Endogenous Regional Development: Perspectives, Measurement and Empirical Investigation (New Horizons in Regional Science)

Endogenous Regional Development

STIMSON PAGINATION (M2469).indd i

20/12/2010 15:12

NEW HORIZONS IN REGIONAL SCIENCE Series Editor: Philip McCann, Professor of Economic Geography, University of Groningen, the Netherlands and Professor of Economics, University of Waikato, New Zealand Regional science analyses important issues surrounding the growth and development of urban and regional systems and is emerging as a major social science discipline. This series provides an invaluable forum for the publication of high-quality scholarly work on urban and regional studies, industrial location economics, transport systems, economic geography and networks. New Horizons in Regional Science aims to publish the best work by economists, geographers, urban and regional planners and other researchers from throughout the world. It is intended to serve a wide readership including academics, students and policymakers. Titles in the series include: Entrepreneurship, Industrial Location and Economic Growth Edited by Josep Maria Arauzo-Carod and Miguel Carlos Manjón-Antolín Creative Cities, Cultural Clusters and Local Economic Development Edited by Philip Cooke and Luciana Lazzeretti The Economics of Regional Clusters Networks, Technology and Policy Edited by Uwe Blien and Gunther Maier Firm Mobility and Organizational Networks Innovation, Embeddedness and Economic Geography Joris Knoben Innovation, Agglomeration and Regional Competition Edited by Charlie Karlsson, Börje Johansson and Roger R. Stough Technological Change and Mature Industrial Regions Firms, Knowledge and Policy Edited by Mahtab A. Farshchi, Odile E.M. Janne and Philip McCann Migration and Human Capital Edited by Jacques Poot, Brigitte Waldorf and Leo van Wissen Universities, Knowledge Transfer and Regional Development Geography, Entrepreneurship and Policy Edited by Attila Varga International Knowledge and Innovation Networks Knowledge Creation and Innovation in Medium Technology Clusters Riccardo Cappellin and Rüdiger Wink Leadership and Institutions in Regional Endogenous Development Robert Stimson and Roger R. Stough with Maria Salazar Entrepreneurship and Regional Development Local Processes and Global Patterns Edited by Charlie Karlsson, Börje Johansson and Roger R. Stough Endogenous Regional Development Perspectives, Measurement and Empirical Investigation Edited by Robert Stimson, Roger R. Stough and Peter Nijkamp

STIMSON PAGINATION (M2469).indd ii

20/12/2010 15:12

Endogenous Regional Development Perspectives, Measurement and Empirical Investigation

Edited by

Robert Stimson Professor of Geographical Sciences and Planning, University of Queensland, Australia

Roger R. Stough Vice President for Research and Economic Development, NOVA Endowed Chair and Professor of Public Policy, George Mason University, USA

Peter Nijkamp Professor of Regional, Urban and Environmental Economics, VU University Amsterdam, the Netherlands NEW HORIZONS IN REGIONAL SCIENCE

Edward Elgar Cheltenham, UK • Northampton, MA, USA

STIMSON PAGINATION (M2469).indd iii

20/12/2010 15:12

© Robert Stimson, Roger R. Stough and Peter Nijkamp 2011 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2010927669

ISBN 978 1 84980 456 1

03

Typeset by Servis Filmsetting Ltd, Stockport, Cheshire Printed and bound by MPG Books Group, UK

STIMSON PAGINATION (M2469).indd iv

20/12/2010 15:12

Contents List of contributors Preface Robert Stimson, Roger Stough and Peter Nijkamp 1 2

3

4

5

6

7

8

9

10

Endogenous regional development Robert Stimson, Roger Stough and Peter Nijkamp The economist’s perspective on regional endogenous development Kenneth Button Endogenous regional theory: a geographer’s perspective and interpretation Michael Taylor and Paul Plummer Endogenous rural development from a sociological perspective Frank Vanclay Rural, urban or regional endogenous development as the core concept in the planning profession Edward Blakely Diversity and endogeny in regional development: applying appreciative intelligence Tojo Thatchenkery and Jessica Heineman-Pieper An exploratory approach to model determinants of endogenous regional growth performance Robert Stimson and Roger Stough A theory of entrepreneurial rents in endogenous growth: implications for regional innovation policies Zoltan Acs and Mark Sanders Foreign direct investment, knowledge assets and the economic geography of growth in the Asian BRIICS countries Tomokazu Arita, Chie Iguchi and Philip McCann Implications of European Union structural assistance to new member states on regional disparities: the question of absorption capacity Daniela Constantin, Zizi Goschin and Gabriela Dragan

vii xi

1

20

39

59

73

83

111

142

160

182

v

STIMSON PAGINATION (M2469).indd v

20/12/2010 15:12

vi

11

12

13

14

15

Endogenous regional development

Macroeconomic and territorial policies for regional competitiveness: theory and empirical evidence from the EU Roberto Camagni and Roberta Capello Endogenous employment growth and decline in Australian capital city statistical divisions Alistair Robson A case study approach to investigating local development initiatives in rural small towns in Victoria John Martin Economic development incentives and the measurement of local endogenous growth: is there a need for modeling adjustment? Terry Clower Regional growth and development theories revisited Roberta Capello and Peter Nijkamp

Index

STIMSON PAGINATION (M2469).indd vi

204

237

268

286 301

325

20/12/2010 15:12

Contributors Zoltan J. Acs is University Professor at the School of Public Policy and Director of the Center for Entrepreneurship and Public Policy at George Mason University, USA. He is also a Research Scholar at the Max Planck Institute for Economics in Jena, Germany, and Scholar-in-Residence at the Kauffman Foundation. He is co-editor and founder of Small Business Economics, the leading entrepreneurship and small business publication in the world. Acs is a leading advocate of the importance of entrepreneurship for economic development. Tomokazu Arita is Associate Professor in the Department of Social Systems and Management, Graduate School of Systems and Information Engineering, at the University of Tsukuba, Japan. After earning a PhD at the Regional Science Department of the University of Pennsylvania, he worked for twelve years at the Japanese Ministry of Land, Infrastructure, Transport, and Tourism (the former Ministry of Construction) before being appointed at Tsukuba. Edward J. Blakely is Urban Policy Professor at the United States Studies Centre at the University of Sydney, Australia, and formerly the Executive Director for Recovery in New Orleans. He has over 30 years as an expert and academic in the fields of regional and local economic development. He is a Fellow of the American Academy of Public Administration and a former Guggenheim Fellow. Kenneth Button is University Professor and Director of both the Center for Transportation Policy, Operations and Logistics and the Center for Aerospace Policy Research in the School of Public Policy, George Mason University, USA. He is also Professor in Civil Engineering at the University of Porto, Portugal and Visiting Professor in Applied Economics at the University of Bergamo, Italy. Button is a Fellow of the Chartered Institute of Logistics and Transport, a Fellow of the Institution of Highways and Transportation and an Academician of the Academy of Social Sciences. He is past President of the Transportation Research Forum. Roberto Camagni is a Professor of Urban Economics at the Politecnico di Milano, Italy. He is a past President of the European Regional Science Association. vii

STIMSON PAGINATION (M2469).indd vii

20/12/2010 15:12

viii

Endogenous regional development

Roberta Capello is Professor of Regional Economics at Politecnico di Milan, Italy. She is the current President of the Regional Science Association International, is Editor-in-Chief of the Italian Journal of Regional Science, and co-editor of Letters in Spatial and Resource Sciences. Terry Clower is Director of the Center for Economic Development and Research Associate Professor of Applied Economics at the University of North Texas, USA. He is Editor (Americas) of Regional Science Policy and Practice and a board member of the Council for Community and Economic Research. Daniela Constantin is a Professor of Regional Economics at the Academy of Economic Studies of Bucharest, Romania, where she is also Director of the Research Centre of Macroeconomic and Regional Forecasting. She is the President of the Romanian Regional Science Association and a member of the European Regional Science Association Council. She is the Director of the Romanian Journal of Regional Science. Gabriela Dragan is a Professor of European Integration at the Bucharest Economic University, Faculty of International Business and Economics and the General Director of the European Institute of Romania. She is a member of the Romanian Regional Science Association, the Team Europe, the European Commission’s panel of independent conference speakers, and Director of the Romanian Journal of European Affairs. Zizi Goschin is a Professor of Statistics, Econometrics and Regional Statistics at the Academy of Economic Studies of Bucharest, Romania, and a Senior Scientific Researcher at the Institute of National Economy of the Romanian Academy. She is a member of the Romanian Regional Science Association Council and the Editor-in-Chief of the Romanian Journal of Regional Science. Jessica Heineman-Pieper is an Assistant Professor in the Master of Science program in Organization Development and Knowledge Management in the School of Public Policy, George Mason University, USA. She has consulted to the National Institute of Mental Health (USA) on science policy, has worked as a management consultant with Booz Allen Hamilton, and is on the Executive Committee of the Critical Management Studies Division of the Academy of Management. Chie Iguchi is Assistant Professor of Global Strategic Management at Rikkyo University College of Business in Tokyo. She gained her PhD at the University of Reading Business School. Before joining Rikkyo University she previously held a full faculty appointment at Ritsumeikan

STIMSON PAGINATION (M2469).indd viii

20/12/2010 15:12

Contributors

ix

University, Kyoto, and was also a Visiting Research Fellow at Rutgers University. John F. Martin is Professor and Director of the Centre for Sustainable Regional Communities at LaTrobe University in Bendigo, Australia. His research interests include the role of local institutions in the sustainability of rural and regional communities across Australia. Philip McCann is Special Adviser to Johannes Hahn, the European Commissioner for Regional Policy and currently holds the University of Groningen Endowed Chair of Economic Geography in the Faculty of Spatial Sciences at the University of Groningen in the Netherlands. He is also a part-time Professor of Economics at the University of Waikato, New Zealand. He has won academic awards in five countries, is a keynote speaker in conferences around the world, and advises major international organizations and governments. Peter Nijkamp is at present Professor of Regional, Urban and Environmental Economics at the VU University in Amsterdam, the Netherlands. He has published extensively in the area of regional development and innovation. For many years, until 2009, he has served as President of the Netherlands Organization for Scientific Research (NOW). He is also a past President of the Regional Science Association International. Paul Plummer is a Professor at the University of Calgary, Canada. He has published widely on the foundations of analytical political economy, competing theories of spatial competition and the dynamics of regional growth and change. Alistair Robson has worked as an economist in both academia and government in Australia. Currently he is an economist in New South Wales Treasury. Prior to that he was a Research Fellow at the Institute for Social Science Research at the University of Queensland, where he also gained his doctorate. Mark Sanders is Assistant Professor in Economics at Utrecht University, the Netherlands, and research fellow of the Max Planck Institute of Economics in Jena, Germany. He conducts research in innovation economics, entrepreneurship and economic development, with a special focus on the role of entrepreneurial activity in the introduction of renewable energy technologies. Robert J. Stimson is Professor of Geographical Sciences and Planning at the University of Queensland, Australia, where he was also Director of the Urban and Regional Analysis Research Program in the Institute

STIMSON PAGINATION (M2469).indd ix

20/12/2010 15:12

x

Endogenous regional development

for Social Science Research, and was the Convenor of the Australian Research Council  Research Network in Spatially Integrated Social Science (ARCRNSISS) until 2009. He is a Fellow of the Academy of the Social Sciences in Australia, a past President of the Regional Science Association International, and the current Chair of the Applied Geography Commission in the International Geographical Union. Roger R. Stough is the Vice President for Research and Economic Development at George Mason University, USA, where he also holds the NOVA Endowed Chair and is an Eminent Scholar and Professor of Public Policy. He is a past President of the Regional Science Association International and holds an honorary doctorate from Jonkoping University in Sweden. Michael Taylor is Professor of Human Geography at the University of Birmingham, UK. He has held academic positions in New Zealand and Australia, and also worked for the Australian Federal Administration in Canberra. His research is focused on enterprise and regional economic development and particularly on the theory of the firm. He is Chair of the International Geographical Union Commission on the Dynamics of Economic Spaces. Tojo Thatchenkery is Professor and Director of the Masters in Science program in Organization Development and Knowledge Management at the School of Public Policy, George Mason University, USA. He is a member of the NTL (National Training Laboratory) Institute for Applied Behavioral Science, the Taos Institute, the Editorial Board of the Journal of Applied Behavioral Sciences, the Journal of Organizational Change Management (JOCM), and the book review editor of JOCM. Frank Vanclay is Professor of Cultural Geography at the University of Groningen, the Netherlands. He is a past President of the International Rural Sociology Association (IRSA).

STIMSON PAGINATION (M2469).indd x

20/12/2010 15:12

Preface The role of endogenous processes in regional economic development and as potential explanatory factors in accounting for spatial variations in the pattern of regional growth and decline has been attracting increasing interest from regional scientists and from regional development policymakers and practitioners. Among researchers there are a range of perspectives on endogeneity and endogenous regional growth processes. There are challenges in developing suitable measures of endogenous regional growth and in building and implementing operational models to analyse endogenous regional development and growth across spatial systems. It is also important for research to generate in-depth analysis of endogenous processes at the scale of the local region to enhance our understanding of the different contextual bases in which endogenous processes occur in local regional economic development. The chapters in this book address these important issues. They are based on papers presented at an international workshop on ‘Regional Endogenous Development: Measurement, Models and Empirical Investigation’ held at The University of Queensland’s Customs House facility in downtown Brisbane, Australia, in February 2008. It was hosted by the Australian Research Council Research Network in Spatially Integrated Social Science, with financial assistance from the Australian government’s Bureau of Infrastructure, Transport and Regional Economics. The workshop brought together researchers in regional science from around the world to address these issues. The workshop was one of many held over the last decade in which a small international group of about 20 to 25 established and emerging researchers come together to discuss topics relating to regional economic development, with an emphasis on innovation and entrepreneurship. The workshops have been held under the banner of the Tinbergen Institute (named after the Nobel Laureate in Economics) and are normally held in Amsterdam. The workshops are sponsored by the Tinbergen Institute; the Department of Spatial Economics at The Free University, Amsterdam; the School of Public Policy at George Mason University, Fairfax, Virginia, USA; and the Australian Research Council Research Network in Spatially Integrated Social Science. xi

STIMSON PAGINATION (M2469).indd xi

20/12/2010 15:12

xii

Endogenous regional development

This book is, we believe, a unique contribution to the growing literature that addresses endogenous issues in regional economic development in that it deliberately presents a series of different disciplinary perspectives on endogeneity and the nature and role of endogenous factors in regional development and growth. Those contributions come from the perspective of an economist, a geographer, a sociologist, a planner and a researcher in organizational management/social psychology. In addition, the book includes chapters in which researchers explore new frameworks to model endogenous regional development, focusing on a range of factors including regional resource endowments, entrepreneurship and institutional factors. Contributions to the book also include empirical investigations conducted at a variety of levels of spatial scale, from cross-national megaregional level of scale, to the national level of scale, to the city level of scale and the local small town level of scale. We thank the participants in the Brisbane workshop for their participation in the processes leading to the publication of this book, which we hope will attract the interest of researchers, students and practitioners in regional science and regional development. The financial and other assistance of the sponsors of the Brisbane workshop is gratefully acknowledged. And we are indebted to the Edward Elgar team for their encouragement to produce this book and for that publisher’s ongoing commitment to publishing research in regional science. Robert Stimson, Roger R. Stough, Peter Nijkamp February, 2010

STIMSON PAGINATION (M2469).indd xii

20/12/2010 15:12

1.

Endogenous regional development Robert Stimson, Roger Stough and Peter Nijkamp

INTRODUCTION Over the past few decades the emphasis in regional development theory has shifted from a focus primarily on exogenous factors to an increasing focus on endogenous factors. Traditional regional economic development approaches were erected on neoclassical economic growth theory, based largely on the Solow (1956, 2000) growth model. The emerging endogenous approaches – while recognizing that development is framed by exogenous factors – attribute a much more significant role for endogenous forces. In that context, a suite of models and arguments that broadly convey the ‘new growth theory’ have been directed towards endogenous factors and processes (see, for example, Johansson et al., 2001). Those developments are of great interest to regional development analysts and also to practitioners for several reasons, including the recognition of the importance of cities and regions in the development process, and also because they introduce an explicit spatial variable into economic development and growth theory, which was a mostly ignored element in neoclassical thinking. This evolutionary development is particularly significant as the importance of regions in national economies – and in particular the role of many of the world’s mega-city regions – has changed considerably since the 1970s as a result of globalization, deregulation and structural change and adjustment. Understanding these recently recognized processes of change are crucial for analysing and understanding different patterns of regional economic performance and in formulating and implementing regional economic development planning strategy. This book focuses explicitly on endogenous regional development, attempting to provide a range of disciplinary perspectives on the nature of endogeneity and what endogenous regional development is about, how it might be conceptualized, measured and modelled, and how empirical analysis taking an endogenous perspective may be conducted at different levels of spatial scale. 1

STIMSON PAGINATION (M2469).indd 1

20/12/2010 15:12

2

Endogenous regional development

THE NATURE OF REGIONAL DEVELOPMENT A Product and a Process Stimson et al. (2006, p. 4) observe that it is often difficult in regional economic development planning strategy formulation and implementation to match desired outcomes of regional economic development with the processes that create them. That gap in understanding the relationship between the apparent causes and effects of development poses a dilemma for those responsible for managing regional economic development in the making of policies and strategies, and the implementation of plans. The dilemma they face is how to achieve some form of congruence between desired outcomes and appropriate and acceptable economic development tools and processes. The dilemma is further compounded by the frequently unstable and changing nature of economic environments, where externalities or exogenous factors (such as exchange rates, new technologies, foreign competition) increasingly impact the decision-making processes that influence economic policy and strategy in cities and regions. Blakely (1994) emphasizes how regional economic development needs to be viewed as both a product and a process but often not by the same groups or actors in the development milieu. For example, economic agents that live, work and invest in regions are those most concerned with economic development outputs or products such as job and wealth creation, investment, quality of life or standards of living, and conditions of the work environment. Contrary to this view is the more process orientation of regional scientists, development planners and practitioners where concern focuses on the creation of infrastructure, labour force preparation, human capital and market development. So it is important when considering regional economic development to maintain an awareness of its product and process aspects. Regional economic development is also known in terms of quantitative and qualitative attributes. In that context, and with respect to the benefits it creates, our concern has typically been with the quantitative measurement of such factors as increasing/decreasing wealth and income levels, job creation or employment levels, the availability of goods and services and improving financial security. At the same time – and especially in recent times – our concern has also been with such qualitative considerations as generating creative capital, creating greater social and financial equity, achieving sustainable development, creating a spread in the range of employment and gaining improvements in the quality of life. Thus, the regional economic development process needs to be informed by both quantitative and qualitative information.

STIMSON PAGINATION (M2469).indd 2

20/12/2010 15:12

Endogenous regional development

3

This multidimensional aspect of economic development led Stimson et al. (2006) to propose the following definition of regional economic development: Regional economic development is the application of economic processes and resources available to a region that result in the sustainable development of, and desired economic outcomes for a region and that meet the values and expectations of business, of residents and of visitors. (Ibid., p. 6)

Changing Paradigms Policy for economic development and regional planning strategy has undergone a series of evolutionary changes since World War II, driven by different paradigms of economic thought as discussed by Stimson et al. (ibid., pp. 11–17). In the post-World War II era there was emphasis on Keynesian approaches to economic policy with an associated focus in regional development planning on master planning and then goals and objectives planning approaches. The era was very much embedded in the notion that differential regional development largely reflected differences in the comparative advantage/disadvantage of regions. Then, from the 1970s and into the 1980s there was a shift in economic policy to embrace monetarist thought and then rationalist economic thought in the 1980s and into the 1990s. That era was also associated with a refocusing in regional development planning on structure and strategic planning approaches. That reflected the notion that the enhancing of regional competitive advantage as a driver of regional development was an important objective. But over the last two decades there has been a further shift to a focus on the principles of sustainable development in regional development and planning, with planning strategies being based on an integrated approach whereby a strategy would seek to advance the capacity and capability of a region to better utilize its resource endowments through a focus on endogenous processes encouraging collaborative advantage across the private, public and community sectors. Those changing paradigms have shaped the way regional and local communities and people think and plan for the future. But much thinking on regional economic development still remains embedded in the paradigms of the 1970s, because of an inherent reluctance of many regions and local communities to proactively embrace change. Subsequently, as suggested by Stimson et al. (ibid., p. 11): many regions are not re-equipping themselves fast enough to compete effectively in the global age of business and technology of the post-industrial economy. To compete successfully in the global economy, regional organizations and

STIMSON PAGINATION (M2469).indd 3

20/12/2010 15:12

4

Endogenous regional development businesses need to understand the implications of the paradigm shifts occurring in economic policy and strategy, and to build the flexible strategic infrastructure to do so.

We now turn to a more detailed consideration of the emergence of the so-called ‘new growth theory’ in regional development, with its increasing emphasis on endogenous factors and processes.

THE EVOLUTION OF THE NEW GROWTH THEORY APPROACH During the 1980s – by which time the focus in economic policy paradigms had shifted from a focus on Keynesian thought and the focus in regional development thinking on the Solow (1956) model emphasizing exogenous forces, to a new paradigm represented by monetarism and economic rationalist thought – there had been a shift from concerns about developing a regional comparative advantage to developing a regional competitive advantage. There had also been a shift in regional development planning strategy from master planning and structural planning to strategic planning paradigms. Thus a new way of conceptualizing regional economic growth and development had begun to emerge, which today is known as the ‘new growth theory’. As early as the late 1970s, Rees (1979) had proposed that technology was a prime driver in regional economic development, and since then over the ensuing two to three decades the regional science literature has shown how technology is directly related to traditional concepts of agglomeration economies in regional economic development. Economists such as Romer (1986, 1990), Lucas (1988), Barro (1990), Rebelo (1991), Grossman and Helpman (1991) and Arthur (1994) sought to explain technical progress in its role as a generator of economic development as an endogenous effect rather than accepting the neoclassical view of long-term growth being due only to exogenous factors. In macroeconomic models of endogenous growth, technological progress was mainly seen as an endogenous process in an economic system, where knowledge is generally embedded in human capital that is enhanced through education, training, creativity and R&D. Thomas (1975) and later Erickson (1994), among others, showed how technological change was related to the competitiveness of regions. Norton and Rees (1979), Erickson and Leinbach (1979) and Markusen (1985) showed how the product cycle, when incorporated into a spatial setting, demonstrated that some regions could be seen as the innovators, while others become the branch plants or recipients of the

STIMSON PAGINATION (M2469).indd 4

20/12/2010 15:12

Endogenous regional development

5

innovation, and those might even then become innovators via endogenous growth. The concept of ‘innovative milieu’ was formulated to explain the ‘how, when and why’ of new technology generation. That notion linked back to the importance of agglomeration economies and localization economies that had been viewed as leading to the development of new industrial spaces (Scott, 1988; Porter, 1990). In particular, research by Krugman (1991, 1995, 1996) and Krugman and Venables (1996) led to a greater emphasis on knowledge as a tacit and primarily local good and the recognition of it as a driving endogenous self-reinforcing mechanism for regional development. There has also been an emphasis on the importance of investment in human capital and its role in regional development, as demonstrated in work by the OECD (2000, 2001). In addition, writers such as Fukuyama (1995) suggest that it is not just economic but also ‘value’ and ‘cultural factors’ – including social capital and trust – that are important in the rise of technology agglomerations as seen in the Silicon Valley phenomenon, where collaboration among small and medium-size enterprises through networks and alliances and links with universities forge a powerful R&D and entrepreneurial business climate. How to enhance regional economic development through engendering regional collaborative advantage (Huxam, 1996) began to be seen as a desirable component of regional development planning strategy. There has emerged as well a considerable emphasis on the role of leadership and institutions as factors that can enhance or even act as a catalytic effect in endogenous regional development, as demonstrated by Stimson and Stough (2009). As Rees (2001) has pointed out, technology-based theories of regional economic development need to incorporate the role of entrepreneurship and leadership, particularly as factors in the endogenous growth of regions, and it is the ‘link between the role of technology change and leadership that can lead to the growth of new industrial regions and to the regeneration of older ones’ (p. 107). Especially in the period since the mid-1970s, the processes of globalization have resulted in the emergence of an increasingly borderless economic world with increasingly unrestricted mobility of capital and labour and increasing freedom of trade in merchandise and services. Seemingly, the influence of the nation state was reduced in a world where cities and particularly mega-city regions assumed increasing importance as strategic hubs and as the drivers of creativity, innovation and entrepreneurial activity and as they increasingly became the dominant engines of economic growth (Knight and Gappert, 1989; Ohmae, 1995; Prud’homme, 1995; Florida, 2002). That created new stresses for both nations and for regions and their governments in developing strategies to find a competitive edge

STIMSON PAGINATION (M2469).indd 5

20/12/2010 15:12

6

Endogenous regional development

in a globalized economy and a highly competitive and rapidly changing world. There was a considerable shift in regional development planning strategies towards the notion of enhancing regional self-help. More recently the emergence of concern for achieving sustainable development has diversified the goals for regional development and intensified competitive pressures. And it is presenting new challenges for institutional reform, leadership and governance. Thus, the ‘new growth theory’ models have allowed for and indeed have implied the importance of both agglomeration effects (economies of scale and externalities) and market imperfections, with the price mechanism not necessarily generating an optimal outcome through efficient allocation of resources. And there has been a considerable emphasis on factors such as leadership, institutions, creativity, innovation and entrepreneurship, which might be regarded as the endogenous ‘intangibles’ that may enhance the performance of cities and regions. The processes of capital accumulation and free trade have not necessarily led to convergence of wage and price levels between regions, with positive agglomeration effects tending to often concentrate activity in one or a few regions in many nations through the self-enforcing effects that attract new investment. That process may be mediated positively by the endogenous ‘intangibles’ we have referred to. The ‘new growth theory’ has actually allowed for both concentration and divergence in regional development. Most importantly, as the spatial distribution of knowledge and its spillovers are now considered to be important success factors in regional development in framing and implementing regional development strategies, it is now seen as being crucial for a city or region to fully understand the nature of the geographical patterns of knowledge diffusion and the barriers to access to knowledge as they relate to creativity, innovation and entrepreneurship as being catalysts for employment and wealth generation (see, for example, Keeble and Wilkinson, 1999; Acs et al., 2002; Döring and Schnellenback, 2006).

THE CONTEMPORARY CHALLENGES We certainly live in a rapidly changing and increasingly competitive world in which uncertainty and risk are considerable. As discussed by Stimson et al. (2006), in many countries the challenge facing economic development planners in contemporary times has been how to formulate economic policy that will respond to both global dynamics and sometimes (or maybe even more often than not) the existence of a national vacuum in adopting a regionally oriented macro-policy.

