return (1963); Growth Theory: An Exposition (1970); Made in America: Regaining the Productive Edge (with M. Dertouzos, R. Lester and the MIT Commission on Industrial Productivity, 1989); The Labor Market as a Social Institution (1990); and A Critical Essay on Modern Macroeconomic Theory (with Frank Hahn, 1995). He is currently Foundation Scholar at the Russell Sage Foundation, where he is participating in an extensive study of the sustainability of high employment. Robert Thornton is the MacFarlane Professor of Economics at Lehigh University. His areas of research include labor market discrimination, unions and collective bargaining, occupational licensing, and forensic economics. He has written or edited many books and collective volumes, most recently (2005) Fundamentals of Labor Economics (with Thomas Hyclak and Geraint Johnes) and Developments in Litigation Economics (with Patrick Gaughan). He is also the author of numerous articles which have appeared in such journals as the Industrial and Labor Relations Review, Journal of Human Resources, Oxford Bulletin of Economics and Statistics, and the Journal of Economic Perspectives. On the (much) less serious side, he has written the Lexicon of Intentionally Ambiguous Recommendations (Barnes and Noble Books, 2005) with the inspiration of Eli Schwartz. Ingo Walter is the Seymour Milstein Professor of Finance, Corporate Governance, and Ethics at the Stern School of Business, New York University, and the Dean of Faculty. He has taught at New York University since 1970 and served a number of terms as Associate Dean for Academic Affairs, Chair of International Business, Chair of Finance and, from 1990 to 2003, as Director of the New York University Salomon Center for the Study of Financial Institutions. Since 1985 he has also been affiliated with INSEAD in Fontainebleau, France, and serves as a consultant to various corporations, banks, government agencies, and international institutions and has authored or co-authored numerous books and articles in the fields of international trade policy, international banking, environmental economics, and economics of multinational corporate operations. Among his recent publications are Governing the Modern Corporation (New York: Oxford University Press, 2006) and “Can Microfinance Reduce Portfolio Volatility?” (Economic Development and Cultural Change, April 2009). Murray Weidenbaum holds the Mallinckrodt Distinguished University Professorship at Washington University in St. Louis, where he also serves as honorary chairman of the Weidenbaum Center on the Economy, Government, and Public Policy. He was the first chairman of the Council of Economic Advisers to President Ronald Reagan and the first Assistant Secretary of the Treasury for Economic Policy in the Nixon Administration. In 1999–2000, he chaired the Congressional Commission to review the trade deficit. Dr. Weidenbaum is the
Contributors
161
author of Business and Government in the Global Marketplace (7th edition) and the Competition of Ideas (in press). He is a fellow of the International Academy of Management and of the American Academy of Arts and Sciences. He holds a Ph.D. in economics from Princeton University.