STIMSON PAGINATION (M2469).indd 6

20/12/2010 15:12

Endogenous regional development

7

At one time regions were protected from outside competition, and to some extent their economies could be manipulated by national governments. But that ability has been overwhelmingly compromised as the economic rationalism pursued by many national governments left many cities and regions to fend for themselves. Many cities and regions have continued to look to higher levels of government for support and resources to provide economic direction and investment to stimulate economic development. Unfortunately, many cities and regions have failed to understand that globalization has left those higher levels of governments relatively weak when it comes to using their inherent power to apply economic and policy mechanisms to enhance the competitiveness of regional economies. In the literature a number of key themes have emerged regarding what constitutes regional growth and development and what drives regional competitiveness. Not surprisingly there have been differences of views among regional development scholars, and some of those differences relate to the relative focus given to the roles of exogenous forces on the one hand and the roles of endogenous processes and factors on the other. But there does now seem to be an almost universal realization of what Garlick et al. (2007) have referred to as the ‘institutional embeddedness’ of endogenous processes and factors in regional development. Of course exogenous factors are likely to remain important to a region’s economic performance and how it develops over time. But increasing importance is being placed on endogenous forces as determinants of a region’s competitiveness. However, regional economic development policy initiatives now tend to be more oriented towards measures that enhance local capacity and capability for a city or region to develop and cope with rapid change in an increasingly competitive global environment. While endogenous growth theory makes mention of leadership, entrepreneurship and institutional factors, little systematic analysis has occurred to thoroughly conceptualize or, even more, measure their roles as endogenous factors in the development process. But as discussed by Stimson et al. (2006), in the contemporary policy era of the last decade or two, it would seem that it has been more and more up to regions to develop and use their own devices to compete internationally in order to survive. Thus, it had become increasingly common in regional development planning strategy for there to be a reliance on endogenous processes, and typically that has been espoused in regional economic development policy. To do that a region would need first to have understood what the factors were that set the dynamics of the new economic age that had emerged in the late twentieth century. In the wake of the current global financial crisis and recession conditions, it will be

STIMSON PAGINATION (M2469).indd 7

20/12/2010 15:12

8

Endogenous regional development

interesting to see whether these much-changed macro-circumstances will set the conditions for a rethinking of that regional self-reliance philosophy and usher in a new era of innovation in institutional arrangements that might incorporate more interventionalist policies in regional development strategy planning. In the regional growth literature there is no doubt that the strategic importance of knowledge for innovation and entrepreneurship has been increasingly recognized. That has built on the notion of the ‘learning region’ as proposed by Simmie (1997). As discussed by Capello and Nijkamp (2009), in a neoclassical framework of analysis, long-range factors – such as education, R&D and technology – have played a critical structural role in the context of the spatial mobility of production factors, which could remove disparities (for example, in terms of per capita income) in the long run and, as a result, may equalize factor productivity across a nation’s regions. And in the endogenous growth literature we have seen how knowledge spillovers and institutional arrangements in local regions are widely acknowledged as factors explaining how knowledge spillovers are spread (as growth spillovers), with those knowledge spillovers representing pure externalities that produce noncompensating advantages for the receivers (Nijkamp and van Hemert, forthcoming). But Capello (2009) has pointed to a discrepancy between the private and social optimum that creates the emergence for ad hoc policy interventions. In the current economic climate of the global financial crisis and recession, Nijkamp and van Hemert (forthcoming) have suggested that in trying to capture the catalytic effect of creativity, innovation and R&D in generating knowledge growth spillovers: ‘more than ever there is a role for government in focusing strong and directed efforts to boost the translation of scientific ideas into useful technologies, and to reinforce the base of science skills that drives this innovation’ (p. 1). They go on to say: ‘Currently, there are different forces at play in the science domain that need attention and support from governments. Besides tensions between local and regional demands, the current crisis has highlighted the growing frictions between the individual and societal needs’ (ibid.). The challenges today include the need to revolutionize transport technologies, meet climate-change targets and secure diversity of energy supply. On a national level, that will require more directed research, education and training innovation to develop the required skills to enact the new technologies, and the active participation of industry in government–science relations to help encourage innovation. This changing socio-political environment, Hertz (2009) suggests, will require different research disciplines to work together more than ever.

STIMSON PAGINATION (M2469).indd 8

20/12/2010 15:12

Endogenous regional development

9

In the context of regional development, Taylor (2009) has referred to the ability to capture ideas and discoveries that flow from research as the main test of whether the UK can recover growth and prosperity. He says that at present the UK does not have the workforce needed to enact new technologies to address the challenges just mentioned, and that is also the case across many if not all countries. While it is a major policy challenge it does, nonetheless, represent an opportunity for local initiatives to be taken to boost investment in education and R&D, particularly in science and technology. The notion is that through what has been termed the ‘triple helix scenario’ (Etzkowitz and Leydesdorff, 1996, 1997, 2000), whereby investment in innovation and R&D inputs will lead to greater innovation outputs when they originate from local sources, cities and regions might be able to catalyse future economic growth. That notion affirms the existence of a spiral pattern of relations and links between the three major institutional actors in a local environment – industry, university and research institutes, and government – in which the education and research sector tends to have a critical part to play in the context of economic growth and regional development in the contemporary knowledge-based economy and in helping societies to address the technological and policy challenges they face with respect to issues such as climate change and achieving more sustainable development. Thus, as Nijkamp and van Hemert (forthcoming) say: ‘Concentrations of outstanding scientific facilities and activities are very important to create challenging and attractive working conditions and opportunities for talented people’ (p. 6). That reinforces what Florida (2002) has suggested in his work on the ‘creative class’ and the emergence of some cities as centres of creativity. Understanding the institutional barriers that mitigate against achieving this creativity and the associated economic dynamism of a city or region and how to unlock those barriers for the emergence of a ‘learning region’ is an obvious priority in regional development strategy planning if the pentagon model proposed by Nijkamp (forthcoming) is to be pursued. A big issue will be the degree to which regional development and growth will converge or diverge as a result of the ‘institutional embeddedness’ of endogenous processes to which Garlick et al. (2007) refer (and which we have discussed in this chapter) and what will be the nature of the ‘jumps and anomalies’ in urban and regional systems that Nijkamp (forthcoming, p. 6) refers to. Endogenous growth theory can help us to understand the complexities of a dynamic space-economy (including shock and bifurcations), but contextual drivers and government policies offer another cause of unexpected dynamics. Consequently, the evolution of complex spatial systems will partly remain an unresolved mystery.

STIMSON PAGINATION (M2469).indd 9

20/12/2010 15:12

10

Endogenous regional development

TOWARDS A SUSTAINABLE DEVELOPMENT MODEL In the contemporary context of a focus on sustainable development in regional economic development strategy it becomes even more necessary to place concerted attention on harnessing endogenous factors in pursuing regional growth and development. Towards that end, from the writings of Nijkamp et al. (1994) and Capello et al. (1999) it is possible to propose a production function for sustainable innovative development based on a pentagon model of five critical success factors (CSFs) as illustrated in Figure 1.1. The five elements of that model that need to be mobilized for the regional development process are: 1.

2.

The availability of productive capital (PC): this corresponds to neoclassical production theory where output is determined by the traditional production factors labour and capital. The presence of human capital (HC): this refers to the quality of labour input obtained by means of education, training or new skills (for example, in ICTs) and may be seen as a productivity-enhancing

PC

EC

HC SID

CC

SC

Source: Compiled by author Peter Nijkamp.

Figure 1.1

A pentagon model of creative forces for sustainable regional development

STIMSON PAGINATION (M2469).indd 10

20/12/2010 15:12

Endogenous regional development

3.

4.

5.

11

factor. Clearly a balanced distribution of human capital over people is of great importance. The access to social capital (SC): this condition comprises interaction and communication between people, socioeconomic bonds, social support systems, business networks (formal and informal), relations based on trust, and so on. The usage of creative capital (CC): this may be seen as a great ability to cope with challenges and new opportunities, and is reflected in entrepreneurial spirit, new ways of thinking and acting, trend-setting artistic expressions, innovative foresights, and so forth. Such a factor is often found in a multicultural urban melting pot. The existence of ecological capital (EC): this condition takes for granted that a favourable quality of life, an ecologically benign condition in a city, presence of green space and water, or an attractive living climate (for example, recreation and entertainment possibilities) contribute significantly to the innovative and sustainable potential of a region.

The pentagon factors can in principle be measured and quantified, and then be put in an explanatory econometric model (for an empirical estimation see Capello et al., 1999).

THE ORGANIZATION OF THIS BOOK The remainder of this book is arranged around a number of themes. These are outlined below. Different Disciplinary Takes on Endogenous Regional Development First there are five chapters in which scholars from different disciplinary backgrounds provide their ‘take’ on endogenous regional development: from an economist, a geographer, a sociologist, a planner and a behavioural/management scientist. Kenneth Button provides an economist’s viewpoint (Chapter 2) in which he refers to the powerful tool-kit of powerful techniques economists have for analysing and understanding why some countries and regions grow faster than others. He refers to a world without endogenous growth and the well-known exogenous growth models in neoclassical economics – as illustrated by the Solow model – before turning to discuss the emergence of the ‘new’ or ‘endogenous growth theory’ from the 1980s, while also reminding us of important precursors to the new growth theories

STIMSON PAGINATION (M2469).indd 11

20/12/2010 15:12

12

Endogenous regional development

such as Schumpeter’s discussion of entrepreneurs in the 1930s. The Romer model in particular is discussed, along with the contributions of Arrow and Lucas, with technology, R&D and knowledge being considerations of key importance in the development process. And there is a discussion of the important issue of the empirical testing for regional convergence and diversity in regional development. Button raises the question that if indeed there are endogenous growth effects and if there is endogeneity in the development process, then the opportunity for wider policy options certainly opens up. The geographer’s perspective is provided by Michael Taylor and Paul Plummer (Chapter 3) in which the discussion of regional development is placed firmly within the contemporary process of globalization and the transition from the Fordist to a post-Fordist production regime. They refer to the ‘new economic geography’ and the ‘new regionalism’ discourses, focusing the discussion mainly on the latter and plugging forcefully for the adoption of both quantitative and qualitative approaches to analysis in bringing these discussions to bear in the empirical investigation of regional development. There is discussion of the ‘abstract’ nature of theoretical models of regional growth and criticism of what are referred to as ‘stylized’ facts about local economic development and a dependence on proxy variables in regional econometrics. At the same time the authors are critical of the simplistic and under-theorized approach often taken in the geographic literature. There is discussion of the notion of ‘embeddedness’ to explain the dynamics of local growth, especially as it related to learning, innovative milieu and regional innovation systems in the ‘new regionalism’ debate, with the added emphasis on the advantages of geographical proximity. Importantly there is a detailed discussion of how mixed-method, but theoretically informed empiricism, research might be of use in framing policy advice for regional development, with the authors providing a case study of a project in which they were involved in Australia with a focus on institutional capability in human capital development to enhance regional development. A sociologist’s perspective is provided by Frank Vanclay (Chapter 4) in which he sets the discussion specifically in the context of rural development in a local situation where the focus is explicitly on place-based dimensions and initiatives. In keeping with much of the concern of the sociologist, the emphasis is on understanding the natural, human and cultural attributes of place. Vanclay refers to the notion of development being about how a place might realize its potentialities to create sustainable activity. Much consideration is placed on ‘values’ and the achievement of ‘multifunctionality’ and a concern about the ‘subsidiarity principle’ to renew and revitalize the countryside and its community. In many ways the

STIMSON PAGINATION (M2469).indd 12

20/12/2010 15:12

Endogenous regional development

13

approach outlined by Vanclay is very micro in scale and takes a cultural economy approach to local development, with detailed reference being made to the operation of the EU’s LEADER (‘Liaison Entre Actions de Développement de l’Economie Rurale’) programme. Edward Blakely (Chapter 5) provides a planner’s perspective that focuses on how to derive development outcomes from local resources, the harnessing of which is necessary to produce results that are tangible. He discusses how planning relies on what he calls ‘indigenous activities and/ or endogenous resources’ to shape outcomes in space and place. To assist this it is necessary to formulate a ‘Plan’, as plans (about which there is a lengthy discussion) are descriptive instruments but they need implementation. Thus, there is a considerable emphasis on what might be called institutional factors. Not surprisingly for a planner, Blakely places considerable emphasis on ‘land’ as a resource – which needs to be capitalized on – and its use as an endogenous base in local or regional development. There is also discussion of the importance of planning strategy that seeks equitable and sustainable wealth outcomes. Institutional capability and capacity building is crucial for the process to succeed. Blakely provides a discussion of how for many places the challenge is how to overcome the ‘product cycle trap’ through harnessing their endogenous resources, and of how social capital can play an important role. A behavioural/management science perspective is provided by Tojo Thatchenkery and Jessica Heineman-Pieper (Chapter 6) in which the emphasis is on human capital, cultural and organizational dimensions. They show how an endogenous ecology coupled with technology innovation may spontaneously create cascading entrepreneurial activity, as highlighted in the case of Silicon Valley and the social and cultural embeddedness of its actors. The authors present a framework they call ‘appreciative intelligence’ and discuss how this concept may be transferable as an endogenous process in a wider context of regional development. The discussion of this perspective also incorporates the broader regional science literature on knowledge spillovers and industrial districts and the long-established notion of agglomeration. Measuring and Modelling Endogenous Regional Growth and Development The second theme that is addressed in this book relates to the measurement and modelling of endogenous regional growth and development. Somewhat surprisingly there are few examples of models that have been explicitly developed and empirically tested to actually measure endogenous regional growth or to investigate the determinants of spatial variations in regional performance on an endogenous growth measure.

STIMSON PAGINATION (M2469).indd 13

20/12/2010 15:12

14

Endogenous regional development

In Chapter 7 Robert Stimson and Roger Stough demonstrate how they have used the regional (differential) shift component derived from a shiftshare analysis of regional employment change, standardized by the size of the regional labour force at the beginning of the period, as a surrogate measure of endogenous regional growth (or decline). The authors present a new model framework for endogenous regional growth. They then provide summary results from two exploratory attempts to model regional endogenous employment change using spatial econometrics regression modelling, first in a study of non-metropolitan government local areas across Australia, and second in a study of metropolitan regions in the US. In the second application of the model framework the authors are able to incorporate proxy variables for institutional and leadership factors and measures of entrepreneurial activity, in addition to a large set of resource endowment measures, as explanatory variables in their model. A specific attention is paid to entrepreneurship in regional development in the chapter by Zoltan Acs and Mark Sanders (Chapter 8). The focus is on analysing rents and endogenous entry in regional growth. The authors present a model in which incumbent firms finance R&D for a clear profit motive but entrepreneurs capture the rents of commercializing the opportunities that this R&D generates through regional intra-temporal knowledge spillovers. The model formalizes the ‘knowledge spillover’ theory of entrepreneurship, predicting that in a decentralized equilibrium R&D will be under-funded and that there is room for welfare gains through policy. But a delicate balance now needs to be found when R&D and entrepreneurship compete over the same resources, and the social optimum is to stimulate both R&D and entrepreneurship and invest resources in facilitating the knowledge spillover to entrepreneurs. This modelling approach is seen to turn the Romer argument on incentive structure on its head. Acs and Sanders show that their results can carry over to the regional level when considering the impact of limited geographical labour mobility, transport costs and communication costs. That has important policy implications at the aggregate and regional levels, but the challenge is to ascertain the exact channels through which such spillovers arise. Addressing Endogenous Factors at Different Levels of Scale There follows a series of chapters in which endogenous factors in regional development are considered at a number of different levels of scale. Taking a national economy perspective, in Chapter 9 Tomokazu Arita, Chie Iguchi and Philip McCann address the role that foreign direct investment plays in three of the world’s largest newly industrializing countries: China, India and Indonesia. The focus of their analysis is on an evaluation

STIMSON PAGINATION (M2469).indd 14

20/12/2010 15:12

Endogenous regional development

15

of institutional factors that are operating in the context of globalization and rapidly improving information technologies. The combination of these features produces complex interrelationships between economic geography, economies of scale and global trade and investment, and the authors review those interrelationships in the three rapidly developing case study countries. The authors highlight how the economic growth of these three countries is dominated by specific large urban regions where agglomeration economies appear to be critical, likely exacerbating regional disparities. But the authors identify differences in roles played by localized technology spillovers. In a study of disparities in inter-regional performance across the EU, Daniela Constantin, Zizi Goschin and Gabriela Dragan (Chapter 10) focus on the performance of ten new member states and evaluate the institutional capabilities affecting their absorption capacity. The discussion focuses on the implications of the EU’s structural assistance programmes (a part of the social cohesion policy) on regional disparities for the new member states and the regions within them. The authors use Gini coefficients and the covariation, weighted by population, to investigate the degree to which convergence is occurring, showing how divergence has actually increased. Institutional inefficiencies and ineffectiveness are highlighted as potential endogenous barriers or impediments. In Chapter 11, Roberto Camagni and Roberta Capello also discuss the issue of differentials in the performance of regions across the EU nations. Their concern is with investigating regional competitiveness through the concept of ‘territorial capital’. They provide a set of theoretical perspectives on the regional convergence/divergence debate, and discuss the relationships between regional performance and exogenous and endogenous factors. Using the regional economic growth simulation model known as MASST, the authors investigate the impact of macroeconomic policy at the regional level across EU regions, the impact of EU regional policies such as the structural funds and the role of specific elements of territorial capital on regional growth patterns. Moving to the level of scale of the city, in Chapter 12 Alistair Robson provides a comparative analysis of endogenous regional employment growth performance for each of Australia’s five largest cities – Sydney, Melbourne, Brisbane, Perth and Adelaide – over the decade 1996 to 2006. He conducts a shift-share analysis of employment change and focuses in particular on the regional shift component as an indicator of endogenous performance. The analysis includes segmenting the analysis by industry sectors. Marked differences are apparent in the performance of the cities. At a much more local level of scale, in Chapter 13 John Martin looks at the performance of four small towns in two rural shires in the state of

STIMSON PAGINATION (M2469).indd 15

20/12/2010 15:12

16

Endogenous regional development

Victoria in Australia in the wider context of population and economic decline that is apparent widely across small towns. Martin focuses on the role of local leadership – in particular the role of local government – and social capital in those four small communities through a ‘sociometry framework’. The author demonstrates how leadership and institutional factors, including strategies enhancing social capital, have played a significant role in the local planning process to try to enhance the survival and functioning of some of those small towns in a rural region. In Chapter 14 Terry Clower investigates the influence of economic development incentives through government industry assistance programmes as an institutional factor in the states of Alabama and Texas in the US. These are common initiatives in response to plant closures at the local level. The author provides a detailed case study of a food manufacturing firm relocation to Grayson County in the Sherman-Denison Metropolitan Area of Texas under the Texas Enterprise Fund. The distorting effect of such a one-off exogenous incentive on the results of a regional employment shift-share analysis is clearly demonstrated. There is also a discussion of what might be better endogenous incentives to grow and attract industry in local development policy. A Way Ahead? In the final chapter (Chapter 15) Peter Nijkamp and Roberta Capello point out that regional growth theory covers a little more than half a century and during that time we have deepened our knowledge and understanding of the complex backgrounds and impediments to balanced regional development. In that context endogenous considerations are of great importance. Nijkamp and Capello provide a provocative review of approaches to investigate regional development, which begins with a discussion of the vexed question of convergence hypothesis in the debate on regional disparities. In particular they draw attention to the policy issues involved, referring to the need for regions to achieve a role in the international division of labour and maintain it over time, and pointing to the implications of regional growth trajectories in coping with sustainability issues and scarcity of environmental resources, such as environmental quality, natural resource security, urban quality of life, or spatial biodiversity. The authors debate the need for theories of regional growth and development to be more realistic and to better incorporate the complex and interactive behaviour and processes that take place in space and to understand regional competitiveness in terms of endogenous factors. In particular they refer to the importance of the spatial distribution of knowledge and

STIMSON PAGINATION (M2469).indd 16

20/12/2010 15:12

Endogenous regional development

17

its spillovers as important success factors for regional development in an open competitive economic system and how they are critical for regional welfare creation. Nijkamp and Capello argue strongly for a more dynamic approach to theories and models while recognizing the difficulties of the complexity of time with spatial analysis through non-linear approaches. Finally, the authors write of the institutional challenges for regions pursuing a sustainable development path and the crucial role of innovation processes. The chapter concludes with the prophetic statement that ‘regional growth is thus a race without a finish; and that also applies to regional growth theory’. Hopefully, the chapters in this book contribute to that race.

REFERENCES Acs, Z.J., Anselin, L. and Varga, A. (2002): Patents and Innovation Counts as Measures of Regional Production of New Knowledge, Research Policy, 31(7), 1069–85. Arthur, W.B. (1994): Increasing Returns and Path Dependency in the Economy, University of Michigan Press, Ann Arbor. Barro, R.J. (1990): Government Spending in a Simple Model of Endogenous Growth, Journal of Political Economy, 98(5), S103–S125. Blakely, E.J. (1994): Planning Local Economic Development: Theory and Practice, 2nd edition, Sage Publications, Thousand Oaks, CA. Capello, R. (2009): Spatial Spillovers and Regional Growth: A Cognitive Approach, European Planning Studies, 17(5), 643–60. Capello, R. and Nijkamp, P. (eds) (2009): Handbook of Regional Growth and Development Theories, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Capello, R., Nijkamp, P. and Pepping, G. (1999): Sustainable Cities and Energy Policies, Springer-Verlag, Berlin. Döring, Th. and Schnellenback, J. (2006): What Do We Know About Geographical Knowledge Spillover and Economic Growth?, Regional Studies, 40(3), 375–95. Erickson, R.A. (1994): Technology, Industrial Restructuring and Regional Development, Growth and Change, 25(3), 353–79. Erickson, R.A. and Leinbach, T. (1979): Characteristics of Branch Plants Attracted to Non Metropolitan Areas, in: Lonsdale, R. and Seyter, H.L. (eds): Non Metropolitan Industrialization, Winston, Washington, DC. Etzkowitz, H. and Leydesdorff, L. (1996): The Future Location of Research: A Triple Helix of University–Industry–Government Relations, II, EAAST Review, 15(4), 20–25. Etzkowitz, H. and Leydesdorff, L. (1997): Universities and the Global Knowledge Economy, a Triple Helix of University–Industry–Government, Pinter, London. Etzkowitz, H. and Leydesdorff, L. (2000): The Dynamics of Innovation: From National Systems and Mode 2 to a Triple Helix of University–Industry– Government Relations, Research Policy, 29(2), 109–29.

STIMSON PAGINATION (M2469).indd 17

20/12/2010 15:12

18

Endogenous regional development

Florida, R. (2002): The Rise of the Creative Class: And How It’s Transforming Work, Leisure, Community and Everyday Life, Basic Books, New York. Fukuyama, F. (1995), Trust: The Social Virtues and Creation of Prosperity, Free Press, New York. Garlick, S., Taylor, M. and Plummer, P. (2006): An Enterprising Approach to Regional Growth: The Role of VET in Regional Development, NCVER (National Centre for Vocational Education Research), Australian National Training Authority, Adelaide. Grossman, G.M. and Helpman, E. (1991): Innovation and Growth in the Global Economy, MIT Press, Cambridge, MA. Hertz, N. (2009): Cooperation Must Rule, Nature, 457(7232), 962–3. Huxham, C. (1996): Collaborative Advantage, Sage, Thousand Oaks, CA. Johansson, B., Karlsson, R. and Stough, R.R. (eds) (2001): Theories of Endogenous Regional Growth, Springer, Berlin. Keeble, D. and Wilkinson, F. (1999): Collective Learning and Knowledge Development in the Evolution of Regional Clusters of High-technology SMEs in Europe, Regional Studies, 33(4), 295–303. Knight, R.V. and Gappert, G. (eds) (1989): Cities in a Global Society, Sage Publications, Newbury Park. Krugman, P. (1991): Geography and Trade, MIT Press, Cambridge, MA. Krugman, P. (1995): Development, Geography and Economic Theory, MIT Press, Cambridge, MA. Krugman, P. (1996): The Self-Organising Economy, Blackwell, Oxford. Krugman, P. and Venables, A. (1996): Integration, Specialization and Adjustment, European Economic Review, 40(3–5), 959–67. Lucas, R.E. (1988), On the Mechanics of Economic Development, Journal of Monetary Economics, 22(1), 3–42. Markusen, A. (1985): Profit Cycles, Oligopoly and Regional Development, MIT Press, Cambridge, MA. Nijkamp, P. (forthcoming): E Pluribus Unum, Region Direct. Nijkamp, P. and van Hemert, P. (forthcoming): Knowledge Infrastructure and Regional Growth, Regional Symbiosis. Nijkamp, P., Vleugel, J., Maggi, R. and Masser, I. (1994): Missing Transport Networks in Europe, Ashgate, Aldershot. Norton, R.D. and Rees, J. (1979): The Product Cycle and the Decentralization of North American Manufacturing, Regional Studies, 13(2), 141–51. OECD (2000): Links between Policy and Growth: Cross-country Evidence, draft paper for Working Party 1, Economics Department, OECD. OECD (2001): The Well-Being of Nations: The Role of Human and Social Capital, Centre for Educational Research and Innovation (CERI), OECD Publishing, Paris. Ohmae, K. (1995): The End of the Nation State: The Rise of Regional Economics, Free Press, New York. Porter, M.E. (1990): The Competitive Advantage of Nations, MacMillan, New York. Prud’homme, R. (1995): On the Economic Role of Cities, in Cities and the New Global Economy, papers from an international conference presented by the OECD and the Australian Government, Australian Government Publishing Service, Canberra, 3, 728–42. Rebelo, S. (1991): Long Run Policy Analysis and Long Run Growth, Journal of Political Economy, 98(5), S71–S102.

STIMSON PAGINATION (M2469).indd 18

20/12/2010 15:12

Endogenous regional development

19

Rees, J. (1979): State Technology Programs and Industry Experience in the USA, Review of Urban and Regional Development Studies, 3(1), 39–59. Rees, J. (2001): Technology and Regional Development: Theory Revisited, in: Johansson, B., Karlsson, Ch., Stough, R.R. (eds): Theories of Endogenous Regional Growth, Springer-Verlag, Heidelberg, pp. 94–110. Romer, P.M. (1986): Increasing Returns and Long-run Growth, Journal of Political Economy, 94(5), 1002–37. Romer, P.M. (1990): Endogenous Technological Change, Journal of Political Economy, 98(5), S71–S102. Scott, A.J. (1988): New Industrial Spaces: Flexible Production Organization and Regional Development in North America and Western Europe, Pion, London. Simmie, J. (ed.) (1997): Innovation, Networks and Learning Regions?, Jessica Kingsley, London. Solow, R.M. (1956): A Contribution to the Theory of Economic Growth, Quarterly Journal of Economics, 70(1), 65–94. Solow, R.M. (2000): Growth Theory: An Exposition, Oxford University Press, New York. Stimson, R.J. and Stough, R.R. (with Salazar, M.) (2009): Leadership and Institutions in Regional Endogenous Development, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Stimson, R.J., Stough, R.R. and Roberts, B.H. (2006): Regional Economic Development: Analysis and Planning Strategy, revised edition, Springer, Berlin. Taylor, I. (2009): Learn to Convince Politicians, Nature, 457(7232), 958–9. Thomas, M.D. (1975): Growth Pole Theory, Technological Change and Regional Economic Growth, Papers of the Regional Science Association, 34(1), 3–25.

STIMSON PAGINATION (M2469).indd 19

20/12/2010 15:12

2.

The economist’s perspective on regional endogenous development Kenneth Button

There must be a purely economic theory of economic change which does not merely rely on external factors propelling the economic system from one equilibrium to another. (J. Schumpeter, Preface to the Japanese edition of Theorie der Wirtschaftlichen Entwicklung, 1911, transl. 1934) Economic progress, in capitalist society, means turmoil. And, . . . in this turmoil competition works in a manner completely different from the way it would work in a stationary process, however perfectly competitive. Possibilities for gains to be reaped by producing old things more cheaply are constantly materializing and calling for new investments. These new products and new methods compete with the old methods not on equal terms but at a decisive advantage that may mean death to the latter. (J. Schumpeter, Capitalism, Socialism, and Democracy, 1942)

INTRODUCTION Economists have developed a tool-kit of very powerful techniques for analyzing a wide range of social phenomena but they have, beyond gaining a very general understanding, been singularly unsuccessful in unanimously agreeing on an explanation of why some regions’ or countries’ economies grow faster than others.1 This is an important intrinsic question, but it also raises issues about the roles of institutions that have been established to narrow the gap between the rich and the poor. Those involved in trying to explain economic growth have included the luminaries of their generations, and their political leanings range from the capitalist market orientation of Joseph Schumpeter to the centralism of Karl Marx. Indeed, if one accepts Milton Friedman’s (1953) criteria for a good model, namely its ability to produce good forecasts, then economic growth models have been sadly inefficient, although, one might add, that by the standards of other social science disciplines, the notions of precision used by economists are particularly demanding. The expansion of material development and enhanced human welfare 20

STIMSON PAGINATION (M2469).indd 20

20/12/2010 15:12

The economist’s perspective

21

has always been at the core of modern economics. Adam Smith, David Ricardo, Malthus and other early classical economists offered a variety of explanations for improving economic efficiency, focusing on the benefits of the division of labor in one way or another, but they were largely focused on squeezing more production from a given resource base than economic growth per se.2 Their concern was essentially with what we now consider comparative statics rather than dynamics. More recently the importance of technological progress in economic growth has attracted attention. Initially, and following the pioneering work in the 1920s of Frank Ramsey (1928), strenuous intellectual efforts have been made to incorporate into economic models the changes in technology that have manifestly affected economic growth since the Industrial Revolution. The basic point being that capital and labor ‘augmenting’ technology continually increases returns to factors of production, leading to additional investment, allowing per capita output to increase seemingly indefinitely. The challenge has been to understand where this technical progress comes from, how it becomes embedded in production processes and how it will, in quantitative terms, influence the economic growth rates of different regions. While this has largely been treated as a question of macroeconomic interest, there are important spatial implications especially because technological progress is not ubiquitous or cannot always be easily and quickly transferred between regions.

BACKGROUND: A WORLD WITHOUT ENDOGENOUS GROWTH In the 1960s the UK’s Royal Economic Society and the American Economics Association commissioned a series of survey papers by leading figures in their fields. Two of these, that by John Meyer (1963) on ‘Regional Economics’, and by Frank Hahn and Robin Matthews (1964) on ‘The Theory of Economic Growth’ provide a useful basis for considering how and why endogenous growth theory has developed over the past 45 years or so. While there was concern at the time amongst regional economists over spatially differential growth patterns, the emphasis of academic work, and possibly reflecting the dominance of Keynesian economics at the time, was more on short-term considerations. In particular there was interest in the tracing through via multiplier and input-output analysis the local implications of expanding a region’s export base. The neoclassical world of macroeconomic growth theory before the 1980s thought that growth was the result of forces that impinged from the outside rather than being a product of the economic system itself.

STIMSON PAGINATION (M2469).indd 21

20/12/2010 15:12

22

Endogenous regional development

Per capita income growth was seen as dependent on saving rates and on exogenous shocks to the steady-state growth path that resulted. The empirical work of the time focused on extracting the inputs of the factors of production – de facto, labor inputs – from the growth process and then treated differences in national or regional growth as an accounting residual. To be fair, however, in part this was as much due to data weaknesses as to intellectual understanding.3 As we see below, endogenous growth theory goes beyond this and seeks to look at the private and public sector choices that cause these residuals to vary. Comparative Statics Traditional macroeconomic, long-run growth theory at the time these papers were written was largely divided between theories that assumed economic growth could occur without technical progress and those that focused, albeit in a particular way, on technology impacts. The former were very much in the classical tradition of Smith with a focus on growth in the labor force. In equilibrium, the growth of output is limited by the growth of the labor force and that implies a constant per capita income. Where regional, as opposed to macro, growth (at any level of aggregation) models differed from this to some extent was to treat labor supply as both dependent on the internal demographics of a region and upon migration between regions. Basically, it allows for the spatial mobility of factors of production. Hence, a specific region may grow when there is capital abundance by ‘importing’ labor from other regions. This, combined with the natural growth in labor supply within the region, can lead to growth through more efficient use of the labor stock although in the long run there will be convergence in per capita income. Figure 2.1, developed from Hart’s (1975a, 1975b) synthesizing work, offers a simple illustration of the point. We assume there to be two regions, A and B. The former enjoys high-income levels and low unemployment whilst B is the mirror image of this.4 There are decreasing returns to factors. The classical economic model assumes that with zero costs of migration and a homogeneous labor force, labor will move from B to A seeking work and higher pay whereas, on the assumption of uniform commercial risk across regions, capital will move from region A to B where it can be combined with abundant, cheap labor to maximize returns. Wages will fall in A, unemployment will increase, and the return on capital will rise as the labor supply increases and capital becomes scarcer. The additional amount of capital and the decreasing size of the labor force in region B will push down the marginal return on investment and concurrently push up wages. The process continues until labor costs and unemployment levels are

STIMSON PAGINATION (M2469).indd 22

20/12/2010 15:12

The economist’s perspective Region A

Capital

23 Region A

Labor

Region B Classical Model

Capital

Skilled Labor

Region B New GrowthTheory Model

Note: I = income; U = unemployment. Source:

Developed form Hart (1975a and 1975b).

Figure 2.1

Simplified theories of migration

equalized.5 This equalization is achieved in a world of zero transportation costs and full information about employment and investment opportunities. There are no legal impediments to factor mobility, and racism is absent. Essentially the model is devoid of any real geographical consideration. Exogenous Growth Modes The neoclassical theory can be treated as an extension to the well-known Harrod-Domar model6 of economic growth developed in the 1940s that includes an additional term reflecting productivity growth. The broad theory is typified by the work of Robert Solow (1956) and the Australian Trevor Swan (1956) on changes in the factor endowment of a country. In their neoclassical models, the long-run rate of growth is exogenously determined – in other words, it is determined outside of the model. A common prediction of these models is that an economy will always converge towards a steady-state rate of growth that depends on the rate of technological progress and the rate of factor accumulation. A country with a higher saving rate, for example, will experience faster growth. In the Harrod-Domar framework, steady-state growth is unstable; it is on a ‘knife-edge’ in the sense that any deviation from that path would result in a further, and subsequently cumulative, move away from that

STIMSON PAGINATION (M2469).indd 23

20/12/2010 15:12

24

Endogenous regional development

path, either up or down. However, Solow, Swan and others argued that the capital–output ratio of the Harrod-Domar model should not be regarded as exogenous. They proposed a growth model where the capital– output ratio was precisely the adjusting variable that would lead a system back to its steady-state growth path. The key assumption of the neoclassical economic growth model is that capital is subject to diminishing returns. Given a fixed stock of labor, the impact on economic output of the last unit of capital accumulated will always be less than the one before. Assuming for simplicity no technological progress or labor force growth, diminishing returns implies that at some point the amount of new capital produced is only just enough to make up for the amount of existing capital lost due to depreciation. At this point, because of the assumptions of no technological progress or labor force growth, the economy ceases to grow. Assuming non-zero rates of labor growth complicates matters at the regional level, where migration may be large, somewhat, but the basic logic still applies – in the short run the rate of growth slows as diminishing returns take effect and the economy converges to a constant steady-state rate of growth (that is, no economic growth per capita).7 Including non-zero technological progress is very similar to the assumption of non-zero workforce growth: a new steady state is reached with constant output per worker-hour required for a unit of output. However, in this case, per capita output is growing at the rate of technological progress in the steady state (that is, the rate of productivity growth). More formally, an exogenous growth model can be formulated by taking a simple Cobb-Douglas production function of the form, Y = Y = A(t)K1−b Lb (where Y is net national product, K is the capital stock, L is the labor stock and A is the level of technology). The fact that A is a function of time (t) indicates the standard neoclassical assumption that technology only improves over time for reasons external to the model.

ENDOGENOUS GROWTH MODELS Adherents to endogenous economic growth theory cite three main limitations of the neoclassical model: 1.

2.

It relies heavily upon technological change to supply growth in per capita income but has no mechanism for explaining the sources for such change. It offers only a very rudimentary framework for assessing the effects of government policy, and while government actions may not be

STIMSON PAGINATION (M2469).indd 24

20/12/2010 15:12

The economist’s perspective

3.

25

able to raise long-run growth rates, government interventions do affect behavior and this, in aggregate, affects the growth path (be it positively or negatively). The model has limited capabilities for analyzing trade between regions or countries and the links between such trade and economic growth.

According to the ‘new’ or ‘endogenous growth theory’ that began to emerge in its current form in the 1980s, economic growth can be understood as a process of learning-by-doing, within a firm, within an industry and within a given spatial jurisdiction such as a region or metropolis. While there were earlier attempts to indigenize technical progress much of the credit for the modern formulation of endogenous growth theory is attributed to Paul Romer (1986, 1990a) and Robert Lucas (1988, 1993). By way of a counterpoint, while exogenous growth theory sees rising output per capita as resulting from externally given increases in the quantities of labor and capital with no internal technological or organizational change, and no economies of scale, but only constant returns in an environment altered solely by investment levels and labor force growth, endogenous growth models see economic growth over time entailing increasing returns to scale for a metropolis or a national economy. A proportionate increase in labor and capital gives rise to more than proportionate gains in output. The explanation lies in better ‘recipes’, as Romer terms innovations, and in spillovers that operate over time, enhancing skill and productivity levels throughout the economy. This approach to growth has the advantage of supplementing the neoclassical model, without destroying it, by refining the ways in which technology change is stimulated and absorbed in economies. The need for a more complete theory of technical change was there in the 1980s, as pointed out by Romer (1994), but earlier efforts aimed at bringing technology in from the cold had not been analytically developed or fully embraced in the theorizing. Precursors to Modern Endogenous Growth Theory In his Capitalism, Socialism, and Democracy, Schumpeter (1942), albeit not in a mathematically rigorous fashion, laid the foundations for much of the subsequent work on endogenous growth theory. The basis of his analysis, however, had more to do with the nature of general political systems, and their implications for national welfare than with per capita income per se, and it was not developed at the regional level.8 Nevertheless, the idea was that within a capitalist system the continual pressure on profit margins of incumbent businesses generated by the pressures of entrepreneurs and

STIMSON PAGINATION (M2469).indd 25

20/12/2010 15:12

26

Endogenous regional development

imitators led to an inevitable push within business to be at the cutting edge of technology. Put another way, in Schumpeter’s interpretation of capitalism, innovative entry by entrepreneurs was the force that sustained long-term economic growth, even as it destroyed the value of established companies that enjoyed some degree of monopoly power. Part of the subsequent stimuli to develop a more technically rigorous endogenous framework in the 1950s and 1960s also came from reassessments of earlier empirical findings.9 According to calculations by Solow (1957), for example, 87.5 percent of growth in output in the US between 1909 and 1949 could be ascribed to technological improvements alone, suggesting that the ‘Solow residual’ was large. One of the first reactions of the neoclassical economists was to argue that by reducing much of that influence to pure capital improvements in the calculations, capitalintensity seemed to play a larger role than suggested by the arithmetic, although, Solow (1960) did later argue that increased capital-intensive investment embodies new machinery and new ideas as well as increased learning for even further economic progress.10 Nevertheless, there were still significant issues to be addressed. Nicholas Kaldor was really the first major post-war economic theorist to consider endogenous technical change as a potentially important element in explaining differences in regional and national growth rates. In a series of papers, including the oft-cited contribution with James Mirrlees (Kaldor and Mirrlees, 1962), Kaldor posited the existence of a ‘technical progress’ function and that per capita income was indeed an increasing function of per capita investment. Thus, ‘learning’ was regarded as a function of the rate of increase in investment. However, Kaldor held that productivity increases had a concave nature (i.e., increases in labor productivity diminish as the rate of investment increases). This proposition falls short of Solow’s insistence on constant returns. Kenneth Arrow (1962) took the view that the level of the ‘learning’ coefficient is a function of cumulative investment (i.e., past gross investment). Unlike Kaldor, Arrow sought to associate the learning function not with the rate of growth in investment but rather with the absolute level of knowledge already accumulated, a stock rather than a flow concept. Because Arrow claimed that new machines are improved and/or are more productive versions of those in existence, investment does not only induce productivity growth of labor on existing capital (as Kaldor would have it), but it would also improve the productivity of labor upon all subsequent machines made in the economy. The trick is to utilize the concept that while firms face constant returns, the industry or economy as a whole takes increasing returns into account. This can be formalized using the Cobb-Douglas production function:

STIMSON PAGINATION (M2469).indd 26

20/12/2010 15:12

The economist’s perspective

27

Y 5 AKaL1 2a where there are constant returns to scale for all inputs together (i.e., a 1 (1 2 a) 5 1). Therefore, it seems as if output per capita, and with it consumption per capita, does not grow unless the exogenous factor, A, grows too. To indigenize A, the Cobb-Douglas production function for each individual firm is defined as: Yi 5 AiKai L1i 2a where the output of an individual firm is related to capital and labor as well as the ‘augmentation’ of labor by Ai. Arrow assumed that Ai, while it looks specific to the firm, is in fact related to ‘knowledge’ in the economy. This knowledge and experience is common to all firms. Arrow argued that knowledge accumulation arises from the past cumulative investment (G) of all firms, that is, the technical augmentation factor is related to economy-wide aggregate capital in a process of ‘learning-bydoing’. In other words, the experience of the particular firm is related to the stock of capital in the economy, G, by Ai 5 Gz. Hence, as the physical capital stock G accumulates, knowledge used by a particular firm also accumulates by a proportion z such that 0 , z , 1. Transferring to the production function for an individual firm yields: Yi 5 G 2zK ai L1i 2a where only G does not have a subscript i, it is a productive force external to the firms and assumed a quasi-public good. Thus, constant returns to scale are maintained at the firm level, but in the aggregate G 5 K because it is only the accumulated stock of capital for the economy. The ‘economywide’ aggregate production function is therefore: Y 5 K a 1 zL1 2a Arrow assumed that a 1 z , 1; hence increasing only capital does not lead to increasing returns. Increasing returns to scale can be obtained by a 1 z 1 (1 2 a) 5 ‘z’ . 0, but capital and labor must both expand. However, by adding this restriction, Arrow’s original model meets Solow’s condition by exhibiting non-increasing returns to scale in aggregate if the rate of growth in an economy is steady.

STIMSON PAGINATION (M2469).indd 27

20/12/2010 15:12

28

Endogenous regional development

The More Recent Models The 1980s saw an upsurge of interest in economic growth theory partly because major structural shifts were occurring as the service sector, and particularly information-based industries, began to grow in importance and as entities such as the EU began to fulfill some of their potential. Essentially, changes were taking place at both the technical and institutional levels that were seen as potentially affecting economic growth. Paul Romer (1986), for example, went to great lengths to disqualify the restriction retained by Arrow – decreasing returns to capital. His main initial thesis, however, was that long-term economic growth stems from the accumulation of knowledge; knowledge being seen as a factor of production embodied in capital and labor.11 Taking Kenneth Arrow’s idea of disembodied knowledge, Romer concludes that there indeed could be constant returns, but argues that the rate of growth of K alone may yield increasing returns; de facto he assumed (a 1 z) . 1 was possible. With this externality, the growth rate of capital gK equals the growth rate of per capita consumption gc 5 gC 2 gL such that gK 5 gc . 0; that is, there is constant positive growth in per capita consumption and capital. It is the externality, from learning-by-doing, that gives a positive growth rate for consumption and output. In addition, the higher the level of disembodied knowledge, the more ‘soil’ exists upon which innovation (i.e., increases in productivity) can work and the higher the rate of technical progress.12 There is a problem with Romer’s model. The size of the labor force enters as an argument. Thus, if L increases, then the growth rate of consumption increases – the Malthusian effect. This is an unattractive feature that could be solved if we had originally divided the accumulated knowledge equation by L. However, the notation in the model gets unnecessarily complicated after this and is not included here.13 Finally, the social planner’s solution to the overall Arrow-Romer model would inevitably be different from the competitive solution presented. Arrow assumed that any firm is so small relative to the overall economy that its investment decisions cannot have any important effect on the aggregate stock of capital or knowledge. In sum, the firm takes no account of the second-round effects of its investments on technological progress, but rather maximizes profits through its choice of labor and capital taking the stock of knowledge as given. A social planner’s idea of the marginal product of capital would be different, as she would take the externality into account and that, in turn and if it could be done effectively, would yield a higher steady-state growth rate. Thus, the equilibrium growth rate under a competitive system is smaller than the optimal growth rate under

STIMSON PAGINATION (M2469).indd 28

20/12/2010 15:12

The economist’s perspective

29

a social planner because firms are not, individually, taking the externality into account. In other words, with full information and an effective toolkit there is a role for public policy. Learning-by-doing within a firm means current unit costs are a function of experience (as measured by the firm’s cumulative past output). Given the learning curve for a single firm, then imitation of successful firms on the part of other firms in the industry spreads the ‘learning’ around, such that the industry as a whole can benefit from falling-forward supply curves. The process links unit costs to cumulative industrial output within a country or region. The ease of imitation and learning then increases within spatial agglomerations, which in turn can be understood as what are often called ‘little nations’ benefiting from increasing returns. Romer refines his work by looking at why private firms directly invest in technical knowledge by undertaking R&D when it has the characteristics of a quasi-public good; knowledge only being partially excludable means that the firm will not reap the full benefits. Patents, trademarks and copyrights are assumed by Romer, as they had been in previous work by Joseph Schumpeter (1942),14 to protect, at least for a time, part of the knowledge acquired, but even without them a firm generally enjoys a first mover advantage and can costlessly reproduce for sale or internal use a new ‘design’. It can sell these new products for a time at prices above marginal cost and earn super-normal profits, but even as the short-term monopoly position is eroded there is the incentive to further innovate because of the lowered R&D costs that stem from the previous experience.15 In effect, R&D activities become a barrier to entry of competition into the market. Robert Lucas (1988, 1993) takes a somewhat different approach to the spillovers from investment, focusing on education and on-the-job training. Essentially, the interest is more on human capital rather than in physical capital and R&D as in Romer’s framework. One approach followed by Lucas involves assuming that personal human capital is built up by workers through their postponing current consumption in the hope that education will allow for greater consumption in the future. To develop an endogenous growth effect from this, he assumes that there is an externality from this activity because the actions of these individuals will enhance the overall productivity of the labor force, including those that do not defer consumption. Lucas found empirical support for this in the spatial agglomeration that can be observed, especially in cities. A second approach is to assume that knowledge is obtained on the job rather than through explicit formal educational programs. Hence, labor does not leave the workforce to acquire and develop knowledge, but gains it from having to produce an ever-increasing range of more sophisticated

STIMSON PAGINATION (M2469).indd 29

20/12/2010 15:12

30

Endogenous regional development

products.16 Lucas supports this line of argument by looking at the experiences of some of the newly industrializing countries of Eastern Asia that have experienced rapid growth in per capita income through selling to diverse and continually changing export markets despite a much slower expansion in the formal education system than his earlier work suggested would be needed. In sum, interpreted variously in terms of Arrow’s learning curves, Romer’s recipe for growth, or Lucas’s vector-autoregressive (VAR) time-series specifications, the key ideas underlying endogenous economic growth are learning-by-doing and cross-fertilization over time. There are effective feedback-loops in the economic system.

IS THERE A SPECIFIC REGIONAL CONTEXT? As we have seen, one implication of Solow’s neoclassical growth model is that each country or region should converge on to its own steady-state growth path at a predictable rate. If one is willing to assume that technology is identical across countries, then one should expect any two countries with identical saving rates and population growth rates to converge to the same level of income. Similarly, all countries should converge to a level of income that can be predicted by its saving rates and population growth rate. Regions, in theory, simply mirror this process, although the greater flexibility in capital and labor markets within a country would suggest that intra-national convergence would be more rapid than international convergence. Empirical evidence for convergence has repeatedly been interpreted, at either level of aggregation, as providing support in favor of the neoclassical model and against endogenous growth models, which we move onto, in both the national and regional context.17 The new growth theory, in contrast to the neoclassical framework, is consistent with divergence in economic growth along the traditional lines of Myrdal’s (1957) ideas of circular and cumulative causation, albeit for somewhat different reasons. The situation can be illustrated by returning to Figure 2.1.18 Taking the initial starting positions for two regions, this approach argues that not only will equalization of real wages and employment levels not be attained but that there may be cases where they diverge further. Labor mobility between regions A and B may be impeded by the various costs of migration – embracing social and search costs as well as simple financial costs – and heterogeneity in the labor market – the jobs available in A not being compatible with the skills and knowledge of labor in region B. This is a potentially serious constraint in knowledge industries in either of the Lucas cases because of the costs of acquiring education in

STIMSON PAGINATION (M2469).indd 30

20/12/2010 15:12

The economist’s perspective

31

his initial model, and the lack of appropriate on-the-job training in the second.19 Equally, capital does not move from region A to B because of the higher returns and less uncertainty that are to be found in regions that already have a high level of prosperity and a pool of complementary labor. While the earlier regional models of Kaldor, in particular, focused largely on divergent growth rates in the context of traditional-style manufacturing industries that still dominated Western economies, the more contemporary form of the theory pays particular attention to the endogenous growth that occurs in regions that have an established high-skilled workforce, and the ability to further develop their knowledge-based industries. Again, there is no substantive difference in the modeling whether it is conducted at the national level or for smaller regions, although institutional and cultural factors may influence the speed and detail of the growth paths of either.

EMPIRICAL TESTING FOR ECONOMIC CONVERGENCE Testing the validity of the alternative theories, in the absence of easily quantifiable counterfactuals, has frequently involved looking at secondary evidence, and in particular at situations that shed light on whether there is convergence in the economic growth paths of regions or, at the macrolevel, nations. The empirical question that is explored at the specific regional level of aggregation becomes one of whether there is convergence in economic growth rates in, generally, per capita income as is a natural outcome of the neoclassical model, but only possible with endogenous growth under rather particular circumstances.20 The aim here is not to even try to provide a comprehensive critique of the numerous studies of growth paths for the variety of economic variables (e.g., income, employment and prices) that have been conducted across a range of different levels of spatial aggregation, locations and time periods, but rather to offer a flavor of what has emerged at the regional level, from some of the more cited, and thus influential, pieces of analysis. The extensive empirical analysis that has emerged has been assisted by improved data sets made available in recent years, as well as new modeling frameworks, enhanced econometric techniques and better understanding of how to measure convergence. In particular, there has been the development of the concept of b-convergence measures (Barro and Sala-i-Martin, 1992; Sala-i-Martin, 1994) that has allowed

STIMSON PAGINATION (M2469).indd 31

20/12/2010 15:12

32

Endogenous regional development

a more rigorous analysis of economic convergence than the more traditional s convergence measure.21 The estimation of possible b convergence involves a mean-reversion calculation and it occurs if there is a negative relationship between the growth rate of income per capita and the level of initial income.22 Much of this type of analysis has been done at the national level, but it, and the more limited body of analysis that has been conducted at the regional level, does offer some insights into the validity of the idea of endogenous economic growth. Much of the early work on spatial economic convergence relied upon aggregate, national data sources and focused on s–convergence measurements (e.g., Abramovitz, 1986; Baumol, 1986; Maddison, 1987). The findings indicated that labor productivity, and with it per capita income in the world, was converging in the long run and thus rendered support to the neoclassical growth theory. The difficulty with this work was that the data sets used only contained countries that had already industrialized, and even for those there were periods of divergence (De Long, 1988).23 Improved data, most notably the Heston-Summers panel data set that embraces a large group of countries, subsequently indicated a lack of any general economic convergence (Romer, 1994). The more recent work that has made use of b–convergence measures, and embraces a number of subnational studies, tends to find little support for overall convergence. Barro and Sala-i-Martin (1991) in a number of studies that, for example, have examined the economies of US states and European Union regions find that while there is evidence of convergence in per capita income, it is slow – about 2 percent per annum and well below the 12 percent or so that neoclassical theory would suggest. These are also conditional convergence measures that allow for homogeneity between, for example, the regional economies within an EU economy but diversity between countries that would suggest potential differences in steady-state growth rates for the regions within them.24 Hence, overall there is little support in this body of work for the exogenous growth idea.

POLICY IMPLICATIONS From a policy perspective, if there were indeed endogenous growth effects this would seem to provide decision-makers with some opportunity to intervene to stimulate growth and to combat spatially divergent growth paths, in effect overcoming the frustration with neoclassical theory expressed by Schumpeter at the beginning of the chapter. This contrasts to the Solow model where only a change in the savings rate could generate long-run growth in per capita income. When in disequilibrium, the

STIMSON PAGINATION (M2469).indd 32

20/12/2010 15:12

The economist’s perspective

33

neoclassical model does allow for fairly limited public policy interventions that would de facto lubricate the system and facilitate a more rapid move to a steady-state growth path. It would not, however, produce movement along it or shift it. In the context of migration, for example, this may involve improved information and enhanced transportation services to allow existing resources to migrate and be used more effectively along Adam Smith’s lines of argument of greater divisions of labor. If there is endogeneity in the growth process then the policy options are somewhat wider. Since knowledge is important, then diffusion of ideas and broader national policies for R&D can be deployed to bring lagging regions up to the production frontier enjoyed by the leading regions. To stimulate a nation’s growth, Romer, for example, argues for a reduction in the US’s federal deficit to reduce interest rates that would in turn increase the amount of human capital devoted to R&D by raising the discounted value of any given stream of future revenues associated with a new design. The Romer framework would also suggest subsidies for R&D because of the currently uncompensated external benefits that it generates; in contrast the Lucas models suggest that these subsidies should largely go to the education and training of workers.25 There is also a case for freer trade in that it allows for the more rapid diffusion of knowledge and thus breaks down the monopoly of those regions and countries that currently enjoy its ‘ownership’. More generally, it releases knowledge workers to invent new designs rather than for those in the lagging regions having to expend energies on catching up and effectively continually having to reinvent the wheel. If one pursues the Richard Florida (2005) line of argument that the creative classes are attracted and retained by the larger environment in which they live and work, then investment in various forms of social and economic infrastructure become important.26

CONCLUSIONS In a way there is no lack of economic theories seeking to explain why some regions or countries grow faster than others. Indeed there may be too many theories. What the recent developments in endogenous growth theory have done is to introduce the less tangible elements of labor and capital (‘knowledge’ in short) much more systematically into the analysis of national and regional economic growth modeling. The accompanying body of econometric and statistical work has helped fine-tune these models and provide numerical methods for policy development and assessment. But the understanding of variations in economic growth paths

STIMSON PAGINATION (M2469).indd 33

20/12/2010 15:12

34

Endogenous regional development

is still far from complete. In part this is because the economic modeling remains controversial and this in turn can partly be explained by gaps in the work that has been completed by complementary disciplines. While there are now relatively powerful economic and programming techniques available to examine macro- and meso-data our understanding of some of the fundamental micro-forces that lead to economic growth is still relatively poor.27 To more completely understand the motivations for companies to undertake R&D, and for individuals to invest differentially in enhancing their knowledge bases, requires more systematic analysis by, in the first case, management scientists, and in the latter by sociologists and social psychologists. Even at the more macro level there appear to be fundamental cultural differences that affect the speed and nature of uptake of new technologies, and especially those that emerge in other regions that are not normally seen as within the scope of economics. Similarly, there appear to be variations in the flexibility to which nations or regions can deal with shocks to their economic systems even within what superficially appear to be relatively similarly educated populations (Pack, 1994). As the old saying goes, ‘The devil is in the detail’, and useful analyses of many of these details require a broader set of tools than those conventionally deployed by traditional economists.

NOTES 1.

2.

We generally discuss development here as an economic phenomenon, and thus in terms of regional incomes. Of course, there are many other elements to development, including such things as the distribution of this income, the condition of the regional environment and a plethora of cultural factors that must be embraced in any holistic approach. The excuse for the narrow approach is, in part, because the author is an economist, but also it allows a focus on the more tangible elements of development. Adam Smith (1776) in explaining the Wealth of Nations posited a supply-side-driven model of growth. Taking the simplest of production functions, Y = g(L, K, T) where, Y is output, L is labor, K is capital and T is land, so output is related to labor and capital and land inputs, output growth (gY) is driven by population growth (gL), investment (gK) and land growth (gT) and increases in overall productivity (g): gY = f(gL, gK, gL, gT). The growth in population is endogenous depending on the sustenance available to accommodate an increasing workforce. Investment is also endogenous, being determined by the rate of savings; land growth was dependent on conquest of new lands or technological improvements of fertility of old lands. Technological progress could also increase growth. Smith’s thesis that the division of labor improves growth was fundamental to his argument. The Smithian and Ricardian models implicitly had technical

STIMSON PAGINATION (M2469).indd 34

20/12/2010 15:12

The economist’s perspective

3.

4. 5.

6. 7. 8.

9.

10. 11. 12. 13.

14.

15. 16. 17. 18. 19.

35

change arising from profit-squeezes or, in the particular case of Smith, arising because of previous technical conditions. This was understandable. At the time of his writing, the UK economy was moving into an extended period when economic growth reached about 1 percent per annum compared with the 0.5 percent per decade that had previously been the norm (i.e., effectively people’s living standards did not change over their life-times) and Smith was, in a way, groping to understand this new world. Harry Richardson (1978, p. 18) says of the empirical work on regional growth up to the 1980s that, ‘The models developed so far have been handicapped by severe data limitations, too few observations, the reliance on static models, their recursive structure, the heavy dependence on national economic variables, and the neglect of space.’ Because of its market-clearing assumptions the pure classical model would be driven only be income differentials. Strictly, with full, instantaneous market clearing there is no unemployment in this type of model, labor movements being determined by real relative wages. The unemployment effect is added to indicate possible imperfections in the short-term labor markets in the two regions. This model synthesizes the work of Evsey Domar (1947) and Roy Harrod (1939). The Malthusian model, because of its implied capital productivity function, may be seen as an extreme case of this whereby society cannot rise above the subsistence level of per capita income. McCraw (2007), for example, argues that part of Schumpeter’s motivation for the style of presentation in the book was due to a degree of chagrin felt over the popular success of John Maynard Keynes’ General Theory that had overshadowed his own analysis of trade cycles. Strict modeling along Schumpeterian lines did not emerge until the 1990s. Another of the stimuli for the interest in developing a more complete understanding of economic growth in the 1950s and 1960s was institutional and founded on concerns about the economic performance of the economies of the newly independent, excolonial nations and the role that institutions such as the World Bank should play in assisting them. This was at a time when it became clear that developing countries were not catching up with higher-income nations. For Solow’s later thoughts on the roles between endogenous and exogenous growth theory see Solow (1994). Romer’s assumption that knowledge and physical capital are related stemmed from empirical observations that there were greater correlations between investment and output than the neoclassical model would predict. Romer (1990b) provides a series of anecdotal examples to support this view. How the stock of knowledge is to be measured is a moot point (Steedman, 2001). There are various elements or components to knowledge that need weighting to determine the stock, even if they can be defined and measured. There are also questions as to the extent that knowledge is non-rival rather than being unique to individuals. Joseph Schumpeter provided a general model of economic growth in which through the forces of creative destruction in oligopolist markets, entrepreneurs were stimulated to develop new technologies. His discussion of endogenous growth was couched very much in terms of a critique of Karl Marx but now emerged into a rigorous model of the type Romer and others later developed. This raises issue of whether endogenous growth is, in effect, an assumption in the model or a result of the model. The model is not closed, however, because Lucas only assumes that new goods are continually being produced without having an explicit model of why. Studies of economic convergence provide a pragmatic way of indirectly testing endogenous growth theory without the need to specifically define and measure ‘knowledge’. For a much more rigorous analysis of endogenous growth involving two regions or countries see, Segerstrom et al. (1990). Richard Florida (2005) has added to these ideas by arguing that ‘creative classes’ having preference for where they live, and that the attributes required to attract and retain

STIMSON PAGINATION (M2469).indd 35

20/12/2010 15:12

36

20.

21. 22.

23.

24.

25.

26. 27.

Endogenous regional development them are more likely found in regions with an established high-income and educated labor force. In either the older or more recent formulations of this framework, the implications are often circular-and-cumulative causation; essentially richer regions get richer and poorer regions, poorer. Theoretical divergence is not an automatic outcome of endogenous growth (Kelly, 1992). Robert Tamura (1991), for example, by making the simple assumption that an idea is more difficult to discover than it is to learn, and following from this that there is a higher payoff from education for people with a lower basic level of education, demonstrates that convergence can occur. Put another way, if a person is at the technology frontier, movement out requires new knowledge whereas for someone within the frontier it is possible, and easier, to move out by acquiring existing knowledge. This approach essentially looks for changes in standard statistical indicators of dispersion – normally the variance. A standard estimating equation takes the form D(yit) = a + b(yit−j) + DiG + e it, where the yit is that being tested for convergence, Di is a (n × k) matrix of k variables capturing spatial economic heterogeneity, G is a (k × 1) vector of parameters, a and b are constants and eit is a white noise error term. The condition b > 0 is a necessary but not sufficient condition for b-convergence (Bernard and Durlauf, 1996). A critique of the empirical work using this and similar tests for convergence is provided by Howard Pack (1994). Bradford De Long and Lawrence Summers (1991) make the general point in their specific analysis of investment in equipment, to which they credit much of the positive spillover effects on economic growth, that markets are not perfect and that bad policy decisions (government failures) inevitably occur and will influence the nature of the growth path for any economy. Some studies seek to embrace such effects in the analysis, for example, Barro (1991) includes political stability, potential rent seeking, and other ‘political’ variables in his calculations while Button and Pentecost (1995) embrace monetary policy within their work on the European Union. Button and Pentecost’s (1995, 1999) work on the regional economies of the EU find that there is evidence of divergence when no allowance is made for possible national steady-state growth path differences. The national boundaries used by Barro and Salai-Martin would not seem to correspond to natural economic units. There would seem to be another policy difference in terms of the types of industry that central policy-makers may seek to encourage to locate in economically slow-growing regions. The tradition of seeking any form of economic activity to stimulate Keynesian multiplier effects is alien to both approaches. Romer’s framework suggests that there is a need to locate more knowledge-based elements of firms and industry in such regions to stop their cumulative decline. Lucas, in contrast, because of the perceived importance of ‘learning-by-doing’, may well argue that provided industry achieved this it would set in motion an effect to stimulate faster long-term growth. The evidence that investment in more traditional infrastructure in slow-growing regions can stem the tide of divergence is less certain (Button, 1998). This is not to say that these techniques are always fully exploited. As Pack (1994) argues, there are likely strong interactive effects between the factors that lead to economic growth but there is a tendency for econometric analysis to miss these interactions in the specifications used.

REFERENCES Abramovitz, M. (1986), Catching up, forging ahead, and falling behind, Journal of Economic History, 46(2), 385–406. Arrow, K.J. (1962), The economic implications of learning by doing, Review of Economic Studies, 29(3), 155–73.

STIMSON PAGINATION (M2469).indd 36

20/12/2010 15:12

The economist’s perspective

37

Barro, R.J. (1991), Economic growth in a cross-section of countries, Quarterly Journal of Economics, 106(2), 407–43. Barro, R.J. and Sala-i-Martin, X. (1991), Convergence across states and regions, Brookings Papers on Economic Activity, 22(1), 107–82. Barro, R.J and Sala-i-Martin, X. (1992), Convergence, Journal of Political Economy, 100(2), 223–51. Baumol, W. (1986), Productivity growth, convergence and welfare: what the longrun data shows, American Economic Review, 76(5), 1072–85. Bernard, A.B. and Durlauf, S.N. (1996), Interpreting tests of the convergence hypothesis, Journal of Econometrics, 71(1–2), 161–73. Button, K.J. (1998), Infrastructure investment, endogenous growth and economic convergence, Annals of Regional Science, 32(1), 145–62. Button, K.J. and Pentecost, E.J. (1995), Testing for convergence of the EU regional economies, Economic Inquiry, 33(4), 664–71. Button, K.J. and Pentecost, E.J. (1999), Regional Economic Performance within the European Union, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. De Long, B. (1988), Productivity growth, convergence and welfare: comment, American Economic Review, 78(5), 1138–54. De Long, B. and Summers, L.H. (1991), Equipment investment and economic growth, Quarterly Journal of Economics, 106(2), 445–502. Domar, E.D. (1947), Capital expansion, rate of growth and employment, Econometrica, 14(2), 13–147. Florida, R. (2005), The Flight of the Creative Class: The New Global Competition for Talent, Harper Collins, New York. Friedman, M. (1953), The methodology of positive economics, in M. Friedman (eds) Essays in Positive Economics, University of Chicago Press, Chicago. Hahn, F.H. and Matthews, R.C.O. (1964), The theory of economic growth: a survey, Economic Journal, 74(296), 779–902. Harrod, R.F. (1939), An essay in dynamic theory, Economic Journal, 49(193), 14–33. Hart, R.A. (1975a), Interregional economic migration: some theoretical considerations (Part I), Journal of Regional Science, 15(2), 127–38. Hart, R.A. (1975b), Interregional economic migration: some theoretical considerations (Part II), Journal of Regional Science, 15(2), 289–305. Kaldor, N. and Mirrlees, J.A. (1962), A new model of economic growth, Review of Economic Studies, 29(3), 174–92. Kelly, M. (1992), On endogenous growth with productivity shocks, Journal of Monetary Economics, 30(1), 47–56. Lucas, R.E. (1988), On the mechanics of economic development, Journal of Monetary Economics, 22(1), 3–42. Lucas, R.E. (1993), Making a miracle, Econometrica, 61(2), 251–72. Maddison, A. (1987), Growth and slowdown in advanced capitalist economies: techniques of quantitative assessment, Journal of Economic Literature, 25(2), 649–98. McCraw, T.K. (2007), Prophet of Innovation: Joseph Schumpeter and Creative Destruction, Belknap Press of Harvard University Press, Cambridge, MA. Meyer, J.R. (1963), Regional economics: a survey, American Economic Review, 53(1), 19–54.

STIMSON PAGINATION (M2469).indd 37

20/12/2010 15:12

38

Endogenous regional development

Myrdal, G. (1957), Economic Theory and Underdeveloped Regions, Gerald Duckworth, London. Pack, H. (1994), Endogenous growth theory: intellectual appeal and empirical shortcomings, Journal of Economic Perspectives, 8(1), 55–72. Ramsey, F.P. (1928), A mathematical theory of saving, Economic Journal, 38(152), 543–59. Richardson, H.W. (1978), The state of regional economics, International Regional Science Review, 3(1), 1–48. Romer, P.M. (1986), Increasing returns and long-run growth, Journal of Political Economy, 94(5), 1002–37. Romer, P.M. (1990a) Endogenous technological change, Journal of Political Economy, 98(5), S71–S102. Romer, P.M. (1990b), Are nonconvexities important for understanding growth?, American Economic Review, 80(2), 97–103. Romer, P.M. (1994), The origins of endogenous growth, Journal of Economic Perspectives, 8(1), 3–22. Sala-i-Martin, X. (1994) Cross-sectional regressions and the empirics of economic growth, European Economic Review, 38(3–4), 739–47. Schumpeter, J.A. (1942), Capitalism, Socialism, and Democracy, Harper and Brothers, New York. Segerstrom, P.S., Anant, T.C.A. and Dinopoulos, E. (1990), A Schumpeterian model of the product life cycle, American Economic Review, 80(5), 1077–91. Smith, A. (1776), An Inquiry into the Nature and Causes of the Wealth of Nations (Oxford University Press edition, Oxford, 1976). Solow, R.M. (1956), A contribution to the theory of economic growth, Quarterly Journal of Economics, 70(1), 65–94. Solow, R.M. (1957), Technical change and the aggregate production function, Review of Economics and Statistics, 39, 312–20. Solow, R.M. (1960), Investment and technical progress, in K.J. Arrow, S. Karlin and P. Suppes (eds) Mathematical Models in Social Sciences, Stanford University Press, Stanford. Solow, R.M. (1994), Perspective on growth theory, Journal of Economic Perspectives, 8(1), 45–54. Steedman, I. (2001), On ‘measuring’ knowledge in new (endogenous) growth theory, paper presented to the Growth Theory Conference, Pisa. Swan, T.W. (1956), Economic growth and capital accumulation, Economic Record, 32(63), 334–61. Tamura, R. (1991), Income convergence in an endogenous growth model, Journal of Political Economy, 99(3), 522–40.

STIMSON PAGINATION (M2469).indd 38

20/12/2010 15:12

3.

Endogenous regional theory: a geographer’s perspective and interpretation Michael Taylor and Paul Plummer

INTRODUCTION The question posed in this chapter is: ‘How do we understand and intervene in processes of regional economic growth and change?’ Globalization is eroding the old economic certainties, increasing competition, reconfiguring the demand for skills, forcing the pace of technological and managerial change, transforming inter-firm relationships and changing (through commodification) the role of finance. Some would argue that there has been a sea change in the workings of national, regional and local economies with the end of Fordism and the coming of post-Fordist flexible specialization and mass customization. For policymakers, the pressing issue is how can appropriate and workable policies and programmes be developed to cope with the internationalization of production and consumption leading to rapid and radical change, foster sustainable and robust local growth and enhance local capacities beyond providing short-term palliatives for unemployment and its associated social problems? From an economic geography perspective, two bodies of discourse currently dominate our understanding of the processes driving local economic change: (1) the discourse of endogenous growth in the ‘new economic geography’ of the economists; (2) the ‘new regionalism’ discourse along with its social constructionist and relational underpinnings in economic geography proper and equivalent areas in other social sciences. Based upon seemingly contradictory criteria of success, derived from either the ‘stylized facts’ of endogenous growth theory or the ‘contingency’ of new regionalism, these bodies of discourse have become reified into a discourse of policy and practice based upon stereotypical understandings of the mechanisms that are supposed to characterize the development of local economies in a world of globalization. Here we argue that, despite 39

STIMSON PAGINATION (M2469).indd 39

20/12/2010 15:12

40

Endogenous regional development

different conceptual foundations, and seemingly contradictory criteria of empirical success, increasingly these two bodies of work are beginning to overlap in terms of the substantive forces that each hypothesize to drive local economic growth. The critical difference between those discourses is placed into sharp relief when we are asked to interpret the causal efficacy of their respective explanatory claims and translate those claims into effective and implementable policies and practice. On the one hand, drawing on the theoretical and methodological presuppositions of neoclassical economics, there is regional endogenous theory. According to its leading proponents this represents a ‘new’ perspective on understanding the process of regional growth and change that is capable of providing clarity, rigour and relevance to a field of research that has hitherto been characterized by anti-clarity, anti-rigour and anti-relevance (Krugman, 1991; Overman, 2004). Typically, the ‘new’ geographical economics is dismissed by many economic geographers as being based on ‘under-socialized’ ill-conceived theory constructs and superficial empirical regularities derived from ‘extensive’ research designs that are considered to be inappropriate means with which to identify the causal mechanisms driving local economic growth. According to Plummer (2007), quantitative methodologies are tolerated in economic geography only to the extent that they are capable of providing background information as a prelude to the real, and preferred, business of the ‘intensive’ qualitative research. Faced with the apparent inadequacies of superficial ‘extensive’ methodologies, many economic geographers have simply abandoned quantification completely in favour of a suite of supposedly ‘deeper’ causal explanations provided by qualitative methodologies (Sayer, 1984; Massey and Meegan, 1985; Clark, 1998). As an alternative discourse, economic geography ‘proper’ has constructed its own ‘new regionalism’ that blends ideas on institutions, individual agency and social regulation (IAR) in an approach, drawing heavily on Polanyi (1957) and Granovetter (1985), which argues that all economies are socially constructed. Eschewing the constraints of theory construction through mathematical modelling and quantitative reasoning based upon stylized facts, this new regionalism attempts to establish the empirical adequacy of its explanatory claims through qualitative description, close dialogue and fine-grained case studies. In contrast with the economists’ emphasis on economic mechanism and quantitative analysis, this more discursive methodology celebrates the supposed inherently complex construction of societal processes and the contingency of the geographical world. In practice, this means exploring the workings of socially constructed regional economies through the lived experience of individuals living and working within it. The perspective that we adopt in this chapter is at odds with both the

STIMSON PAGINATION (M2469).indd 40

20/12/2010 15:12

A geographer’s perspective and interpretation

41

‘new’ geographical economics and economic geography proper. We are, in effect, advocating a critically engaged pluralist approach to understanding local economic growth and change (Plummer and Sheppard, 2007). Unlike many geographers, we reject what we believe to be an unsustainable dualism between quantitative and qualitative economic geographies in favour of a meaningful and grounded multi-method research design that allows us to triangulate the insights of both economists’ and geographers’ ways of knowing. Put differently, as economic geographers, we reject the tout court dismissal of ‘extensive’ research designs of the type employed by endogenous regional theory. Instead, we argue that the way forward is to blend the economism of ‘endogenous theory’, and its ‘extensive’ research strategy, with the social constructionism of ‘new regionalism’ and its ‘intensive’ research strategy. We want to suggest that multi-method research strategies can provide us with critical understanding and explanation of social reality, which can both empirically evaluate the explanatory claims and inform policy debate (King et al., 1994).

CONTRASTING RESEARCH STRATEGIES: CARICATURES AND STYLIZATIONS It can be argued that both endogenous regional theory and geography’s ‘new regionalism’ are caricatures of functioning regional economies, both of which posit processes that are necessary to developing an understanding of economic change at the regional scale, but neither of which offers sufficient explanation on its own. Put differently, all conceptualizations are partial representations of complex real world processes; where they differ is over which aspects of geographical reality they choose to prioritize and how those characteristics that are identified are thought to be related to one another. It is in this sense that all theories or models are stylized descriptions of imitations or local economies, and hence caricatures. The trick, however, is to avoid the exaggerating, comic and/or grotesque characteristics that are typically associated with caricatures! Endogenous Regional Theory With its origins in contemporary economics, endogenous regional theory is driven by a methodological commitment to accounting for a set of established ‘stylized facts’ using abstract mathematical reasoning. To constitute an adequate explanation of these stylized facts, economists’ models of local economic growth need to be formulated in a particular style of mathematical reasoning in which representative agents maximize

STIMSON PAGINATION (M2469).indd 41

20/12/2010 15:12

42

Endogenous regional development

their inter-temporal objective function, subject to resource constraints. Ideally, the actions of these representative agents are supposed to ensure an optimal and stable equilibrium configuration whose properties act as a guide to empirical work (see Sheppard, 2000; Plummer and Sheppard, 2006). Notwithstanding the existence of path dependence and multiple equilibria that are characteristic of models derived from the core model of geographical economics (Brakman et al., 2001), there remains an ontological commitment to the proposition that geographical reality can be understood using equilibrium-based models. Within the constraints of optimum-and-constraints theorizing, endogenous regional theory is grounded in the logic of aggregate production functions relating per capita output (income) to the degree of ‘capital’ intensity and technological change. In its classical Solow-Swan formulation, long-run growth of per capita output (income) is sustained through a constant and smooth, but exogenous, process of technological change (Fingleton, 2003). As a corollary, technological change represents the untheorized determinant of long-run local economic growth and change. In contrast, the ‘new’ regional growth theory attempts to ‘endogenize’ technological change by explicitly modelling processes of learning-bydoing, knowledge spillovers and/or Schumpeterian ‘creative destruction’ (Martin and Sunley, 1998). Typically, endogeneous technological change is introduced either through the ad hoc augmentation of a conventional production function with ‘factors’ such as ‘social’ and/or ‘human’ capital or in the form of a knowledge production function, which is derived from the decision of a rational profit-seeking capitalist to invest in ‘knowledge’ and ‘innovation’ (Jones, 1998). These theoretical specifications form part of a wider class of models, which form the core of geographical economics, and have replaced the conventional assumption that competitive markets exhibit constant returns to scale with the assumptions specifying increasing returns and imperfect competition based upon product differentiation (Fujita et al., 1999; Brakman and Heijdra, 2004). As has been pointed out by many observers, the abstract nature of the theoretical models proposed by endogenous regional theory, coupled with the unrealism of the assumptions that are necessary to generate closedform solutions to these abstract models, means that empirical testing has lagged behind theoretical developments (Neary, 2001; Fingleton, 2007). Nonetheless, recently we have witnessed an explosion in empirical models attempting to evaluate the explanatory claims of endogenous regional theory using ‘third-way’ growth econometrics (Fingleton, 2001). The new growth econometrics represents a particular research strategy that has become the work-horse of economists’ attempts to account for both the convergence that is hypothesized to operate between regions, resulting from

STIMSON PAGINATION (M2469).indd 42

20/12/2010 15:12

A geographer’s perspective and interpretation

43

the inter-regional mobility of factors of production and commodities, and the set of ‘structural’ conditioning factors that are hypothesized to produce persistent, and perhaps divergent, regional growth differentials. Much of the literature emphasizes five ‘stylized facts’ determining, to varying degrees, aspects of change at the regional level, including (Glaeser, 2000): ● ● ● ● ●

technological change and innovation; human capital, embracing research and education; agglomeration and externalities; knowledge spillovers, including entrepreneurship and new firm formation; sectoral specialization/diversification.

Fundamentally, we should remain sceptical about the existence of the economists’ ‘stylized facts’ about local economic growth. Indeed, it may well be the case that diminishing returns have set in regarding the empirical efficacy and policy relevance of growth econometrics. In studies that have adopted the growth regression approach, over 50 different ‘proxy’ variables have been identified as potential conditioning variables, ranging across societal preferences, techniques of production, rates of population growth and government policy regimes (Durlauf et al., 2004). Whilst ad hoc ‘proxy’ variables may fit satisfactorily, the significance of individual variables is not robust, and tends to be sensitive to the set of other conditioning variables that are included in the model (Durlauf and Quah, 1999). Furthermore, it may be difficult to give these ‘proxy’ variables a meaningful interpretation in terms of the economic reasoning underlying the economists’ growth theory, especially when mechanistic variable selection searches are driven by data availability rather than theoretical relevance or plausibility. It is, perhaps, ironic that economists insist on employing sophisticated mathematics to ensure that the parameters related to a local economy’s production function are derived from economic reasoning based upon inter-generational rational choice models, yet they seem willing to rely on casual empiricism when selecting the appropriate (assumed exogenous) set of conditioning variables. At the same time, from the perspective of both theoretical and empirical research in geography, many of the relationships posited in the endogenous regional approach appear simplistic and under-theorized, merely a set of control variables that allows us to focus on the convergence amongst and between local economies. For example, uncritically, agglomeration is assumed to be a source of external economies, reducing transaction costs when there is empirical evidence that agglomeration offers not cheaper production, even in transaction cost terms, but simply easier production

STIMSON PAGINATION (M2469).indd 43

20/12/2010 15:12

44

Endogenous regional development

in purely behavioural terms (Taylor, 1975). Similarly, the presence of knowledge in a place is assumed to lead to spillover effects without any real conceptualization of the transmission mechanism through which knowledge is supposed to diffuse between individuals and/or regional economies. Even casual empiricism would suggest that the facts that: contract law is one of the fastest-growing components of the legal profession; the protection of IPR is a growing corporate concern (reflected in the ‘inimitability’ version of the competencies theory of the firm); and that the spat between Ferrari and McLaren teams in Formula 1 racing in 2007 was so acrimonious when an employee moved between the two organizations and took technical data with him that it brought a multi-million dollar fine all point to knowledge spillover effects being a far more complex process than endogenous regional theory would imply. The point we want to make here has been made before (Clark, 1998); that endogenous regional theory is based on stylized facts that need seriously to be unpacked. Even within the terms of reference of endogenous regional theory, in order to understand both the existence and persistence of local growth differentials, we need to understand the nature of the mechanisms that are hypothesized to diffuse technological change through a regional economic system. Certainly, this form of analysis offers a way to prioritize the theorized drivers of regional economic change that ‘new regionalism’ lacks. However, the whole comes together not as a consistently articulated theory of local economic growth but as a series of conceptual and/or empirical explorations of the components of what might be developed into something fuller. In practice, attempts to bridge the gap between abstract, but untestable, models and the ‘messy’ contingency of the social world using growth econometrics has produced some insights, but can be more broadly caricatured by the use of sophisticated techniques and poor data to make inferences and gain insights into the processes driving local economic growth (Plummer, 2001). New Regionalism and Embeddedness In the past decade, an increasing body of research has built upon Granovetter’s (1985) concept of ‘embeddedness’ to explain the dynamics of local economies. This perspective emphasizes the role of social relations in economic transactions, holding that anonymous markets are non-existent and that economic life and transactions are built on social connections. The concept of embeddedness has given rise to a powerful model of local economic growth that draws on a range of complementary literatures on ‘new industrial spaces’, ‘learning regions’, ‘innovative milieus’ and ‘regional innovation systems’ (e.g., Storper, 1997; Braczyk

STIMSON PAGINATION (M2469).indd 44

20/12/2010 15:12

A geographer’s perspective and interpretation

45

et al., 1998; Porter, 1998; MacKinnon et al., 2002) – now more generally captured under the term ‘new regionalism’ (Rainnie and Grobbelaar, 2004). Building directly on the ideas of Granovetter, the new regionalist writing suggests that economic growth is linked not only to market conditions, but also to repeated inter-firm interaction and knowledge exchange, collaborative long-term buyer–supplier relationships, the creation of social capital (including trust, reciprocity and loyalty) and a supportive tissue of local institutional thickness (see Putnam, 1993; Cumbers et al., 2003; Malmberg and Maskell, 2006). The ‘new’ regionalism and embeddedness ideas have prompted renewed interest in the advantages of geographical proximity between firms in related industries (Keeble and Nachum, 2002) and it is a significant shift away from traditional understandings of agglomeration based on input– output relations and transaction costs, focusing instead on the social and institutional drivers of growth. Whilst the approach has theorized new drivers and has added richness to the local growth debate, it has been less rigorous in how we might measure those processes or detect their impact on long-run economic growth. It is also a model of economic growth with great policy appeal, but the application of this new orthodoxy has often been implemented uncritically and in the absence of detailed empirical evaluation (i.e., evidence-based policy). The new regionalism approach may be popular but it also has significant and prominent limitations (see also Taylor, 2005). First and most importantly, the mechanisms of local economic growth transmission are not clearly articulated in the model, and this limitation has four interconnected components: 1.

2.

The model lacks a meaningful explanandum. It does not measure growth or change, just ‘success’ in that the qualitative data on which the model is based are drawn from apparently successful regions and places, with the processes identified within them being taken to account for that ‘success’. As a consequence, sample selection bias is combined with a truncated and ‘censored’ distribution to increase the threats to validity when attempting to draw inferences about the processes driving local economic growth from self-elected case studies. Time, too, is poorly incorporated into the model, and local firm networks are seen to translate unproblematically and instantly into local structures of knowledge mobilization and exchange, learning, innovation and social capital that create growth. This is a caricature of transmission mechanisms, leaving us without any knowledge of how and when economic potentialities are actualized within a local economy. All that the model alludes to is a mass (or morass) of contingency.

STIMSON PAGINATION (M2469).indd 45

20/12/2010 15:12

46

3.

4.

Endogenous regional development

New knowledge is assumed to translate unproblematically into new business ventures, neglecting processes of coalition formation, third party referral, the commercial translation of ideas and knowledge, the securing of finance, for example. Transmission mechanisms are again ignored. The model fetishises proximity (Oinas, 1999). As a consequence, it underplays geographically more extensive transmission mechanisms and networks that might be as, if not more, important than those of a particular locality (see Glückler, 2007).

Second, the new regionalist model fails to recognize the importance of unequal power relations between firms; inequalities that privilege some and peripheralize others within both local and geographically more extensive networks (Taylor, 2000). Third, the model also fails to deal adequately with the capitalist imperatives of profit generation, the price mechanism and wage discipline as mechanisms of personal wealth creation (Hudson, 1999). Finally, much of the work on new regionalism is based on theoretical speculation, rather than empirical research, thus providing only a limited understanding of processes at work (Maskell and Malmberg, 1999). Empirical research has focused on high-technology clusters and dynamic growth regions, neglecting those localities, places and regions experiencing decline, industrial restructuring, or socioeconomic marginalization. Taken together, these limitations raise the spectre that without clearly articulated theories specifying the manner in which posited mechanisms operate both within and between local economies, the ‘new regionalism’ is in danger of degenerating into a morass of contingency, the qualitative analogue of over-fitting econometric models. It is not clear whether this observed contingency is the result of local capacities and competitive mechanisms working themselves out in different contexts or whether there are simply different drivers of local economic growth that operate idiosyncratically in each local economy at any time. As a corollary, the new regionalism fails to address the methodological questions of how, and in what ways, local economic capacities might be actualized in a particular empirical context. New regionalism, therefore, offers a differently limited caricature of regional economic processes. It is a caricature built on over-theorization in which layers of contingency have been laid one over another, obscuring as much as clarifying processes of growth and change. In part, this accounts for its popularity, especially in policy circles, because it can be all things to all people. Again, it is built on stylized facts, in this case relating to trust, networks, institutional support and knowledge transfer, for

STIMSON PAGINATION (M2469).indd 46

20/12/2010 15:12

A geographer’s perspective and interpretation

47

example. However, as more theorizing has been added to the model, it has become evident that there is no mechanism to say what the importance of the separate, postulated processes is in promoting or retarding regional economic growth. It is here that the modelling strategy of endogenous regional theory has much to offer.

ANALYTICAL CAPABILITIES AND POLICY ADVICE There is, perhaps, a more targeted and appropriate way to develop a richer understanding of growth and change at the level of the local economy that goes beyond the dualism of ‘stylized facts’ and ‘contingency’, and yet is capable of providing policy advice that can be used to formulate effective strategies that are capable of being implemented in practice. We advocate a mixed-method research design that is grounded in critically engaged pluralism (Taylor and Plummer, 2003). This necessitates moving beyond the types of econometric modelling strategies employed by endogenous regional theory and the intensive case studies of qualitative geographers to triangulate our knowledge claims about local economies. In practice, this entails a research design that can accommodate and use a blend of qualitative and quantitative information to articulate local capacities and transmission mechanisms and evaluate empirically the likelihood that these capacities are actualized in any particular context (Plummer and Taylor, 2001a, 2001b). In short, we reject neither approach to regional analysis but suggest that their combination offers important synergies for both analysis and policy formulation. It is the combination of these approaches that we illustrate in the analyses of the remaining sections of this chapter. A recently completed study relating to policy advice on vocational education and training undertaken for the National Centre for Vocational Education Research (NCVER) in Australia (Garlick et al., 2007) explored the potential and limitations of such a mixed-method research design. This study was motivated by the need to develop evidence-based policy outcomes in the context of vocational education and training (VET) in Australia. We can use our experience of trying to implement this research project to illustrate the virtues of a pluralist methodology when developing analysis to inform policy advice. Based upon these methodological norms, this study produced controversial and provocative conclusions that, we would contend, are essential if new policy approaches are to be developed that go beyond repeating the tired and increasingly hackneyed initiatives suggested in the one-size-fits-all ‘guru’ theorizing on ‘clusters’ and ‘creatives’.

STIMSON PAGINATION (M2469).indd 47

20/12/2010 15:12

48

Endogenous regional development

Theoretically Informed Empiricism The Garlick et al. (2007) study of vocational education and training (VET) provision in Australia focused on providing advice on policy alternatives that might better target VET to foster regional economic growth that was consistent with the long-run processes shaping those regional economies. The study asked two key questions: (1) What processes are driving economic change at the regional scale in Australia? (2) How can VET work sympathetically and in a targeted manner to modify those processes, create growth and, therefore, more fully realize national economic potentials? From an institutionalist perspective, successful local economies are said to depend on complex processes of integration and embedding – built on trust, reciprocity and loyalty – that create social capital. They involve the exchange of information, ideas and innovation through mechanisms of quasi-integration that creates new knowledge through processes of situated learning. In addition, the support of local institutions (labelled as institutional thickness; Amin and Thrift, 1994) is seen as bolstering these local economic processes. Enmeshed in webs of global coordination and value transfer, these local economies are seen as nodes of untraded interdependencies (Storper, 1997) that are tapped into by large corporations and TNCs, which, for their part, act as global information arbitrageurs (Economist, 2003). To reflect institutionalist thinking on the propulsive elements of local economic growth, eight ‘drivers’ were distilled from the broad range of theories, each of which prioritized different sub-sets of drivers as operating in different ways, depending on local context. These were: ● ● ● ● ● ● ● ●

technological leadership at the enterprise level; knowledge creation and access to information; the local integration of small firms – their propensity to deal one with another within a locality; infrastructure support and institutional thickness; human resources – the skills and capacities of a local workforce; the power of large corporations – the assets and authority they are able to wield; local demand and inter-regional trade; local sectoral specialization – the existing specialist skills and knowledge of a locality or region (see Plummer and Taylor, 2001a).

To perform the ‘quantitative’ element of the multi-method project, these drivers were incorporated into a growth econometrics model consistent with existing models derived from endogenous regional theory.

STIMSON PAGINATION (M2469).indd 48

20/12/2010 15:12

A geographer’s perspective and interpretation

49

The eight drivers distilled from theory were measured empirically using a unique Australian data set generated across Australia between 1984 and 2002, with the country divided into 94 functional regions. The nature and origins of the data set are outlined in Appendix 3A.1 at the end of this chapter. These empirical measures were incorporated into a conventional gap convergence (mean reversion) econometric model (Plummer and Taylor, 2001b; Taylor and Plummer, 2003) in which regional growth (measured in terms of unemployment relativities) is broken down into three components: 1. 2. 3.

transition dynamics: the speed at which a region’s growth rate returns to a long-run equilibrium after a shock; that is, mean reversion; ‘structural’ characteristics: the extent to which there are growth differences between neighbouring regions; random shocks: the unanticipated and unpredicted factors that can impact on the regional growth rate (Martin and Sunley, 1998).

The modelling exercise is fully explained in Garlick et al. (2007). Here, however, it is the final model specification that is important. The findings are quite stark. Regional growth in Australia in the last two decades has been driven by four mechanisms: 1. 2. 3. 4.

human capital: fundamentally the local stock of skilled and educated people; access to high technology and technological leadership at the enterprise level; industry specialization: reflecting local stocks of specialist knowledge; while, interestingly, government intervention: through direct institutional support held back regional growth.

Very clearly, human capital and enterprise are the life blood of regional economic growth, especially when there is appropriate government intervention. Qualitative Information and Facilitated Workshops However, the study recognized that not only must positive drivers of growth be found at the regional scale as the quantitative modelling shows, but, to achieve success, it is also important that growth transmission processes at that scale are faced by as few impediments as possible. To address this aspect of the local growth process, qualitative information was

STIMSON PAGINATION (M2469).indd 49

20/12/2010 15:12

50

Table 3.1

Endogenous regional development

Australian case study regions and their characteristics

Region (State)

Growth Characteristics

Orange (NSW) Western Sydney/ Penrith (NSW) Wollongong (NSW) Mount Isa (Qld) Wide Bay Burnett (Qld) Shepparton (Vic.) Horsham (Vic.) Burnie (Tas.) North-East Adelaide (SA) Pilbara/Port Hedland (WA) Alice Springs (NT)

A rural city with high growth A peri-urban metropolitan area with high-technology enterprise A provincial city with static growth A remote mining centre with low growth A region with rural cities, and low and declining growth Centred on a rural city and shows virtually no growth Centred on a rural city, and with relatively high growth A rural city with declining growth A peri-urban metropolitan area with low growth Remote mining communities with rapidly declining growth A remote city with recent relatively high growth

Source: Authors’ compilation.

collected through facilitated workshops in 11 case study regions (Garlick et al., 2007), and this information adds significantly to the interpretation of growth dynamics across Australia’s regions. Workshops were held in regions across the spectrum of growth rates in Australia, spread across states and territories and embracing a mix of urban and rural regions. These regions and their growth characteristics are listed in Table 3.1. Between 12 and 30 people attended each workshop, with participants from VET and technical and further education (TAFE), local government, state government, business, schools, universities, regional development organizations and social development bodies. There were managers from small and large business, institutional managers, students, teachers, social entrepreneurs and others. A number of salient findings from these workshops are disturbing. In particular, there was a general inability to fully appreciate the global dynamic influencing business development. Also, there was an apparent naivety about the way companies use their capital and labour and the way networks are used for corporate rather than community gain. There was a penchant for off-the-shelf, ‘quick-fix’ solutions brought in from outside the region, and a parallel lack of local strategic thinking. In some regions, with economic problems, there was a disturbing tendency towards selfdelusion. Though local economic problems were recognized there was a

STIMSON PAGINATION (M2469).indd 50

20/12/2010 15:12

A geographer’s perspective and interpretation

51

tendency to argue that, of course, things had changed and that the future promised growth. This could only be interpreted as ‘the optimism of the immediate’. It highlights, however, the potential limitations of qualitative analyses and, possibly, the unreal expectation that consultants can create. Not all regions were the same, and in some outstanding initiatives had been taken. However, where plans were developed, there was a tendency for people to be transfixed by the plan and not to move to implementation. It was concluded that 20 years of bottom-up regional development policy and practice in Australia had produced disturbingly low levels of local action and that government-mandated regional leadership might not be the most effective way to encourage local growth that depends on free-flowing ideas and enthusiasm across the whole of a region (Garlick et al., 2007). On a more positive note, the workshops did recognize a range of impediments to growth transmission. In particular, they recognized a general complacency and lack of dynamism in some regions. They pointed to human capital being exported to major metropolitan regions across the country, with this exodus being seen as ‘normal’. Key industries were not seen as producing business spin-outs at the regional level, and that facilitative regional planning to realize opportunities appeared to be disorganized. At the same time, there was a recognition that the education system was unable to inspire a region’s human capital to be enterprising. What is more, the workshops saw an overemphasis on specific sectoral ‘winners’, pushed by institutions and consultants from outside the region, as the ‘silver bullets’ to achieve local economic growth rather than building on unique local capacities, attributes and abilities (ibid.). Enterprising Human Capital and Regional Engagement When the ‘extensive’ and ‘intensive’ analyses of regional economic growth in Australia are combined, a clear picture of its dynamics and transmission emerges that highlights the roles of enterprising human capital and regional engagement (ibid.). It is a picture that neither research strategy could generate independently of the other. Enterprising human capital, we would contend, is created by ‘enterprising’ people who take ideas and turn them into outcomes using the attributes, competencies and capabilities at their disposal. This is very different from the ‘creative’ human capital that Richard Florida (2002) has emphasized that sees people with skills and knowledge generating new ideas. The translation of new ideas into socially and commercially viable enterprises is a non-trivial process. Rather, it is a process at the very heart of the regional growth process, which needs to create wealth, income

STIMSON PAGINATION (M2469).indd 51

20/12/2010 15:12

52

Endogenous regional development

and jobs at least to maintain and possibly to enhance the well-being and standard of living of entire communities that are currently facing global competitive pressures. It involves having a strong grasp of, amongst many other things, markets and market potentials, sourcing, financing, business planning, accountancy, cost control, timing, coordination and negotiation. It involves outcome-orientated people who, at their most successful, are able to work in teams that blend complementary skills. The enterprising outcomes of these people are not necessarily just economic in the sense of creating new businesses. They may also emerge as new social, cultural and environmental outcomes that reshape the very nature of the communities within which they operate and may, in turn, create a climate for further ‘enterprising’ in that place. All the case study regions pointed to the lack of a local enterprising culture within them and none had mechanisms in place to facilitate the creation of such an enterprising capacity. Essential to the creation of enterprising human capital are mutual and creative connections among a region’s human capital – what more broadly can be called regional engagement. It is, at its most fundamental, a framework of mutual cooperation and dialogue that complements and extends the processes of enterprising human capital. However, it is more than a process. Regional engagement is also about achieving results and generating outcomes. It is the community equivalent of the enterprising human capital of individuals. Towards a Policy The question posed by these empirical findings on regional growth dynamics in Australia is: how do you translate these ideas into a policy that can be used to guide the development of vocational education and training (VET) in Australia’s regions? The report we prepared (Garlick et al., 2007) suggested four new roles for VET, building on local knowledge and skills, consistent with the processes shaping the country’s regional economies. Those roles are described below: 1.

VET should aim to develop competencies within communities by building ‘enterprising skills’. Consistent with the theoretically informed analysis, this role might be achieved in two ways: by developing locally the skills to enable enterprising opportunities of all types to be turned into outcomes; and creating greater knowledge and awareness of how the global economy works. The first of these approaches to competency building is essentially entrepreneurship education. It is hard, if not impossible, to teach people to be entrepreneurial.

STIMSON PAGINATION (M2469).indd 52

20/12/2010 15:12

A geographer’s perspective and interpretation

53

However, it is possible to equip those who are keen to try to take risks with the skills that can help them succeed, including: –

issues involved in identifying commercial and other ‘enterprising’ opportunities; – business planning; – marketing contracts – their drawing up and enforcement; – accessing venture capital and development finance; – entrepreneurialism; – networking; – risk management; – communication.

2.

3.

4.

The second approach to competency building is a broader educational provision that discusses the workings of the global economy, the competition it brings and the opportunities and limitations it offers together with the mechanisms to facilitate change. Here it is important to understand the inter-company relationships in supply chains, and the unequal power relationships that operate within them. VET can assist local capacity building by taking a lead in developing coalitions of people with ‘enterprising’ ideas in competitive areas, and equipping them with connections and synergies with other businesses and institutions. Firms are essentially temporary coalitions of people who deploy assets in an effort to achieve anticipated commercial outcomes. As yet, our understanding of the dynamics of those coalitions is relatively underdeveloped (Taylor, 2006), but it is an area where ‘intensive’ case study evidence can be used to great effect. VET needs to target 40+ year-olds to provide them with the skills to be ‘enterprising’ and to build businesses, and realize their capacities and capabilities. In this context, vocational education and training has the opportunity to realize the otherwise lost potential of this section of society as was pointed out in the workshop element of the Australian study. VET has an important role in bringing together information of a region’s knowledge resources. This is a significant additional facilitative role that can speed up processes of ‘enterprising’.

The reaction to these policy proposals for VET has proved very contentious. It proposes a very different educational agenda from the one that has traditionally been followed. It makes educators uncomfortable because it disturbs people’s comfort zones and involves doing something new!

STIMSON PAGINATION (M2469).indd 53

20/12/2010 15:12

54

Endogenous regional development

CONCLUSION In developing a geographic perspective on endogenous processes operating in regional and local economies, this chapter has sought to highlight the differences in methodology in this discipline compared with economics: between the ‘intensive’ qualitative approach in geography’s ‘new regionalism’ and the ‘extensive’ modelling strategy of economics’ endogenous growth theory. Both are distinctive caricatures of local economic processes built on sets of differently stylized facts. It has been suggested that the two approaches are complementary and, through integration, have the potential to bring a fuller and more nuanced perspective on local economic growth processes and issues that are capable of reflecting the unique characteristics and capabilities of places rather than generating a one-size-fits-all policy prescription. As such it is an approach that has the potential to suggest new policy initiatives. A multi-method analysis of Australian regional change using theoretically informed empirical modelling coupled with qualitative information collected through facilitated workshops in case study regions, threw new light on regional growth processes in this economy. Focused on a review of VET in Australia, this approach was shown to have the capacity to generate new policy perspectives that fall outside the usual comfort zone of bureaucratic policy-makers. From this it might be suggested that a multimethod approach to the analysis of endogenous processes in regional economies might be a useful way of developing new regional policies and programmes.

REFERENCES Amin, A. and Thrift, N.J. (1994), ‘Living in the Global’, in A. Amin and N.J. Thrift (eds), Globalization, Institutions and Regional Development in Europe, Oxford University Press, Oxford, pp. 1–22. Braczyk, H.J., Cooke, P. and Heidenreich, M. (eds) (1998), Regional Innovation Systems: The Role of Governances in a Globalized World, UCL Press, London. Brakman, S. and Heijdra, B. (2004), The Monopolisitic Competition Revolution in Retrospect, Cambridge University Press, Cambridge, UK. Brakman, S., Garretson, H. and Von Marrewijk, C. (2001), An Introduction to Geographical Economics, Cambridge University Press, Cambridge, UK. Clark, G. (1998), ‘Stylized Facts and Close Dialogue: Methodology in Economic Geography’, Annals of the Association of American Geographers, 88(1), 73–87. Cumbers, A., MacKinnon, D. and Chapman, K. (2003), ‘Innovation, Collaboration and Learning in Clusters: A Study of SMEs in the Aberdeen Oil Complex’, Environment and Planning A, 35(9), 1689–706. Durlauf, S.N. and Quah. D. (1999), ‘The New Empirics of Economic Growth’,

STIMSON PAGINATION (M2469).indd 54

20/12/2010 15:12

A geographer’s perspective and interpretation

55

in J.P. Taylor and M. Woodford (eds), Handbook of Macroeconomics, NorthHolland, Amsterdam, pp. 235–308. Durlauf, S., Johnson, P. and Temple, J. (2004), ‘Growth Econometrics’, in P. Aghion and S.N. Durlauf (eds), Handbook of Economic Growth, North-Holland, Amsterdam. Economist, The (2003), ‘The New Geography of the IT Industry’, The Economist Newspaper, 368(8333), 53–5. Fingleton, B. (2001), ‘Equilibirum and Economic Growth: Spatial Econometric Models and Simulations’, Journal of Regional Science, 41(1), 117–48. Fingleton, B. (ed.) (2003), European Regional Growth, Springer-Verlag, New York. Fingleton, B. (ed.) (2007), New Directions in Economic Geography, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Florida, R. (2002), The Rise of the Creative Class: and How It’s Transforming Work, Leisure, Community and Everyday Life, Basic Books, New York. Fujita, M., Krugman, P. and Venables, A. (1999), The Spatial Economy: Cities, Regions, and International Trade, MIT Press, Cambridge, MA. Garlick, S., Taylor, M. and Plummer, P. (2007), An Enterprising Approach to Regional Growth: The Role of VET in Regional Development, report prepared for the National Centre for Vocational Education Research (NCVER), funding provided by the Australian National Training Authority (ANTA). Glaeser, E. (2000), ‘The New Urban and Regional Growth’, in G. Clark, M. Gertler and M. Feldman (eds), Handbook of Economic Geography, Oxford University Press, Oxford, pp. 83–99. Glückler, J. (2007), ‘Geography of Reputation: The City as the Locus of Business Opportunity’, Regional Studies, 41(7), 949–61. Granovetter, M. (1985), ‘Economic Action and Social Structure: The Problem of Embeddedness’, American Journal of Sociology, 91(3), 481–510. Hudson, R. (1999), ‘The Learning Economy, the Learning Firm and the Learning Region: A Sympathetic Critique of the Limits of Learning’, European Urban and Regional Studies, 6(1), 59–72. Jones, C. (1998), Introduction to Economic Growth, Norton, London. Keeble, D.E. and Nachum, L. (2002), ‘Why do Business Service Firms Cluster?’, Transactions of the Institute of British Geographers, 27(1), 67–90. King, G., Keohane, R. and Verba, S. (1994), Designing Social Inquiry: Scientific Inference and Qualitative Research, Princeton University Press, Princeton, NJ. Krugman, P. (1991), Geography and Trade, MIT Press, Cambridge, MA. MacKinnon, D., Cumbers, A. and Chapman, K. (2002), ‘Learning, Innovation and Regional Development: A Critical Appraisal of Recent Debates’, Progress in Human Geography, 26(3), 293–311. Malmberg, A. and Maskell, P. (2006), ‘Localized Learning Revisited’, Growth and Change, 37(1), 1–18. Martin, R. and Sunley, P. (1998), ‘Slow Convergence?: The “New” Endogenous Growth Theory and Regional Development’, Economic Geography 74(3), 201–27. Maskell, P. and Malmberg, A. (1999), ‘Localised Learning and Industrial Competitiveness’, Cambridge Journal of Economics, 23(2), 167–90. Massey, D. and Meegan, R. (1985), Politics and Method: Contrasting Studies in Industrial Geography, Routledge, London. Neary, P. (2001). ‘Of Hype and Hyperbolas: Introducing the New Economic Geography’, Journal of Economic Literature, 39(2), 536–61.

STIMSON PAGINATION (M2469).indd 55

20/12/2010 15:12

56

Endogenous regional development

Oinas, P. (1999), ‘Activity-Specificity in Organizational Learning: Implications for Analysing the Role of Proximity’, GeoJournal, 49(4), 363–72. Overman, H. (2004), ‘Can We Learn Anything From Economic Geography Proper?’, Journal of Economic Geography, 4(5), 501–16. Plummer, P. (2001), ‘Vague Theories, Sophisticated Techniques, and Poor Data’, Environment and Planning A, 33(5), 761–4. Plummer, P. (2007), ‘Economic Geography, by the Numbers’, in A. Tickell, E. Sheppard, J. Peck and T. Barnes (eds), Politics and Practice in Economic Geography, Sage, London. Plummer, P. and Sheppard, E. (2006), ‘Geography Matters: Agency, Structures and Dynamics at the Intersection of Economics and Geography’, Journal of Economic Geography, 6(5), 619–37. Plummer, P. and Sheppard, E. (2007), ‘Toward Critically Engaged Pluralism in Geographical Debate’, Environment and Planning A, 6(5), 1275–81. Plummer, P. and Taylor, M. (2001a), ‘Theories of Local Economic Growth: Concepts, Models and Measurement’, Environment and Planning A, 33(2), 219–36. Plummer, P. and Taylor, M. (2001b), ‘Theories of Local Economic Growth: Model Specification and Empirical Validation’, Environment and Planning A, 33(3), 385–98. Plummer, P. and Taylor, M. (2003), ‘Theory and Praxis in Economic Geography: “Enterprising” and Local Growth in a Global Economy’, Environment and Planning C, 21(5), 633–49. Polanyi, K. (1957), The Great Transformation, Beacon Press, Beacon Hill, MA. Porter, M. (1998), On Competition, Harvard Business School Press, Boston, MA. Putnam, R. (1993), Making Democracy Work: Civic Traditions in Modern Italy, Princeton University Press, Princeton, NJ. Rainnie, A. and Grobbelaar, M. (eds) (2004), New Regionalism in Australia, Ashgate, Aldershot. Sayer, A. (1984), Method in Social Science: A Realist Approach, Hutchinson, London. Sheppard, E. (2000), ‘Geography or Economics?; Conceptions of Space, Time, Interdependence, and Agency’, in G. Clark, M. Gertler and M. Feldman (eds), Handbook of Economic Geography, Oxford University Press, Oxford, pp. 99–125. Storper, M. (1997), The Regional World: Territorial Development in a Global Economy, Guilford Press, New York. Taylor, M. (1975), ‘Organizational Growth, Spatial Interaction and Location Decision-Making’, Regional Studies, 9(4), 313–23. Taylor, M. (2000), ‘Enterprise, Power and Embeddedness: An Empirical Exploration’, in E. Vatne and M. Taylor (eds), The Networked Firm in a Global World, Ashgate, Aldershot, pp. 199–234. Taylor, M. (2005), ‘Embedded Local Growth: A Theory Taken Too Far?’, in R. Boschma and R. Kloosterman (eds), Learning from Clusters: A Critical Assessment from an Economic-Geographical Perspective, Springer, Dordrecht, pp. 69–88. Taylor, M. (2006), ‘The Firm: Coalitions, Communities and Collective Agency’, in M. Taylor and P. Oinas (eds), Understanding the Firm: Spatial and Organizational Dimensions, Oxford University Press, Oxford, pp. 87–116. Taylor, M. and Plummer, P. (2003), ‘Promoting Local Economic Growth: The Role of Entrepreneurship and Human Capital’, Education and Training, 45(8/9), 558–64.

STIMSON PAGINATION (M2469).indd 56

20/12/2010 15:12

A geographer’s perspective and interpretation

57

APPENDIX 3A.1 THE NATURE AND ORIGINS OF THE DATA SET The variables used in this analysis were generated initially in the 1980s and early 1990s for a series of studies published through the Australian Government Publishing Service (AGPS) in the Office of Local Government’s (OLG) Australian Regional Developments series. The publications were part of a large database describing the economic, social and accessibility vulnerability and resilience of 94 Australian regions (see Australian Regional Developments monograph A Regionalisation of Australia for Comparative Economic Analysis, 1988). The data are unique, developed from initially unpublished data sources and major empirical analyses. Where possible, they continue to be updated. Technological leadership at the enterprise level (HITECH) The data surrogate for this measure is the regional significance of R&D-intensive industries. The Department of Industry, Technology and Commerce (DITAC) identified the following ASIC industry groups and classes as Measures of Science and Innovation (1987, pp. 378–9) containing high-technology components: (1) Pharmaceutical and Veterinary produce; (2) Aircraft Manufacturing; (3) Photographic, Professional and Scientific Equipment: (4) Data Processing Services; (5) Research and Scientific Institutions. The HITECH measure is calculated as the proportion of employment in each region in these four-digit ASIC industries using the unpublished data from the Integrated Regional Information System (IRIS) database. Knowledge creation and access to information (INFOACC) This measure is based on a simple interaction model in which the size of information activity at a place is measured as employment in professional and managerial jobs in each region, and the distance between pairs of regions measured as time-distance by the quickest means available. The measure is more fully described in the OLG Working Paper, ‘Accessibility and Remoteness’ (1992, pp. 21–4), which describes six elements of regional accessibility (access to goods and services, access to intermediate goods markets, the cost of access to intermediate goods markets, access to export ports, access to information and the cost of access to information). The local integration of small firms (MLOCN) This measure is the inverse of the proportion of establishments in a region that belong to multi-location enterprises, taken from the Australian

STIMSON PAGINATION (M2469).indd 57

20/12/2010 15:12

58

Endogenous regional development

Bureau of Statistics’ IRIS database. This enables the removal from the region of establishments with the weakest local affiliations. Infrastructure support and institutional thickness (PROT) This measure is the regional effective rate of industry protection drawn from the OLG Working Paper (1992) ‘The Regional Impact of Changing Levels of Protection in Australian Industries’. A weighted averaging method has been used to allocate industry-effective rates of protection to regions using sectoral employment levels using the IRIS database. Local human resource base (DEGREE) The measure used here is the proportion of the population in each region without university degrees. This is a surrogate not only for local skill levels but also for issues of income and, indirectly, for the local availability of capital. The power of large corporations (TOTPOP) An index of corporate control in the region was constructed using the head office regional address for companies listed in the top 1000 in the Business Review Weekly magazine. Employment for each business was expressed as a quotient of the total employment of the particular region in which it is located using IRIS data. Inter-regional trade (MKTACC) The measure for this driver for regional growth is taken as a region’s accessibility to intermediate goods markets within Australia and has been taken from the OLG Working Paper ‘Accessibility and Remoteness’ (1992, p. 16). It is expressed as a function of the employment level in each pair of regions in intermediate goods industries such as manufacturing and construction, and the direct road distance between the pairs of regions. Local sectoral specialization (SPEC) This is a simple measure of the numbers of business establishments taken from the IRIS database for each industry category in each region.

STIMSON PAGINATION (M2469).indd 58

20/12/2010 15:12

4.

Endogenous rural development from a sociological perspective Frank Vanclay

DEFINING ERD SOCIOLOGICALLY For many sociologists and some social geographers1 endogenous regional development is synonymously ‘endogenous rural development’ (both abbreviated ‘ERD’). It is a multidimensional concept, ‘a multi-level, multi-actor and multi-faceted process’ (Van der Ploeg et al., 2000, p. 391). Much more than a statement about the origin of development initiatives, in the sociological perspective ERD has layers of meaning. Above all, ERD is a new paradigm, worldview and philosophy about appropriate development, rural renewal and a multifunctional, post-productivist countryside (Slee, 1994; Van Broekhuizen et al., 1997; Ray, 1997; Van der Ploeg et al., 2000; Marsden, 2003; OECD, 2006). At a very basic level, if exogenous development is development that is initiated outside a local region (i.e., externally) then endogenous development is the opposite. In other words, ERD is development that is initiated and controlled by the local community. However, from a sociological perspective, ERD implies much more than this. Whereas exogenous development is seen as modernist, Fordist and top-down, ERD is seen as bottom-up and a reaction to modernization, or as resistance to it (Bassand et al., 1986). To some extent, post-modernism and post-Fordism are characteristics of ERD in that it accommodates niche-marketing of value-added product and flexible specialization, while at the same time valorizing local culture, tradition, artisanal production and regional typical food. In general, ERD refers to the utilization and celebration of local and place-based dimensions of a region as the basis of its economic activity and livelihood. The emphasis in most sociological approaches to ERD is in understanding the characteristics (natural, human and cultural) of a place that makes it special and/or distinctive (different from other regions), and how these may become the focus of sustainable economic activity. Some sociologists critique the concept and especially how it has been 59

STIMSON PAGINATION (M2469).indd 59

20/12/2010 15:12

60

Endogenous regional development

appropriated by the state to devolve not power and control but responsibility and costs (see, for example, Herbert-Cheshire, 2000). ERD has had much policy significance in the European Union (EU) because of the realization that conventional subsidies to agriculture were not effective in promoting growth or development in rural areas and that a new approach focusing on local potentialities was needed (OECD, 2006). Considerable interest in ERD occurred because of the EU LEADER programme (discussed later in this chapter), which ‘was conceived as an integrated and endogenous approach to rural development’ (ibid., p. 90). There is not a singular coherent position to ERD in the sociology discipline, but a range of approaches. Furthermore, there is no agreed-upon definition (Ray, 1999; Van der Ploeg et al., 2000). Nevertheless, perhaps Long and Van der Ploeg (1994) best sum up the sociological perspective. For them ERD is ‘founded mainly, though not exclusively, on locally available resources, such as the potentialities of the local ecology, labour force, knowledge, and local patterns for linking production to consumption, etc’ (pp. 1–2). ERD and exogenous development should be seen as a dualism of ideal types that blend in the development strategies of regions rather than as a mutually exclusive dichotomy (Cristovao et al., 1994; Kneafsey et al., 2001). They are, in essence, a heuristic device (Van der Ploeg et al., 2000).

CHARACTERISTICS AND VALUES OF ERD Despite the lack of a common definition, there is general agreement around the characteristics of ERD in the sociological perspective. Expanding on Slee (1994) and Bowler (1999), the key elements are: ● ● ● ● ● ● ● ●

a goal to create diversified, resilient and sustainable local economies; local determination of development options; local control over the development process; retention of benefits locally; utilization of locally available resources (natural, human and cultural); valorization of ‘the local’ and ‘place’, especially what is locally unique or special, and respect for local values; awareness of the rural as being post-productivist – that is, being a site of consumption as well as a site of production; appreciation of multifunctionality.

The multifunctionality element recognizes that rural areas (and agriculture) provide a range of non-market (as well as market) goods and services

STIMSON PAGINATION (M2469).indd 60

20/12/2010 15:12

A sociological perspective

61

such as environmental protection, landscape management, preservation of biodiversity and habitat protection, ecosystem services, carbon sinks, maintenance of cultural heritage, employment and livelihoods for rural people and food security. The social values implied by ERD in a sociological approach primarily relate to development as a social concept rather than to economic growth (Brugger, 1986). Perhaps in contrast to some other chapters in this book, ERD should not be taken to infer regional economic growth! Sustainability, in its broadest possible understanding, is a key value (Murdoch et al., 1994). Local control and self-determination are also fundamental – ERD highly values the rights of people to have a say in the things that affect their lives. The principle of subsidiarity – that decisions should be taken as closely as possible to the citizen and that larger groupings should encourage the autonomy of smaller groupings (Carozza, 2003) – is firmly embodied in the sociological and policy approach to ERD. Notions of equity, capacity building, community development and building resilient communities that have vitality, viability and health, are all part of the concept.

CONTRASTING ENDOGENOUS AND EXOGENOUS ERD arose because of the failing of exogenous development to deliver benefits to rural areas. Exogenous development, with its focus on economies of scale and concentration of activity, has led to the industrialization of urban growth poles, despite regions attempting to attract multinational corporations by providing incentives, lowering taxes and sometimes reducing other regulations in a ‘race to the bottom’. In that model, the role of rural areas was to provide cheap food, and as a result they were increasingly marginalized (Lawrence, 1986; Vanclay and Lawrence, 1993, 1995; Vanclay, 2003). In the words of Ward et al. (2005, p. 4; italics original): [Exogenous development] was criticised as dependent development, reliant on continued subsidies and the policy decisions of distant agencies and boardrooms. It was seen as distorted development, which boosted single sectors, selected settlements and certain types of business (e.g. progressive farmers) but left others behind and neglected the non-economic aspects of rural life. It was cast as destructive development which erased the cultural and environmental differences and considered to be dictated development devised by external experts and planners from outside local rural areas.

There has been some criticism of ERD along the lines that it is a nice ideal but not practical, that local areas will never be free of external

STIMSON PAGINATION (M2469).indd 61

20/12/2010 15:12

62

Endogenous regional development

influences especially in a globalizing world, that rural will always exist in juxtaposition to urban, that local will always interact with extra-local, and that exogenous and endogenous processes need to be balanced (Lowe et al., 1995; Ray, 2001a; Ward et al., 2005). As a result, Ray (2001a) coined the term, ‘neo-endogenous (rural) development’ to mean ‘endogenous-based development in which extra-local factors are recognised and regarded as essential but which retains a belief in the potential of local areas to shape their future’ (p. 4). Ray’s term ‘neo-endogenous development’, has generated much discussion in the academic literature and at conferences. Personally I am not fully convinced of its merits – as indicated earlier, the concepts were never intended to be a mutually exclusive dichotomy. However, perhaps the popularity of the new term can be attributed both to the need for increased emphasis that an integrated approach is necessary, and the erroneous assumption that ERD was necessarily in opposition to extra-local factors. ERD is meant to stimulate local business to supply local markets, and also to create external interest in local places and stimulate place-based consumption, for example through ecotourism, cultural tourism, farm-stays, rural B&Bs and the purchase of place-based products for consumption in situ or at home. External interest is thus vital to an ERD approach.

THE CULTURE ECONOMY APPROACH In ERD the focus of attention of development initiatives is local resources, not as traditional inputs for large extra-region or multinational firms, but as the basis of the economic activity of local small to medium firms. The concept of what a resource or asset can be thus expands from natural and human capital to cultural capital. Ray (1998) calls this the culture economy approach to rural development. With the valorization of local culture, many things potentially become of interest to people with spare time and/or disposable income in an information-rich, post-Fordist, postmodern society. Expanding considerably from Ray (1998, 2001a), the potential local cultural assets and how they could be utilized by regions to promote tourism and/or add value through place-based marketing are listed in Appendix 4A.1 at the end of the chapter. Regions that have highly significant features, for example that might qualify for UNESCO World Heritage status, have a clear advantage. The intention with the culture economy approach is for all regions to identify the particular strengths they each have that could be capitalized on, even though they may not be globally significant. The list in Appendix 4A.1, long as it is, nonetheless remains a small

STIMSON PAGINATION (M2469).indd 62

20/12/2010 15:12

A sociological perspective

63

selection of the many possible cultural assets regions can identify to make their region special. Many regions may also have natural resource assets that would also attract interest. Perhaps it should be noted that there has been some critique of this approach. Graham Day (1998), for example, argues that it shows a misunderstanding of the concept of culture in that it focuses on the material artefact rather than on social relationships and shared understandings, meanings and values.

THE SIGNIFICANCE OF THE EU ‘LEADER’ PROGRAMME LEADER is an acronym deriving from the French expression ‘Liaison Entre Actions de Développement de l’Economie Rurale’,2 which means ‘links between actions for the development of the rural economy’ (European Commission, 2007). The LEADER programme was conceived as ‘an integrated and endogenous approach to rural development’ within the European Union (OECD, 2006, p. 17). Ray (2001b, p. 280) suggests that: the official rationale of the LEADER intervention portrays it primarily as a tool with which to redirect the trajectory of the rural economy so as to meet concerns about the failure of regional economic convergence, the fragility of primary sectors of the economy (especially agriculture) and the financial burden of the [Common Agricultural Policy].

It was targeted at ‘territories’ of less than 100 000 inhabitants. Phases There were three phases to the LEADER programme: ● ● ●

LEADER 1 from 1991 to 1994; LEADER 2 from 1994 to 1999; LEADER + from 2000 to 2006.

The LEADER programme is now regarded as having ‘reached a level of maturity enabling rural areas to implement the LEADER approach more widely in mainstream rural development programming’ (European Union, 2005, Clause 50). From 2007 on, the LEADER approach will continue to persist, but has been integrated into mainstream rural development programmes (European Commission, 2006). LEADER  1 with a budget of EUR417 million was implemented in 217 territories. LEADER 2 was

STIMSON PAGINATION (M2469).indd 63

20/12/2010 15:12

64

Endogenous regional development

applied in approximately 1000 territories with a budget of EUR1755 million (LEADER European Observatory, 2001, p. I.14). There were 893 LEADER  + local action groups in 2004 prior to enlargement (i.e., admitting the new member nations) with a budget of EUR2143 million (European Commission, 2007, pp. 2, back cover). Components There are seven components to the LEADER approach (condensed from European Commission, 2007, pp. 2–3): 1.

2.

3.

4.

Area-based approach: this entails defining a local development policy on the basis of an area’s particular situation, strengths and weaknesses. Areas should be fairly homogeneous and characterized by internal social cohesion, shared history and traditions, a sense of common identity, and so on. Bottom-up approach: this aims to encourage participatory decisionmaking at local level for all development policy aspects. Its objective is the involvement of local players, including the community as a whole, economic and social interest groups and representatives of public and private institutions. The bottom-up approach relies on two major activities, ‘animation’ (facilitation of activities) and training of local communities. It is important that the project is initiated by local actors and that the public concerned with the action has been consulted. Partnership approach and the local action group: the LAG is a body of public and private actors, united in a partnership that identifies a joint strategy and a local action plan for developing an area. The LAG represents a model of organization that can influence the institutional and political balance of the area concerned. The LAGs provide appropriate mechanisms for participation, awareness raising and organization of local actors in favour of rural development. Innovation: LEADER demands that proposed actions are innovative. They may be: actions to promote local resources in new ways; actions that are of interest to local development but not covered by other development policies; actions providing new answers to the weaknesses and problems of rural areas; or actions that create a new product, new process, new forms of organization, or a new market. Innovation is also embodied in the programme’s pedagogical and networking components: disseminating information to other groups of players wishing to gain inspiration from achievements elsewhere, or to carry out joint projects.

STIMSON PAGINATION (M2469).indd 64

20/12/2010 15:12

A sociological perspective

5.

6.

7.

65

Integrated approach: the actions and projects in the local action plan should be linked and coordinated as a coherent whole. Integration may concern actions within a single sector, all programme actions or specific groups of actions, or, most importantly, links between the different economic, social, cultural, environmental actors involved in the area. Networking and cooperation between areas: by facilitating the exchange and circulation of information on rural development policies and the dissemination and transfer of good practice and innovative strategies and actions, the LEADER network aims to limit the isolation of LAGs and to create a source of information and analysis of the actions. Some LAGs have spontaneously organized themselves into informal networks. Another core part of LEADER is the cooperation between rural areas. Cooperation between areas can be transnational but may equally take place between areas within the same member state. Local financing and management: delegating a large proportion of the decision-making responsibilities for funding and management to the LAG is a key element of the LEADER approach. However, the degree of autonomy varies considerably depending on the member state’s specific mode of organization and institutional context.

Significance LEADER was significant in many ways. The foundations of LEADER directly addressed the sociological critique of earlier models of development and incorporated the sociological advocacy for a more participatory, cultural economy approach. Sociologists were directly involved in devising LEADER as an EU-level framework and often at national and local levels. They were involved in evaluating individual LEADER projects and the LEADER programme as a whole. The various evaluations of LEADER at project, member-nation and programme-wide (EU) level, although identifying a number of specific shortcomings (see, for example, Barke and Newton, 1997), overwhelmingly praised the approach. While Ray (2000) somewhat cynically argued that the funding levels were a very small proportion of the EU funding allocated for rural development and that this reflected a lack of commitment, he also admitted that LEADER was of considerable sociological interest. From an Australian perspective, a country that eschews rural subsidies, the fact that the approach potentially can be applied with a low level of government funding is important. It was feasible in Europe to divert finances from a very expensive system of payments to farmers under the Common Agriculture Policy to fund LEADER, especially since this was expected to

STIMSON PAGINATION (M2469).indd 65

20/12/2010 15:12

66

Endogenous regional development

have a greater return on investment (benefit to rural areas) and because this partly addressed World Trade Organization concerns about protectionism and trade distortion. Whether such an approach would be feasible, politically and otherwise, in countries without Europe’s resources and past experience of government investment in rural areas is another question. Therefore, the fact that this programme can be carried out ‘virtually without money’ (ibid., p. 165) is a virtue.

CONCLUSION Endogenous rural development (ERD) is a powerful way of renewing and revitalizing the countryside. Its benefits have been documented in the OECD (2006) report, The New Rural Paradigm, and in much literature emanating from the LEADER programme (e.g., European Commission, 2007). Because of a focus on local resources and on mobilizing (‘animating’ in LEADER language) local community activity, ERD has far greater potential to be sustainable than, typically, do exogenous forms of development (Gralton and Vanclay, 2006, 2009). The benefits of ERD extend far beyond what is normally considered by traditional economic indicators or measured as ‘growth’. They include, for example: ● ● ● ● ●

increased pride in where people live; increased sense of being part of a community; increased interest in participating in community activities; increased social networks (social capital); an increased sense of place (Vanclay, 2008).

All of these can be summarized as increased social well-being and quality of life. Most likely this translates into improved health as people lead more active, engaged lives. Despite the practical experience provided by LEADER, ERD is primarily a philosophy about appropriate development at the local level. ERD argues for a shift in thinking about development initiatives and what local resources might entail, and how endogenous and exogenous development can work together for sustainable regional development.

NOTES 1. As I have been requested to take a ‘sociological perspective’, I will use the words ‘sociology’, ‘sociologist’ and ‘sociological’. However, I point out that a sociological perspective

STIMSON PAGINATION (M2469).indd 66

20/12/2010 15:12

A sociological perspective

67

is not necessarily distinguishable from a more generic social science position and certainly blends with and draws heavily on the social and human geography literature. Further, although I have been a member of the Australian Sociological Association for over 25 years and am a former President of the International Rural Sociology Association, I also identify as a human geographer and am a member of the Institute of Australian Geographers and a fellow of the Royal Geographical Society in the UK. In general, I take an eclectic approach and ascribe to transdisciplinarity. 2. It is curious that the OECD report The New Rural Paradigm (OECD, 2006, p. 101) has a peculiar wording of ‘Liaison Entre Activités du Développement de l’Economie Rurale’ rather than what is most likely correct, ‘Liaison Entre Actions de Développement de l’Economie Rurale’, at least according to the majority of sources consulted including the FAQ website for LEADER + http://ec.europa.eu/agriculture/rur/leaderplus/faq_en.htm and the French language version of the EU portal: www.welcomeurope.com (both accessed 19 May 2010.).

REFERENCES Barke, M. and Newton, M. (1997), ‘The EU LEADER Initiative and Endogenous Rural Development: The Application of the Programme in Two Rural Areas of Andalusia, Southern Spain’, Journal of Rural Studies, 13(3), 319–41. Bassand,  M., Brugger, E.A., Bryden, J.M., Friedman, J. and Stuckey, B. (eds) (1986), Self-reliant Development in Europe: Theory, Problems, Actions, Gower, Aldershot. Bowler, I. (1999), ‘Endogenous Agricultural Development in Western Europe’, Tijdschrift voor Economische en Sociale Geografie, 90(3), 260–71. Brugger, E.A. (1986), ‘Endogenous Development: A Concept between Utopia and Reality’, in M. Bassand et al. (eds), Self-reliant Development in Europe, Gower, Aldershot, pp. 38–58. Carozza, P. (2003), ‘Subsidiarity as a Structural Principle of International Human Rights Law’, The American Journal of International Law, 97(1), 38–79. Cristovao,  A., Oostindie, H. and Pereira, F. (1994), ‘Practices of Endogenous Development in Barroso, Northern Portugal’, in J.D.  Van der Ploeg and A.  Long (eds), Born from Within: Practices and Perspectives of Endogenous Rural Development, Van Gorcum, Assen, pp. 38–58. Day, G. (1998), ‘Working with the Grain?: Towards Sustainable Rural and Community Development’, Journal of Rural Studies, 14(1), 89–105. European Commission (2006), The EU Rural Development Policy 2007–2013 (Fact Sheet), Office for Official Publications of the European Communities, Luxembourg. European Commission (2007), A Selection of Leader  + Best Practices 2007/1, European Observatory of Rural Areas, European Commission, Brussels. European Union (2005), ‘Council Regulation (EC) No. 1698/2005 of 20 September 2005 on support for rural development by the European Agricultural Fund for Rural Development (EAFRD)’, Official Journal of the European Union L  277 (21 October 2005). Gibson, C. and Connell, J. (2005), Music and Tourism: On the Road Again, Chanel View Publications, Clevedon. Gralton,  A. and Vanclay, F. (2006), ‘Quality Food Production, Diversity and Sustainability: Opportunities for Small Towns’, in M.  Rogers and D.  Jones

STIMSON PAGINATION (M2469).indd 67

20/12/2010 15:12

68

Endogenous regional development

(eds), The Changing Nature of Australia’s Country Towns, VURRN Press (Victorian Universities Regional Research Network Press), Ballarat, pp. 126–38. Gralton, A. and Vanclay, F. (2009), ‘Artisanality and Culture in Innovative Regional Agrifood Development: Lessons from the Tasmanian Artisanal Food Industry’, International Journal of Foresight and Innovation Policy, 5(1/2/3), 193–204. Herbert-Cheshire, L. (2000), ‘Contemporary Strategies for Rural Community Development in Australia: A Governmentality Perspective’, Journal of Rural Studies, 16(2), 203–15. Kneafsey,  M., Ilbery, B. and Jenkins, T. (2001), ‘Exploring the Dimensions of Culture Economies in Rural West Wales’, Sociologia Ruralis, 41(3), 296–310. Lawrence, G. (1986), Capitalism and the Countryside, Federation Press, Sydney. LEADER European Observatory (2001), LEADER, From Initiative to Method: A Guide to Teaching the LEADER Approach, available online at: http://www.fao. org/sard/static/leader/en/index.htm; accessed 20 May 2010. Long,  A. and Van der Ploeg, J.D. (1994), ‘Endogenous Development: Practices and Perspectives’, in J.D. Van der Ploeg and A. Long (eds), Born From Within: Practice and Perspectives of Endogenous Rural Development, Van Gorcum, Assen, pp. 1–6. Lowe, P., Murdoch, J. and Ward, N. (1995), ‘Networks in Rural Development: Beyond Exogenous and Endogenous Models’, in J.D.  Van der Ploeg and G.  Van Dijk (eds), Beyond Modernization: The Impact of Endogenous Rural Development, Van Gorcum, Assen, pp. 87–105. Marsden, T. (2003), The Condition of Rural Sustainability, Van Gorcum, Assen. Murdoch,  J., Ward, N. and Lowe, P. (1994), ‘Sustainable Agriculture and Endogenous Development: A Socio-Political Perspective’, in J.D. Van der Ploeg and A. Long (eds), Born From Within: Practice and Perspectives of Endogenous Rural Development, Van Gorcum, Assen, pp. 262–79. OECD (2006), The New Rural Paradigm: Policies and Governance, Organisation for Economic Co-operation and Development, Paris. Ray, C. (1997), ‘Towards a Theory of the Dialectic of Rural Development’, Sociologia Ruralis, 37(3), 345–62. Ray, C. (1998), ‘Culture, Intellectual Property and Territorial Rural Development’, Sociologia Ruralis, 38(1), 3–20. Ray, C. (1999), ‘Towards a Meta-framework of Endogenous Development: Repertoires, Paths, Democracy and Rights’, Sociologia Ruralis, 39(4), 521–37. Ray, C. (2000), ‘The EU LEADER Programme: Rural Development Laboratory’, Sociologia Ruralis, 40(2), 163–71. Ray, C. (2001a), Culture Economies: A Perspective on Local Rural Development in Europe, University of Newcastle, Centre for Rural Economy, monograph available online at: http://www.ncl.ac.uk/cre/publish/books.htm; accessed 19 May 2010. Ray, C. (2001b), ‘Transnational Co-operation Between Rural Areas: Elements of a Political Economy of EU Rural Development’, Sociologia Ruralis, 41(3), 279–95. Slee, B. (1994), ‘Theoretical Aspects of the Study of Endogenous Development’, in J.D.  Van der Ploeg and A.  Long (eds), Born From Within: Practice and Perspectives of Endogenous Rural Development, Van Gorcum, Assen, pp. 184–94. Van Broekhuizen,  R., Klep, L., Oostindie, H. and Van der Ploeg, J.D. (1997), Renewing the Countryside: An Atlas with Two Hundred Examples from Dutch Rural Society, Misset, Doetinchem.

STIMSON PAGINATION (M2469).indd 68

20/12/2010 15:12

A sociological perspective

69

Vanclay, F. (2003), ‘The Impacts of Deregulation and Agricultural Restructuring for Rural Australia’, Australian Journal of Social Issues, 38(1), 81–94. Vanclay, F. (2008), ‘Place Matters’, in F. Vanclay, M. Higgins and A. Blackshaw (eds), Making Sense of Place, National Museum of Australia Press, Canberra, pp. 2–11. Vanclay,  F. and Lawrence, G. (1993), ‘Social and Environmental Impacts of Economic Restructuring in Australian Agriculture’, International Journal of Sociology of Agriculture and Food, 3, 97–118. Vanclay,  F. and Lawrence, G. (1995), The Environmental Imperative: Ecosocial Concerns for Australian Agriculture, Central Queensland University Press, Rockhampton. Van der Ploeg, J.D., Renting, H., Brunori, G., Knickel, K., Mannion, J., Marsden, T., Roest, K., Sevilla-Guzman, E. and Ventura, F. (2000), ‘Rural Development: From Practices and Policies Towards Theory’, Sociologia Ruralis, 40(4), 391–408. Ward, N. et al. (2005), ‘Universities, the Knowledge Economy and “Neoendogenous Rural Development”’, Centre for Rural Economy Discussion Paper Series, 1, University of Newcastle upon Tyne.

STIMSON PAGINATION (M2469).indd 69

20/12/2010 15:12

70

Endogenous regional development

APPENDIX 4A.1 POTENTIAL LOCAL CULTURAL ASSETS ●

●

● ●

●

●

●

●

●

●

Regional typical food, traditional foods, bushfood (native or wild food), place-based foods and new foods (see Gralton and Vanclay, 2006, 2009). Winery trails, real-ale trails (and micro-breweries) and, for example (in Scotland especially) promoting visits to malt whisky distilleries and/or the sampling of malts at pubs near to where they are distilled. Regional languages and dialects and their associated cultures. Local folklore and, for example, developing and promoting community museums (ecomuseums), festivals and fairs that celebrate local culture and folklore. Local arts and crafts, especially those of high quality and that are inspired by and/or depict local settings and/or utilize local materials, for example, the extensive craftwork using Tasmanian native timbers (e.g., Huon Pine, myrtle, sassafras) that is on sale at the weekly Salamanca Craft Market and in Tasmanian galleries and boutiques. Local festivals, especially where they celebrate sense of place such as the Mountain Festival in Hobart, Tasmania; or some other activity with local relevance such as the Wooden Boat Festival in Tasmania, which partly celebrates Huon pine as a timber that was once used to make boats. Sites that have architectural significance, either because of a distinctive architectural style (e.g., Art Deco), or as examples of the works of specific architects such as, for example, colonial Tasmanian Georgian architect, John Lee Archer, or Gothic revival architect and designer, Augustus Pugin, who designed the interior of the English Houses of Parliament as well as numerous churches and religious paraphernalia in Australia and especially in Tasmania. Sites that have musical significance, such as the buildings where now renowned composers or performers once played, the houses they lived in, or their birthplaces, deathbeds and final resting places; or significant concert halls, recording studios, opera houses, organs or carillons around the world (see further discussion in Gibson and Connell, 2005). Sites that are depicted in the movies or on television, for example New Zealand has promoted the settings used in the Lord of the Rings films, Barwon Heads in coastal Victoria promotes itself as the location of a popular Australian TV series, SeaChange. Sites that have literary significance, for example locations mentioned by Wordsworth, Thomas Hardy’s Wessex, or Lorna Doone

STIMSON PAGINATION (M2469).indd 70

20/12/2010 15:12

A sociological perspective

●

●

●

●

●

●

●

71

in Exmoor, or Gundagai in NSW where the ‘dog sits on the tuckerbox’ (a statue nearby). Sites associated with particular artists, for example Constable country on the border of Essex and Suffolk in the UK, Albert Namatjira in Central Australia, or Hans Heyson in South Australia. Sites that have religious significance, such as places where there may have been apparitions, places along the pilgrim route(s) to Santiago de Compostela in Spain (El Camino), the journeys of Saint Paul in modern-day Turkey, or places visited by significant religious figures, for example, Sister Mary MacKillop, the only Australian to have been beatified. Another example is Walla Walla, a small town of around 600 people in south-central NSW, which boasts not only the largest Lutheran Church building in Australia, but a story of an 1890s trek of German Lutherans cross-country from the Barossa Valley in South Australia. Sites that have sporting significance, such as the locations of sport museums, famous playing fields, locations where some champion was born, raised, played as a youngster, had success (or failure), lived or died. For example in 2008, the centenary anniversary of the birth of cricketer Don Bradman, many regions in Australia advertised their connection to him. Sites that have political or ideological significance, or where there were political struggles or massacres (e.g., Robben Island in South Africa where Nelson Mandela was incarcerated); places where political leaders were born or grew up, or where significant political events took place. Sites with military significance, such as Gallipoli, the battlefields of Flanders, the Kokoda Trail in Papua New Guinea, or the sites of various incidents during the Vietnam war, or for example, the location of the memorial to the loss of the HMAS Sydney in Geraldton Western Australia, or the many battlements installed around the coast of Australia. Sites with industrial significance, such as the locations of mines or factories, especially ones that have become culturally iconic, such as the Cadbury chocolate factory in Hobart, or the Line of Lode in Broken Hill (NSW). Sites with historical significance, for example Burra in South Australia, which had the largest mine in Australia in the middle of the nineteenth century, and even though now a small town of only around 2000 people has become famous in the form of the Burra Charter (The Australia ICOMOS Charter for Places of Cultural Significance).

STIMSON PAGINATION (M2469).indd 71

20/12/2010 15:12

72

Endogenous regional development ●

●

●

Sites with criminal or penal significance, for example buildings that were built by convicts, that housed convicts, or places where bushrangers hid (e.g., Ben Hall’s cave in Grenfell NSW) or conducted their hold-ups (e.g., Glenrowan Victoria where the Ned Kelly Gang had their last stand). Sites of current indigenous or archaeological/anthropological significance, for example any place where there are Aboriginal rock paintings or engravings, sites where significant artefacts or middens have been found, sites where indigenous people may have lived, burial sites, massacre sites, spiritual or sacred areas. [Note that there will need to be negotiations with traditional owners before any attempt is made to promote these areas.] Even locations that were notorious can make a feature of their infamy.

STIMSON PAGINATION (M2469).indd 72

20/12/2010 15:12

5.

Rural, urban or regional endogenous development as the core concept in the planning profession Edward Blakely

INTRODUCTION The basic concept of endogenous development – or deriving outcomes from local resources – is based on the notion that the local resource is primary or essential to organizing or producing any results or tangible products such as a building location or an intangible, such as a public policy. Planning in all forms relies on elements of indigenous activities and/or endogenous resources because planning as a policy science is based on the use of rules to shape outcomes in space and place. All communities in the world – no matter how large or small – declare themselves to be unique in some dimensions. In some respects this is correct because no bounded area, such as a metropolitan area, is identical with any other. On the other hand, human habitats in systems do not vary markedly around the world from the most primitive to the most advanced. There are similar features in all landscape and social organizations. Nonetheless, the particular mix of these resources is endogenous – that is, it is local-specific. No matter what the prime resource is – from a coal mine to a park or housing development – some rules have to be devised to determine how the natural resource is extracted or modified or manipulated, requiring decisions on how the resource is to be exploited to enhance benefit for the local community. Thus, planning is the means to provide order to protect the environment, public safety and movement of goods. ‘The Plan’ is the base of the resource exploitation. Plans are descriptive instruments. That is, a plan in the sense of rural, urban or regional planning is contained in a specific locality and meant to engage the resource of that locality for advancement, enhancement or preservation. In a much larger sense, planning and plans are ways to articulate the endogenous character, assembly and use of human and physical resources. The one that they use frequently becomes 73

STIMSON PAGINATION (M2469).indd 73

20/12/2010 15:12

74

Endogenous regional development

the regulatory and monitoring instrument that exemplifies the resources’ depictions. Even building height limitations act to provide ways for the visual ‘uniqueness’ of a place to be preserved. For example, the sight-lines to the US national capitol, Washington DC, are preserved so that it is the central feature of and the manifestation of the US nation state. The evolution of the planning discipline comes from these former notions of ‘The Plan’ as the defining instrument for the preservation and articulation of the relationship between humankind and space. The notion of ‘plan’ has refined the ways in which a plan becomes a code or legislative instrument for local endogenous resource control. In many respects, planning as a science rests on the concept of how well local spatial and social/human resources are mobilized and utilized for endogenous development. The plan is thus the mechanism that matches human and physical resources for developing the better use of local resources in the development of a high quality of life in a specific place. Planning styles arise from various ways of understanding how local human and spatial endogenous resources can be combined within a city or other jurisdictional form. Planning as described here is as proclaimed by leading US planning theorists Charles Hoch, Linda Dalton and Frank So (Hoch et al. 2000): ‘A community’s physical aspects – its size, climate, location and geography – shape many of its planning issues. Other less immediately visible feature factors – such as a community’s history and social and demographic make up – also contribute to its uniqueness’ (p. 33). However, some scholars consider planning as being a-spatial; that is, planning is organizing any activity from running companies to organizing social organization. In this chapter we focus on the professional practice of planning that creates the link between organizing and controlling an indigenous land resource base or creating ways to develop, organize and use resources within a landdefined space that is in a rural, urban or metropolitan setting.

THE EVOLUTION OF PLANNING AS INDIGENOUS/ ENDOGENOUS ASSESSMENT The earliest formal literature of settlement describes how the local earth endowed and the resources generated a rationale for placing certain infrastructure, from roads and bridges to aqueducts, in an orderly fashion designed to maximize resource exploitation and protect or advance some form of settlement scheme. This set of activities evolved into what we now term professional planning. The earliest records indicate that the Romans, the Egyptians, the Greeks and other early civilizations had a very high regard for laying out settlements to advance specific purposes. The

STIMSON PAGINATION (M2469).indd 74

20/12/2010 15:12

The core concept in the planning profession

75

Pharaohs of early Egypt were particularly adept at settlement design and planning for the use of natural local resources. The Bible extols the virtues of layout and design of places like Babylon. In this respect the notion of indigenous planning was central to the development of the mighty Inca and Aztec empires as well as the guiding principles of China’s great cities. So, the idea of an indigenous base for plans is as old and as deep as the notions of professional planning. Indigenous asset development in this concept is the bedrock of the modern planning profession and manifests itself in several contemporary ways that are a continuum of the field’s ancestral roots. Here we will only describe the current form of local and parochial resources as they provide the template for the profession of planning. Planning needs to be regarded as being both a process and a form. As a process, planners in the last century have placed as a central tenet the notion of citizen engagement and assessment of local needs as the critical bedrock of ‘The Plan’. In essence, a plan can only hold legitimacy if it grows out of a collective engagement with locality (defined as the geographic area the plan embraces) and deals with a collective understanding of the resources that the plan purports to advance. As a result, the planning process considers all local indigenous resources – human, physical, architectural, historical and cultural – as necessary elements to form the base for the planning process. This means the reification of local assets is part of the indigenous resource base. Collective description of resources is as important as the actual scientific investigation and revelation of any local indigenous asset. That is, if the community does not see a local stream or creek as a value, then the planner’s science is challenged by community ascription versus real evidence. One of the classic examples is the often cited instance where a naive planner attempts to convince villagers in the Amazon to build a vegetable garden to improve local nutrition. The villagers object because they will not build any new projects until a monument is built honouring a local earth god. In essence, the planner has to consider belief as central to the plan as material goods. The villagers view plans as incidental to larger cosmic forces over which humankind has little control. So, planning as a practice as we know it recognizes the local endowments and is rooted in the notion of human control over natural environments. This has led to the sad consequences of climate change through too much planned carbon use with automobiles and factories. Clearly, communities can view cultural and related practices as indigenous assets even if they are contentious, such as religious historical roots or community practices that may be at variance with larger mores or laws.

STIMSON PAGINATION (M2469).indd 75

20/12/2010 15:12

76

Endogenous regional development

Several obvious examples of this are well documented by Tomlinson et al. (2003) and others as they describe the use of planning processes to justify racial segregation as valued as indigenous resources in South Africa and Northern Ireland, Israel and Palestine. For many years, so-called Jim Crow Laws were codified in planning codes in the US to spatially separate blacks and whites. In essence, planning is a policy tool that incorporates and institutionalizes human values spatially. Clearly, there is more illustration of the reification of local indigenous/endogenous cultural resources. New Orleans, post-Katrina in 2005, is using its past indigenous cultural base as the primary object in re-planning the city after the hurricane. In the case of New Orleans, the old indigenous spatial separations were challenged in a post-storm setting when people could articulate another set of values that were not based on race and class as the way space can and should be organized.

LAND AND ITS USES AS ENDOGENOUS PLANNING BASE Thus, the larger and more profound aspect of indigenous resources for planning is how land is used and controlled, ranging from parcel sizes and heights to access and transportation. In this case the use of land for various purposes is based on an assessment of how the in-place resources might be used for various purposes and linked together to form new opportunities. Planning a railroad to carry natural mineral resources is a good example. In this case, the natural resource that is a collective resource is exploited as a reasonable plan for current and future use. But as the exploitation takes place other natural resources may well be jeopardized. So, the fact that the plan is to use something indigenous is not in itself an attribute so the planning process of collective assessment has to include all of the options, opportunities, barriers and threats this resource requires. This is how planning is based on the endogenous aspects of the setting. As we enter a new millennium the local asset is no longer what the place is but who is in the place and how they understand the use of place assets in presenting the human knowledge base to the local region and the world. Computer technology transforms the notion of endogenous advantage since global interchange can potentially now take place from any location thus making many activities completely footloose in that regard. Until the advent of computer technology, places were valued by their connectivity via water, rail, road or air. Now, the place as natural setting or collection of human skills can be an endogenous value. For example, call centres handling airline reservations around the world can be located

STIMSON PAGINATION (M2469).indd 76

20/12/2010 15:12

The core concept in the planning profession

77

in a picturesque rural area that is attractive for lifestyle reasons for the call centre workforce or because the area is less expensive but the quality is very high. Many small rural communities in the US are now economically robust because they are attracting a cross-section of people who can work from these locations locally and globally. In Australia, the so-called ‘sea-changers’ who move to coastal locations are founding new business in small coastal villages and linking to the world (Gurran et al., 2007). So the endogenous resource to be planned in this case is how to provide locally access to the ITC infrastructure that is necessary for small- and mediumsized businesses (including one person operations) to have access to the web, so that a local endogenous advantage in this regard is equivalent to that elsewhere where such infrastructure access is available.

ENDOGENOUS ECONOMIC DEVELOPMENT PLANNING Within planning the activity of local economic development planning is based on endogenous processes. The basic theory associated with local economic development planning is that capitalizing on local indigenous resources is the springboard for creating a sustainable economy. Blakely and Bradshaw (2002) have laid a strong foundation for the notion that jobs and wealth are not produced by attraction or altering incentives for firms but must be induced by the quality of the location and the use of local resources to develop equitable and sustainable wealth. Blakely and Bradshaw reject the old theories of development that based the process of progress on the attraction and retention of firms to what they term the Third Wave Theory: 1. 2.

3.

4. 5.

It assumes that better and not more populations is the key to economic improvement. It bases economic growth not on attracting firms but attracting and retaining high-quality human resources with the institutions, learning, research and innovation to retain them. It places all communities in the global economy and not competing with one another within their regions for the same market share of a diminishing resource base. It proposes multi-economic bases versus single industrial or sector bases for all communities, both urban and rural. It views social equity as critical to community well-being and stability. It places technological and social innovation at the forefront of economic expansion. (Ibid. pp. 47–8)

STIMSON PAGINATION (M2469).indd 77

20/12/2010 15:12

78

Endogenous regional development

Sustainable wealth cannot be imported – as Blakely and Bradshaw illustrate above – but must rest on local endogenous resources that are not transferable. Even if the form of the indigenous resource is altered – that is, the factory is not competitive – the people and place can alter their competitive strategy by reconfiguring their resources to compete and not attempt to copy some other place or relocate a firm from another location. As the base for manufacturing changes from machine and natural resource, a community must maintain its leadership by increasing the skills of its population or adding other value to the goods or services. In this way the real assets, the people and place, are recreated by planning. In this model Blakely and Bradshaw (ibid.) assert that local economic development planning is not a one-time but a continuous process of adjusting resources internally and developing new means to compete with those resources in different ways. To illustrate this point, communities that redefine their current position can use a combination of place features and institutional resources. A community that was a fishing village and has a picturesque setting can leverage the setting after the fish are gone by repositioning itself with its local college as a new music and arts venue that attracts artists and musicians all year round using the old fish factories as studios for music and art as well as the teaching staff at the college as the artists in residence for the programme. This type of transformation might attract new year-round residents who retire to the place for continuing artistic education for themselves or to be part-time teachers and thus improve their own lives in the process of developing a new economic niche. The Product Cycle Trap Places that import economic activity become victims of the import rather than the resource to build continuing wealth. Thus, it is no accident that poverty and natural resource extraction are so closely linked because the extraction of the natural resources usually degrades the human resources as well. Of course, when the natural resource is gone, the human resources are discarded. Factory towns are the prime example. In the early years, the town has a competitive advantage with energy and or labour cost being low. The wage, transport or resource arbitrage as it is called – the difference between the costs and profit margin – is reduced over time as other places closer to the markets or with lower labour or energy cost compete for the same goods production. As a result, the factory town dies and its workers are stranded. The only way for places to remain competitive in a global economy is to build on local resources and retain the wealth or added-value components as close to the source of exploitation as possible. Similarly, manufacturing and related mass production activities – as

STIMSON PAGINATION (M2469).indd 78

20/12/2010 15:12

The core concept in the planning profession

79

standardized as Vernon (1966) in his classic work has shown – will move to the lowest-cost areas, leaving the last host denuded to the capacity to adjust and use the same skilled workforce or other resources to maintain economic perch. The product cycle theory goes like this according to Vernon (ibid., p. 52): Economic development is defined as the creation of new products and the diffusion of standardized products. Development originates in the more developed region and is exported to the less developed region through trade and investment. Establishing a new industry in the less developed region creates a progressive force that can help eliminate the barriers to interregional equality.

Beyond the Product Cycle Youngstown in Ohio is a classic example of the deindustrialization process. Once a thriving rubber manufacturing centre that could boast that it produced the majority of rubber tyres in the world, it was to lose all its rubber factories in the 1970s. Now Youngstown is a model of economic recovery as it incubates with the local university to support new-technology firms in rejuvenated old derelict downtown retail spaces. Similarly, Emeryville in California, once a troubled industrial suburb of San Francisco, is home to some of the world’s fastest-growing biotechnology firms because it created good links with the nearby University of California at Berkeley, luring firms to old factory spaces with local land use zoning for start-up firms in biotechnology. So in these cases new resources were generated from old endogenous resources that were thought to be of low or no value. Planning endogenous resources is thus essential in local economic development planning. Planning is the key in local economic development. Plans must be based on the flexible and continuous improvement of local resources so they meet the needs of changing times and markets. Local economic development planning then is the epitome of endogenous planning. Local resources put together in the correct manner can remain competitive only through planning processes that recognize how these resources can be positioned, organized and reorganized to meet altering conditions regionally, locally and globally. Moving to a new strategic approach does not neglect the older assessments and engagement of community in determining local physical assets but provides a new crucible in which these resources fit. This is a stage and not a finish of indigenous asset incorporation into the planning process. As human resources are increasingly important the social links among people as discussed previously are being examined in new ways to help understand how people can form the endogenous base for planning.

STIMSON PAGINATION (M2469).indd 79

20/12/2010 15:12

80

Endogenous regional development

ENDOGENOUS PLAN AS SOCIAL SPACE Social capital is a new vocabulary but an old idea. For many years, the concepts of social capital have been understood. Cities and towns have reputations with respect to how people feel included or excluded and can play a beneficial or hostile part in the social dimensions of the place. Physical endowments are sometimes described as drivers for social interactions and the formation of social space. For example, people who live in cold wet places are said to be less sociable than people who live in warm sunny places. In this sense the climate is an endogenous resource that planners acknowledge in the planning process that promotes or impedes social capital formation. Furthermore, people who live in some cities seem to be bound by traditions that impede new projects or programmes. As mentioned earlier, New Orleans has experienced many failures when ideas that succeeded elsewhere were attempted there. The New Orleans World Fair in 1984 is an illustration of this. The city newspaper and leadership of the old guard did not like the idea so they conspired to let it fail rather than embrace it. On the other hand, San Jose in California built a downtown arena with no sports team and eventually got an ice hockey team to play in the arena. The idea of a northern-temperature ice hockey team surviving in sunny San Jose was derided nationally but the team and arena led to a downtown renaissance because of the civic spirit that led the effort (Henton et al., 2006). So, human civic spirit is an endogenous resource that firms and global capital is beginning to recognize in the form of places like Dubai that offer almost no other resource than a strong ‘can do’ attitude from its civic leaders. As discussed earlier, natural assets are not as pivotal in planning. Furthermore, social and cultural attributes are increasingly important as human capital attractors. Social capital is an endogenous resource that comes from social contacts that generate social contracts among a set of people who usually bond by space. That is, bonds of social relationships can be a resource as much as any hard fixed assets. Social capital is integral to planning since it is the interactions that create new ideas as well as act as the glue for capital formation and wealth generation. Individual reliability is enhanced with collective obligation. Thus, social capital in planning means sharing opportunities and resources to maximize desired outcomes. So, simple social capital might be thought of as in friends and colleagues working to fix a group member’s home for habitation post a disaster, and it can in fact extend across multiple geographic scales from a local community to the nation to the world (Putnam et al., 2003). Some communities treasure this form of community collective action. There are

STIMSON PAGINATION (M2469).indd 80

20/12/2010 15:12

The core concept in the planning profession

81

several excellent examples of endogenous social capital as the base for planning both economic and land use activities. The Basque Madrigon region is a prime example of collective social capital infusing all aspects of community life and planning all activities related to it. And a recent and well-known illustration of planned social capital is the Grameen Bank pioneered in Bangladesh with many off-shoots around the world in various formulations of micro-credit (Yunus, 2007). These forms of social capital are not part of the lexicon of planning at every level. The concept of restoring social capital as one of the essential pillars of planning is increasingly popular in the planning literature. Communities that lack endogenous social capital cannot determine their own destiny because they lack the social glue to articulate a direction and the common sense of long-term vision and spirit to hold that vision through time. No plan can survive if the society does not view the plan as an articulation of common purpose and protect that purpose through an array of social capital institutions.

SUMMARY AND TAKE-HOME POINTS Regional is local in the sense of endogenous planning (Stimson et al., 2002); that is, the base resources set in a locality, no matter how defined, are essentially at the smallest organized units of economic and social space. So a region is a collectivity of interlocking resources, human and physical, that can articulate a common planned use of these resources. The planning paradigm: ● ● ● ●

incorporates the notion of endogenous to assets or locates the common regional resource base; creates a vision for these resources that can be communicated through an instrument called a plan; generates the social cohesion/capital to forge a destiny based on the local/regional resources; creates an economic system from these resources that is in a continuing state of planning so that the region/locality remains competitive by inducing a consistent flow of ideas and resources to it and through it.

In these ways endogenous planning is basically the bedrock of quality social, physical and economic development. Without a plan to link and forge common purpose new endogenous resource won’t be sustainable for long. Planners are not merely police officers of rules for the present

STIMSON PAGINATION (M2469).indd 81

20/12/2010 15:12

82

Endogenous regional development

but crafters of the future by organizing and expressing how to use local/ regional resources optimally in ways that sustain any region or community.

REFERENCES Blakely, E.J. and Bradshaw, T.K. (2002), Planning Local Economic Development, Sage Publications, Thousand Oaks, California. Gurran, N., Blakely, E.J. and Squires, C. (2007), Governance Responses to Rapid Growth in Environmentally Sensitive Areas of Coastal Australia, Coastal Management, 35(4), 445–65. Henton, D., Melville, J. and Walesh, K. (2006), Grassroots Leaders for a New Economy: How Civic Entrepreneurs Are Building Prosperous Communities, Jossey Bass Nonprofit and Public Management Series, San Francisco. Hoch, C., Dalton, L. and So, F. (eds) (2000), The Practice of Local Government Planning (3rd edn), Washington, DC: International City/County Management Association, Municipal Management Series. Putnam, R.D., Feldstein, L. and Cohen, D. (2003), Better Together: Restoring the American Community, Simon & Schuster, New York. Stimson, R.J., Stough, R.R. and Roberts, B.H. (2002), Regional Economic Development: Analysis and Planning, Springer, New York. Tomlinson, R., Beauregard, R.A., Bremner, L. and Mangcu, X. (2003), Emerging Johannesburg, Routledge, London. Vernon, R. (1966), The Myth of Urban Problems, Harvard University Press, Boston. Yunus, M. (2007), Creating a World Without Poverty, Public Affairs Books, Washington DC.

STIMSON PAGINATION (M2469).indd 82

20/12/2010 15:12

6.

Diversity and endogeny in regional development: applying appreciative intelligence Tojo Thatchenkery and Jessica Heineman-Pieper

INTRODUCTION Endogenous processes in regional economic development have received renewed attention by regional scientists and regional development policymakers, due in part to the increasingly apparent contradictions within increasingly globalizing economic development. Globalization is a largely homogenizing force, and so the diversity required for innovation must come from residual uniqueness of local cultures and contexts (Sachs, 1992; Shiva, 2000; Thatchenkery, 2006). Endogenous vibrancy is both threatened by and a requirement for the engines of globalization, which are thus self-limiting (Sachs, 1992). Both as oases of possibility within a globalizing world and as reservoirs of possibility for a post-globalizing future, endogenous vibrancy is indispensable. Most importantly, it is a fundamental value in its own right. Focusing on human, cultural and organizational dimensions, this chapter examines a case study in how an endogenous ecology spontaneously created cascading entrepreneurial activity (one of infinite possible forms of vibrancy that could be studied, and a popular focus in economic development studies). In particular, the chapter analyzes the endogenous ecology of Silicon Valley in the US, and highlights some of the key cultural, organizational and human factors contributing to the technological innovation, creativity and entrepreneurial impact of the region. Many of these factors can be situated within the framework of ‘appreciative intelligence’ to provide generative and transferable lessons. At the same time, these lessons apply not at the level of content (replicating structures and systems) but rather at the level of ‘being’ or fundamental stance. At this level, the message also challenges some deeply held theories and beliefs about what leads to a thriving economy and a thriving society. 83

STIMSON PAGINATION (M2469).indd 83

20/12/2010 15:12

84

Endogenous regional development

DIVERSITY AND ENDOGENY IN EXPLANATORY PARADIGMS Diversity and endogeny are important not just as the topical focus but also at the level of underlying theoretical orientations and explanatory paradigms. One of the recurring tensions in social science theory and research comes from different views about the nature of social reality (Burrell and Morgan, 1985; Gergen, 1994) between researchers in search of a unitary, ‘objective’ truth and those who treat social reality as constituted by multiple, often competing perspectives on reality, arising out of the subjective interpretations and sense-making processes of a variety of actors and coalitions. Some would even argue that the interpretation of reality that usually seeks and gains dominance is the one actively promoted by those in power and that the challenge of building an egalitarian social order would be one of creating emancipatory spaces in which suppressed or muted views are allowed room for expression. For example, Zinn’s (2003) classic, A People’s History of the United States, demonstrates the impossibility of providing an ‘objective’ account of US history and the importance instead of listening to perspectives that are usually elided in dominant narratives. Based on extensive archival data and interviews, Zinn argues for recognizing the salience of multiple histories, and he demonstrates empirically that African American, Native American and European Caucasian settlers’ perspectives on and experiences of US history differ dramatically from each other in ways that cannot merely be assimilated one to the other. To speak about US history as if it were a singular, objective representation of facts would be simplistic, distorting and highly political. These considerations are not limited to history but infuse the sciences as well. In biology, a field that is analogous to economics in its combination of historical and experimental methods with complex systems, Lewontin (1994a, 1994b) has shown how the conceptual resources of the entire field are shaped by ‘organizing metaphors’, which structure, frame and circumscribe the search for evidence while operating outside of and immune to evidence. Lewontin demonstrates how these organizing metaphors are also laden with values, ideology and politics. At stake is not just representational correspondence but, much more importantly, what kind of world we contribute to creating (HeinemanPieper, 2009). Ever since the publication of Kuhn’s (1962) The Structure of Scientific Revolutions, a growing literature has demonstrated as untenable the earlier views of science as a neutral, value-free window onto the world (Wimsatt, 1976; Galison, 1987; Gergen, 1994; Lewontin, 1994a, 1994b; Rosnow and Rosenthal, 1997). Biological, ecological, psychological, social and cultural realities are open, dynamic and complex systems

STIMSON PAGINATION (M2469).indd 84

20/12/2010 15:12

Diversity and endogeny in regional development

85

(Wimsatt, 2007; Heineman-Pieper, 2009). Moreover, human beings (including researchers) do not stand outside of these systems but rather are participants in them. As a result, how we interact with these systems makes a difference in what we come to know about them (Wimsatt, 1976; Rosnow and Rosenthal, 1997; Heineman-Pieper, 2009), especially when our own human limitations and imperfections are further taken into account (Wimsatt, 2007). In this context, it should not be surprising that researchers and the results they produce are impacted by their interests, values, culture and disciplinary socialization, among other factors (Haraway, 1990; Sachs, 1992; Rahnema and Bawtree, 1997; Gergen and Thatchenkery, 2004). For example, Alvares (1992) has shown how at its very core the concept of ‘efficiency’ has been defined so as to focus on factors that favor high-temperature industrial production over ambienttemperature natural production while effacing considerations favoring natural, ambient temperatures over industrial production, even though from a long-term systems perspective this choice is leading to far greater and potentially deadly inefficiencies (global warming, destruction of natural ecosystems, dispossession of highly sustainable traditional people and cultures by ‘development’ projects, etc.; Rahnema and Bawtree, 1997; Gowdy, 2000; Shiva, 2000; Farmer, 2004). Much of the work in economics and regional science proceeds without reflective attention to the underlying framework assumptions and values that structure the research, while simultaneously reproducing and often even enforcing adherence to the dominant frameworks. The dominant framework assumptions and values that are usually reproduced occur both at the level of the content and at the level of the ‘rules of the game’ – what counts as doing ‘good’ research in the field (Gergen and Thatchenkery, 2004; Wimsatt, 2007; Heineman-Pieper, 2009). For example, at the level of the ‘rules of the game’, readers embedded in positivist methodology have objected to our approach on the grounds that good scientific research should either allow for verification (Hempel, 1965) or refutation/ falsification (Popper, 2002). These positivistic requirements on research impose a particular set of values and assumptions that have cascading effects (Heineman-Pieper, 2009) and ignore half a century of scholarship across a variety of fields showing these strictures to be untenable, misleading and distorting (Heineman, 1981; Gergen, 2009). For example, Galison (1987) has shown how Hempelian verification and Popperian falsification are unviable even in experimental particle physics (so how much more in the social sciences!). Galison shows how, in the debate over the existence of neutral current, the ‘critical tests’ required by verificationist/ falsificationist models amounted to a subjective process of social consensus building because the vital distinction between experimental fact

STIMSON PAGINATION (M2469).indd 85

20/12/2010 15:12

86

Endogenous regional development

and experimental artifact can be decided neither in advance nor on the basis of logic (ibid.). In biology, and on grounds equally applicable in economics, Lewontin (1994a, 1994b) shows how, contrary to the claims and requirements of verificationism and falsificationism, biologists are never forced to abandon their theories but can always merely revise their range of applicability – and yet these same theories embed values of social and political import. Kohn (1993) provides a parallel example of this in psychology and economics regarding the inexhaustible faith in the value of incentives, despite decades of evidence showing that incentives undermine intrinsic motivation, reduce creativity, destroy relationships and have a variety of other unintended consequences. Kohn shows how for 50 years researchers continue to ignore or dismiss this and other evidence and, determined not to admit to the detrimental effects of incentives per se, continue to search for the ‘right’ incentive schemes. At the level of content, regional economics has many standard explanatory constructs, ranging from external economies of scale and agglomerations (Marshall, 1890; Hoover, 1948, 1971; Blumenthal, 1955; Krugman, 1991) to industrial clusters (Porter, 1998), to regional competition (Isard, 1949; Begg, 1999; Camagni, 2002), to knowledge spillovers (Henderson, 2007), to high-technology networks (Audretsch and Stephan, 1996), to market structure and firm size (Acs and Audretsch, 1987). This chapter adds an important perspective to the literature by offering a novel way of framing, understanding and supporting endogenous development based on a principle that is incommensurable with a core framework principle in economics and the existing literature on regional economic development, namely: valuing people, communities and activities as ends in themselves rather than as means to other ends (such as making money). Rather than attempting to reconcile or homogenize these divergent frameworks and explanatory strategies we will allow both to stand on their own terms, like the multiple histories of the US (Zinn, 2003). For the benefit of those who nonetheless insist on seeing their familiar categories, we will briefly mention the following points. Alchian’s (1950) account of adopted outcomes in regional economies relates only to sufficiency and not optimality of outcomes and thus has no bearing on the case studies that ensue. Other constructs operated in both Route 128 and Silicon Valley (the comparison set for the case study) and thus are not sufficient to explain the special vibrancy of Silicon Valley. Examples of these constructs include: external economies of scale, knowledge spillovers and academic/industry relationships. In the case of the more specific investigations of Acs and Audretsch (1987), the small-firm advantages they show still do not explain why cooperative-culture firms outperformed standard autocratic firms in the region (Baron and Hannan, 2002), and thus only

STIMSON PAGINATION (M2469).indd 86

20/12/2010 15:12

Diversity and endogeny in regional development

87

underline the contribution of the present chapter. The chapter does not claim to provide a comprehensive account of all of the factors involved in the vibrancy of Silicon Valley; however, it does claim to elucidate a very important factor that is systematically overlooked by the intersecting disciplines that constitute regional economics. The chapter further claims that this factor has been overlooked for reasons related to systematic biases in the values, framework assumptions and organizing metaphors through which these disciplines structure, relate to and attempt to explain reality.

CASE STUDY: ECOLOGY OF SILICON VALLEY VERSUS ROUTE 128 High-tech regions, clusters and industrial districts have attracted serious scholarly attention in the last two decades. Porter’s (1998) treatment of industrial clusters highlights the importance of the local to the global: ‘enduring competitive advantages in a global economy lie increasingly in local things – knowledge, relationships, motivation – that distant rivals cannot match’ (p. 78). Even within high-technology clusters in the US, different locales have had very different histories and trajectories. In this vein, many scholars have contrasted Silicon Valley, California, and the older technology corridor on the East Coast near MIT and Harvard in Massachusetts, Route 128. Numerous studies (Weiss and Delbucq, 1987; Saxenian, 1999, 2006; Delbecq and Weiss, 2000; Florida, 2002, 2003; Lee et al., 2000; Shipley, 2006; Hulsink et al., 2007) have documented organizational, cultural and regional differences between Silicon Valley and the long-standing East Coast high-technology corridor, Route 128. Both of these regions, Silicon Valley and Route 128, are characterized by a concentration of high-technology firms and access to top tier universities (e.g., Stanford, MIT and Harvard). However, Silicon Valley has bypassed the longerestablished Route 128 in number of viable firms created as well as in other Western management measures of economic and innovative success. As Saxenian (2006, pp. 27–8) writes: The region’s technology upstarts proved more adaptive than their older, vertically integrated counterparts in an environment of fast-changing markets and technological advances. In the 1960s and 1970s, Silicon Valley’s semiconductor companies outperformed older East Coast corporations like RCA and Sylvania, and in the 1980s the region’s personal computer industry outperformed large, established companies including NCR, Honeywell, and Digital Equipment Corporation.

STIMSON PAGINATION (M2469).indd 87

20/12/2010 15:12

88

Endogenous regional development

Using data from interviews, questionnaires, lists of associations and custom databases, researchers (Weiss and Delbecq, 1987; Saxenian, 1999, 2006; Delbecq and Weiss, 2000; Florida, 2002) have identified several key factors as contributing to the high levels of IT entrepreneurialism in Silicon Valley as compared with Route 128. Whereas Route 128 was dominated by traditional, highly vertically integrated firms with hierarchical management and control cultures seeking to hoard as much as possible within the walls of their own firms (a zero sum game attitude), in Silicon Valley there was a fluid flow of people and ideas among firms and a relatively flat organization. Some of the early firms, such as Varian Associates, were explicitly founded as an ‘association of equals’, and the founders stated that ‘We did not want to have the hierarchy of a company owning facilities and employing employees’ . . . [and ] . . . ‘We wanted to be a cooperative. We wanted to create a cooperative organization’ (Lecuyer, 2007, p. 98). For example, employees switched firms regularly and without penalty in Silicon Valley, whereas such movement was frowned on and often contractually limited in Route 128 firms. Whereas Route 128 took a judgmental attitude towards failures, in Silicon Valley, failures were viewed as just part of a process of getting things right. Saxenian quotes WebEx cofounder Min Zhu: ‘The advantage of Silicon Valley is that you can fail and learn and try again. It’s the only place where you can screw up once and try again’ (Saxenian, 2006, p. 31). Researchers have also documented key differences in the cultures underlying Route 128 versus Silicon Valley. Whereas Route 128 is traditional, ‘stodgy’, hierarchical and conformist, in Silicon Valley technological fluency made relative equals of people of diverse ages, nationalities, backgrounds, levels of social skill and various other ancillary characteristics (Delbecq and Weiss, 2000). However, this egalitarianism is not absolute but relative (to Route 128); limitations and exceptions existed within Silicon Valley, such as the persistence of discriminatory limitations in advancement for some groups (Saxenian, 2000). In Silicon Valley, and especially in the early days, many people engaged in innovation primarily as a form of self-expression and creativity, with financial gain as merely a secondary benefit. This important and prevalent quality can be seen at cultural, organizational and individual levels. For example, at an organizational level, useful data can be found in Baron and Hannan’s (2002) longitudinal study of nearly 200 Silicon Valley start-ups. In addition to collecting a variety of other data, these authors conducted interviews with founders and CEOs in order to excavate the assumptions and mental models that ‘guided their thinking about how to organize employment relations and manage personnel’ (ibid., p. 10). The authors found the ‘notions about how work and employment should be organized

STIMSON PAGINATION (M2469).indd 88

20/12/2010 15:12

Diversity and endogeny in regional development

89

varied along three main dimensions . . . each characterized by three or four fairly distinct options or approaches’ (ibid.). The three main dimensions are: ● ● ●

the basis of employee ‘attachment’ to the company; the criteria for coordinating and regulating employees; criteria for selecting employees.

In the first of these, employee attachment, the authors noted three bases envisioned by founders, which they called ‘love, work and money’ (ibid.; italics original). ‘Work’ corresponds to founders’ recognition that ‘the primary motivator for their employees is the desire to work at the technological frontier’ (ibid.). ‘Love’ corresponds to ‘creating a strong family-like feeling and an intense emotional bond with the workforce that would inspire superior effort and increase retention’ (ibid.). This is a form of social motivation that could be seen as providing a supportive relational context in which the employees’ could express themselves freely. It is a culture that values people over money and creates an environment of trust and broader human valuing that can encourage more courageous self-expression. The final form of attachment is ‘money’ – an exchange relationship of labor for money. After coding the data into all possible combinations of subtypes across the three main factors, the authors found that firms clustered into five main types or organizational blueprints. Only one of these involved money as a form of attachment, and this blueprint formed only 6 percent of the distribution of organizational types (ibid., p. 13). Moreover, even when the authors controlled for a variety of variables, the money-based blueprint was the most likely to fail of all the blueprints, including a catch-all ‘nontype’ category blueprint. In addition, although the underlying statistics are not sufficiently presented to clarify the precise meaning of this result, the magnitude of the percentage difference in failure likelihood that the authors report was greater between this worst-performing money-based blueprint and the next worst-performing blueprint than existed between the second-worst performing blueprint and the best-performing blueprint. While it is true that the focus on money to attract and retain employees was only one of three elements within this poorly performing blueprint, the other two qualities in the other two dimensions – selecting employees based on specific skills and a hierarchical form of control – are not independent of the money-based mindset (as suggested by the fact that not all of the possible cells represented by the intersection of the three dimensions were occupied) and are part of the same world-view and gestalt. The commitment to meaningful activity for its own sake, as a form of

STIMSON PAGINATION (M2469).indd 89

20/12/2010 15:12

90

Endogenous regional development

self-expression, rather than to make money can also be seen at a cultural level. Florida (2002) notes some of Silicon Valley’s roots in 1960s’ counterculture. Unlike organized labor, which fought for ‘powers and rights within the framework of the existing economic system’ (p. 203; italics original), the counter-culture in northern California ‘included a wide spectrum of views on work and economics’, including operating outside of it: Some in the hippie milieu preferred simply to ignore the world of work, perhaps living by their wits or the generosity of friends or parents. Some sought to rob ‘the system’, as described how-to-do-it style in Abbie Hoffman’s Steal this Book. For many the strategy was to grudgingly coexist with the system. Get a job, even a haircut if you must; earn the money you need to do what you have to do, but no more. (Ibid., p. 203)

Freed from the bonds of greedy materialism, the innovators of Silicon Valley were free to focus on following what they most enjoyed, innovating for the love of it, and making things work. Economic and capitalist ideology pretends that people can be driven primarily by a profit motive and then claims that this assertion represents either the base reality of human nature, a useful lubricant for the market mechanism, or else a neutral fact. The success of Silicon Valley offers a counterweight to that view and shows that it matters why we do what we do and that there’s no substitute for doing things for their own sakes and treating people as ends in themselves rather than as means (e.g., rather than as in the labor for money model in Baron and Hannan, 2002). At an individual level, the importance and prevalence in Silicon Valley of people motivated not by money but by love of what they do – work as play or hobby – can be seen both in famous founding stories (e.g., Apple) as well as in stories from the more ‘everyday’. For example, Google Desktop came to be because an employee had decided to create that function for himself on his own computer in his spare time. Simultaneously, at an organizational level, Google had structured the work week to give employees a free day every week to pursue their own personal interests – and did not demand that anything ‘productive’ or ‘useful’ come of it – in recognition of the fact that employees were intrinsically motivated to create and would do this best when they were given the freedom to follow what they enjoyed, unconstrained by top-down agendas (Elgin, 2005). Delbecq and Weiss (2000) write that people who would be considered ‘misfits’ in regions (such as Route 128) that are concerned with ‘superficial impressions’ such as those based on elegance of expression (versus unfluent English as a second language), elegance of social skills, or elegance of physical appearance (note that many psychology studies have shown the biasing effects in business and other environments of what mainstream

STIMSON PAGINATION (M2469).indd 90

20/12/2010 15:12

Diversity and endogeny in regional development

91

society identifies as canonical ‘good looks’) are given prime projects to lead and prime resources to command in Silicon Valley. According to Delbecq and Weiss (ibid.), Silicon Valley is much less concerned with superficial impressions than are other parts of the US (such as Route 128): In Silicon Valley, it is the quality of ideas, the willingness to creatively problem solve, the ability to arrive at a breakthrough technical solution, which is valued. I have heard senior managers listen with rapt attention in briefings to young talent whose modality of presentation is rough. . . . As one venture capitalist expressed it, ‘I don’t care how the talent is wrapped’ (ibid., p. 40).

Also, these authors report that in Silicon Valley there is an ethos that gains are shared not based on hierarchical position, as in Route 128 firms, but instead based on contribution. Fleming and Marx (2006) present the concept of ‘small worlds’, social networks that cut across individual firm boundaries, in explaining how the effective management of innovation in an environment of ongoing knowledge exchanges was a critical factor underlying the success of Silicon Valley. Small worlds, families of interconnected firms, are more likely to be the source of innovation than individual firms (see also O’Brien, 2007). Saxenian (1999 and 2006) shows how this small worlds phenomenon was also made especially effective by the many highly educated immigrants who came to Silicon Valley. Saxenian (1999, 2000, 2006) shows how these families of interconnected firms went global as a result of the social networks of immigrants/expatriates largely from Asian countries (also reported in Adams, 2005). Saxenian documents the social networks of Indian, Chinese, Taiwanese and other immigrant groups and how these associations, interactions and knowledge-sharing not only substantially advanced the development within Silicon Valley, but also created global technological and business networks beyond what would otherwise have been possible. In both of these contexts, domestic and international, the immigrant communities significantly augmented the overall ‘size of the pie’. While Route 128 firms were busy protecting their turf, Silicon Valley firms and individuals, inspired by innovation itself and coming from a mindset of abundance rather than scarcity, shared freely and thereby created a better result at each level from the individual through to the whole. Across diverse metrics, Saxenian (2000) shows that immigrants, often from Asian countries such as China and India, were a vital part of the Silicon Valley technological and entrepreneurial success. For example, approximately one-quarter of ‘Silicon Valley technology firms founded between 1980 and 1998 had Chinese or Indian executives’ (p. 252), and this number is likely to be an underestimate because racism powered by and

STIMSON PAGINATION (M2469).indd 91

20/12/2010 15:12

92

Endogenous regional development

instituted through restrictions on venture capital required many non-white immigrant founders to choose white American CEOs (Saxenian, 2000). The (likely underestimated) immigrant-founded firms created significant wealth and employment in Silicon Valley, and in 1998 ‘collectively accounted for more than $168 billion in sales and 58 282 jobs’ . . . [and the] .  .  . ‘rate of immigrant entrepreneurship in Silicon Valley has increased significantly over time’ (ibid., p. 253). The experience of immigrants in Silicon Valley points to a serious shortcoming of the culture and place. While it may have been more open than Route 128, Silicon Valley was still a place of advancement limitations and discrimination against the very immigrants who were in large part fueling its success (ibid., p. 251). Saxenian cites a 1991 survey of Asian professionals in Silicon Valley that showed that ‘two-thirds of those working in the private sector believed that advancement to managerial positions was limited by race’, and this is further supported by the fact that ‘Chinese and Indian engineers remain concentrated in professional rather than managerial positions, despite superior educational attainment [to the white Americans occupying equal or higher positions]’ (ibid.). Interestingly, ‘those surveyed attributed these limitations less to “racial prejudice and stereotypes” than to the perception of an “old boys” network that excludes Asians’ . . . [and the] . . . ‘“lack of role models”’ (ibid., p. 25). At the same time, overt discrimination also existed, as is demonstrated by the requirement of many venture capitalists that non-white immigrant founders choose a white American CEO instead of running the company themselves. For example, Saxenian reports the story of David Lee, who ‘left Xerox in 1973 to start Qume after a less experienced outsider was hired as his boss. Lee was able to raise start-up capital from the mainstream venture capital community, but only on the condition that he hire a non-Asian president for his company’ (ibid., p. 251). Despite (or perhaps partly because of) these barriers, immigrants in Silicon Valley created local ecologies within the larger Silicon Valley ecology to support themselves and their compatriots. Immigrant groups and subgroups founded associations that provided both social networks  and professional contacts, as well as connecting people with role models and sources of tacit knowledge and experience. For example, David Lee, the founder of Qume described earlier, and a handful of others like him, became ‘community leaders and role models for subsequent generations of Chinese entrepreneurs’ (ibid., p. 252). Putnam (2000) has shown the importance of these forms of civic engagement in building social capital and community (see also Beugelsdijk, 2003). While the immigrant experience shows how the social networks were racially divided, indicating that Silicon Valley was not quite as free

STIMSON PAGINATION (M2469).indd 92

20/12/2010 15:12

Diversity and endogeny in regional development

93

of racism as some accounts suggest, all of these accounts point to the importance of social networks as a factor in Silicon Valley entrepreneurship. Audia and Rider (2005) show how the ‘garage legend’ creates a false impression of entrepreneurs as individual founders, whereas these authors document the importance of organizational ties even in the case of canonical founder stories such as Apple and Hewlett-Packard. Ferrary (2003) explicates one of the many modalities in which social networks are important. He outlines three mechanisms of exchange that underlie the circulation of economic goods: ● ● ●

the ‘arm’s length transaction’; the ‘power relationship’; the ‘gift exchange’.

Those correspond to the modalities of the market, the organization and the network, respectively. The social embeddedness of the actors in Silicon Valley (arising from having worked in the same firm, graduating from the same university or sharing a common ethnicity) resulted in a majority of the exchanges being informal exchanges based on reciprocity and a strong sense of community and modeled on the metaphor of the gift exchange. By this argument, Ferrary (ibid.) would suggest that the social architecture that normally accompanies urbanization forces the actors unwittingly toward more legalistic and formal forms of exchange and may explain the difficulties in mechanistically replicating Silicon Valley’s success if the underlying culture of the system is not oriented toward a sense of community (‘gemeinschaftfuhl’), a feature that is much harder to copy through rational intention. Another dimension of distinctiveness for the region is the way in which the different strains of the underlying culture and counter-culture themselves in turn created a variety of organizational cultures. As we have seen, Baron and Hannan (2002) contrasted five different blueprints of human resource practices among Silicon Valley companies. These are: the Star (marked by the attitude that ‘We recruit only top talent, pay them top wages, and give them the resources and autonomy they need to do their job’), the Commitment (represented by ‘I wanted to build the kind of company where people would only leave when they retire’), the Bureaucracy (exemplified by ‘We make sure things are documented, have job descriptions for people, project descriptions, and pretty rigorous project management technique’), the Engineering (captured by ‘We were very committed. It was a skunk-works mentality and the binding energy was very high’) and the Autocracy (implied by ‘You work, you get paid’.). They found that firms whose founders practiced the Commitment

STIMSON PAGINATION (M2469).indd 93

20/12/2010 15:12

94

Endogenous regional development

blueprint have performed better over the ensuing years compared with ventures based on the other blueprints. The Commitment blueprint highlights a reliance on emotional connections and family-type relationships; Baron and Hannan (ibid.) argue that perhaps given the rarity of such signals in the corporate world, the few companies that send out these signals end up enjoying a competitive advantage that is hard to imitate, particularly because it runs counter to conventional corporate ideology in ways that are humanistic and thus that people readily find valuable. It should perhaps not be surprising that the most successful variant, the Commitment blueprint, is one in which people are treated and valued as people rather than instrumentally. In short, researchers have identified a number of unique cultural features of the Silicon Valley region that have contributed to its ability to bypass the more traditional high-tech corridor, Route 128. These features include but are not limited to: ● ● ● ●

● ●

vibrant social networks; an underlying counter-culture; intense diversity of national background; relative (to Route 128 – though still less than it might be) valuing of persons as persons rather than as objects in the production process and sources of income and other ‘measurable results’ for distant owners; willingness to fail without passing judgment, to learn, and to try again; relatively flat organizational structures versus hierarchies.

These features will next be considered from the framework of ‘appreciative intelligence’ (Thatchenkery and Metzker, 2006).

APPRECIATIVE INTELLIGENCE Howard Gardner (1993) famously demonstrated that, contrary to notions generally believed at the time in the US, intelligence is not unitary. In the US, there has been a long (and unfortunate) tradition of viewing intelligence as springing from an essential, innate and fixed source in individuals, and measurable through standardized IQ tests. The resulting measure, the ‘G-factor’, was presumed to reflect the quantity of intelligence in any individual. The IQ assumptions and methodology have all been roundly critiqued by a host of scholars (see, e.g., Block and Dworkin, 1976), who have shown that:

STIMSON PAGINATION (M2469).indd 94

20/12/2010 15:12

Diversity and endogeny in regional development ● ●

●

95

the assumption of an innate fixed ‘quantity’ of intelligence is entirely unfounded; intelligence is highly situated for all of us, such that each of us may show high levels of a particular kind of intelligence in certain kinds of situations or contexts and at certain times and not in others; intelligence itself has myriad forms (as demonstrated by Gardner, 1993).

Thus, in this part of the chapter we consider a quality that has been ignored in the literature, namely ‘appreciative intelligence’, or the ability to focus not on problems and shortcomings but on the beauty and potential inherent in any moment, including moments of suffering (e.g., to ‘see the mighty oak in the acorn’). This is the type of intelligence that our best school teachers have shown when they saw and brought out more in us than we may even have known at the time that we had. In contrast to intelligence focused on ‘problem-solving’ – which inevitably thereby everywhere sees problems and solutions – appreciative intelligence is a way of being in the world that always sees the constructive side of persons and situations without being utopian or unrealistic. It is about seeing and taking a stand for our own and each other’s best selves. When we take this stand completely and holistically, rather than partially and with limits, it gains enormously in power and authenticity. In other words, if we see an opportunity to make money by exploiting others, the narrowness of our ‘appreciative’ focus, and especially the lack of appreciative intelligence towards other persons, constitutes a blind spot that will in some way feed back around to impact our own well-being. One of the most powerful aspects of the appreciative intelligence shown in Silicon Valley is the very fact that it was focused on creating possibilities that were positive intrinsically and in multiple dimensions – in how people organized work together, in the authenticity of their intrinsic enjoyment in creating helpful and useful technologies and in their love of what they did – rather than being focused on competing with others and winning. We have already seen several examples of this, such as the case of Google seeking ideas from anywhere in the organization (rather than just the top) and providing its engineers with a day a week to work on whatever personal pet projects they wish. This practice demonstrates both how the employees were motivated by the love of discovery and problem-solving, such that the day a week to work on their pet projects was a perk of working at Google, and the organizational wisdom of Google that people usually do their best work when they are doing what they most enjoy. In the end, the organization will be enhanced from allowing people to be maximally self-expressive, whether or not that self-expression ever yields financial or

STIMSON PAGINATION (M2469).indd 95

20/12/2010 15:12

96

Endogenous regional development

other ‘measurable results’. In allowing employees total freedom to choose to engage in what they enjoy on those days and to pursue it for their own personal interest and benefit, Google recognized and refused to be caught in the irony that as soon as people’s work is held to the standard of ‘measurable results’, the freedom to pursue and experiment with what they truly and intrinsically enjoy has been destroyed – along with the profound breakthroughs (such as Google Desktop in the example previously described) that it may – or may not – spawn. Components and Qualities of Appreciative Intelligence Appreciative intelligence has three components, and leads to four qualities (Thatchenkery and Metzker, 2006). The three components or elements are: ● ● ●

reframing; appreciating the positive; seeing how the future unfolds from the present.

The four qualities that result from appreciative intelligence are: ● ● ● ●

persistence; conviction that one’s actions matter; tolerance for uncertainty; irrepressible resilience.

The three components are discussed in turn. Reframing This is the ‘psychological process whereby a person intentionally views or puts into a certain perspective any object, person, context or scenario. One of the most common examples of framing is that of calling a glass “half empty or half full”’ (Thatchenkery and Metzker, 2006, p. 6). We have tremendous choice in how we see situations, ourselves and others: what we pay attention to; what we take to be important; whether we see things as a liability or an opportunity for learning and growth, and so on. Often life appears to us as a ‘given’ reality, and older traditions in Western social science reinforced that way of seeing (contemporary social sciences and philosophy of science have seriously revised this view, however). In fact, we can always choose how we make meanings out of situations and how we respond, and people exercising their appreciative intelligence do just that. Viktor Frankl (1963, p. 104), a concentration

STIMSON PAGINATION (M2469).indd 96

20/12/2010 15:12

Diversity and endogeny in regional development

97

camp survivor, famously wrote: ‘everything can be taken from a man but one thing: the last of the human freedoms – to choose one’s attitude in any given set of circumstances, to choose one’s own way’. Appreciating This second element of appreciative intelligence builds on the first and is the most constructive aspects of a situation in a realistic fashion. We can see the power of this element most forcefully when we think about being on the receiving end of it. If you go well out of your way to do something nice for someone, and they focus only on flaws in the content of how you selected to express this goal, you can see that they are missing the most powerful and generative aspect of the situation – your good intentions. If, instead of focusing on the problems in the execution, they focused on the beauty of your intentions and from there, in a space of alliance, requested a different mode of expressing the intention, the experience and result of the interaction, as well as of downstream results and interactions, would be very different indeed. In Silicon Valley, this positive valuing was largely present along some dimensions and in some ways, and not in others. For example, relative to Route 128, Silicon Valley often embraced people regardless of the superficial polish and sheen important in most traditional business environments – the pattern quoted earlier of not caring ‘how the talent is wrapped’. This valuing sometimes also may have extended to seeking to value employees as persons rather than as means to ends, as is suggested somewhat by the prevalent ‘commitment blueprint’ of organizations described by Baron and Hannan (2002). At the same time, this valuing was incomplete in that immigrants faced discriminatory limitations in advancement in many firms and from many venture capitalists. Moreover, the valuing was somewhat limited in that, while technical insight may have functioned to create greater equality across people who might have been discriminated against based on other superficial dimensions, this valuing did not extend universally to persons as persons such that people of all sorts and in all professions (janitors, construction workers) were accorded the same levels of social appreciation, support and opening of possibilities, along the lines Paolo Freire sought to establish in his emancipatory educational institutions (Freire, 1996). While much more common than in other parts of the country, this valuing wasn’t universal, as the example of PayPal shows, where employment was based on being ‘like’ the founders along many superficial dimensions including aversion to sports (O’Brien, 2007). At the same time, these exceptions existed in a context in which organizations themselves had diverse cultures and thus, across organizations, the result would have been a valuing of a distribution of characteristics. While less

STIMSON PAGINATION (M2469).indd 97

20/12/2010 15:12

98

Endogenous regional development

than holistic, this organization-level distribution still provides greater opportunities for valuing more diverse people (or aspects of people) than other parts of the country were doing at the time. In other words, while far from perfect, Silicon Valley offered greater opportunity and valuing than the other members of its reference class. Seeing how the future unfolds from the present This is the third element of appreciative intelligence. Since in fact, as the second element has implied, ‘useful, desirable or positive aspects already exist in the current condition of people, situations or things’ (Thatchenkery and Metzker, 2006, p. 33), the final element involves attunement to how to nurture and amplify these constructive elements. In short, when people are showing high levels of appreciative intelligence, they: see the oak in the acorn. They also go beyond – they plan[t] their acorns and persevere to help them grow. While others may doubt the potential of the acorns, these leaders believe in their own and others’ abilities to water and fertilize the plants from sapling to tall oak. They deal with the risk and uncertainty that comes with planting something new and hoping for growth. Finally, they find a way for the oaks to survive and thrive despite unpredictable circumstances or a challenging environment. (Ibid., p. 33)

Relevance to Silicon Valley How does all of this relate to the Silicon Valley experience? Examples abound along many of the dimensions (reframing, appreciating the positive, seeing the future unfold in the present) and qualities (resilience, persistence, conviction that one’s actions matter, and comfort with ambiguity) of appreciative intelligence. For example, we have just encountered elements of a general Silicon Valley culture of allowing everyone to be who they are, regardless of how the more traditional firms might view them as ‘misfits’, socially inept, counter-establishment, or otherwise failing to have the outer demeanor and sheen that traditional business culture unfortunately trains people to value. In the words of the venture capitalist quoted earlier from Delbecq and Weiss (2000), ‘I don’t care how the talent is wrapped’. There are, as we have seen, limitations and exceptions to this, such as the fact that at the organizational level, PayPal was fairly intolerant of various dimensions of difference internally, though those dimensions were themselves offbeat – as an early employee described it, ‘We’re all a little weird’ (O’Brien, 2007, p. 106). Likewise, consider the application of appreciative intelligence to technological problems by the young techies. This was a culture of making things work with whatever materials might be available in the garage

STIMSON PAGINATION (M2469).indd 98

20/12/2010 15:12

Diversity and endogeny in regional development

99

and through social networks (Saxenian, 2000; Audia and Rider, 2005). It also shows a willingness to fail and to persevere in the face of failures. YouTube and Yelp founders ‘learned a valuable lesson from PayPal: The first idea isn’t always the best. Yelp was a convoluted e-mail referral service before becoming a top review site. YouTube started as a video dating play’ (O’Brien, 2007, p. 106). Google’s Marissa Ann Mayer says that Google’s innovation relies on its ‘fearlessness’ – she launches products ‘early and often’ without fear of failure (Elgin, 2005). Referring to Apple Computer and Madonna, Mayer says ‘Nobody remembers the Sex Book or the Newton. Consumers remember your average over time. That philosophy frees you from fear’ (ibid., p. 90). A founder of PayPal had so much confidence in his own actions that he founded a financial transactions company even though when he started, ‘Peter Thiel didn’t know what a chargeback was.. . . That’s one of the fundamental things of any credit card payment system. Chargebacks almost killed the company’ (O’Brien, 2007, p. 106). At PayPal, ‘the executive team made up for nonmastery of details with unwavering vision, which inspired the troops’ (ibid., p. 106). Moreover, a can-do approach of getting in and trying things and learning on the fly inspired other employees to found their own businesses. For example, Chad Hurley, CEO of YouTube remembers his PayPal days as an education in business. When he arrived in California with a degree in art from Indiana University of Pennsylvania, building a successful company seemed like something other people did. ‘“You never think it could happen to you,” Hurley says: “But seeing Peter and Max and the guys come up with ideas and seeing how to make things work gave me a lot of insight. You may not have a business degree, but you see how to put the process into effect”’ (ibid., p. 106). We also should not underestimate the fact that appreciative intelligence was frequently applied not individualistically to advance one’s own personal gain at the expense of others, but rather collectively, holistically, and in a more communitarian way to appreciating, supporting and valuing the success of others. Rather than a scarcity-based zero-sum-game mindset, Silicon Valley was frequently characterized by a mindset of sharing, abundance and mutual encouragement and support. Founders of one firm would support former employees in founding their own firms – with capital, knowledge-sharing and other resources. This happened within myriad social networks, from the immigrant associations documented by Saxenian (2000), to the organizational support in which employees would be supported by their current and former employers to start companies (Audia and Rider, 2005; O’Brien, 2007).

STIMSON PAGINATION (M2469).indd 99

20/12/2010 15:12

100

Endogenous regional development

PayPal founders claim to have supported employees to such an extent that the ‘PayPal mafia’ now runs $30 billion in businesses (O’Brien, 2007). This even occurred sometimes in cases of separations under contentious and acrimonious circumstances, as in the case of Musk at PayPal, who continued to provide financial support to projects by people from PayPal even after he’d been removed by the board while on a flight to Australia for his first vacation in years (ibid.).

TRANSFERABLE LESSONS The history of management thought is full of examples in which a meteoric ‘success’ is identified and everyone rushes to imitate – often even as the original inspiration for the success story is itself being indicted for accounting fraud (e.g., Enron) or going extinct (e.g., Wang Labs). What makes us think that Silicon Valley will not burn itself out and end up worse off than if it had developed more quietly, moderately and sustainably? When we think about transferable lessons we must first make sure we really want what we are seeking to transfer. A second important irony is that Silicon Valley was not the result of deliberate planning. How then can we expect to create by deliberate planning something that itself emerged organically through a confluence of complex and historically contingent factors? What, if any, lessons can be learned from Silicon Valley? Metcalfe (1998, p. 123), 3Com’s founder, pointed out that Silicon Valley is the only place on Earth not trying to figure out how to become Silicon Valley. There we argue that the most important, generative and transferable lessons from Silicon Valley are not at the level of replicating the content or structure of so-called success factors. Instead, the most powerful lessons are to be found at the level of a fundamental stance. In particular, what worked in Silicon Valley was authentically and thoroughly valuing, supporting and inhabiting its own positive uniqueness – not trying to be Route 128, not trying to be Detroit and not trying to connect with its best qualities in order to make money. The example of Silicon Valley forces us to recognize the deepest level of appreciative intelligence: seeing and celebrating the most life-affirming aspects of each community, rather than trying to import a set of ‘best practices’ from one place to another. When seen to operate at the level of a fundamental stance, the lessons from Silicon Valley also challenge some deeply held theories and beliefs about what leads to a thriving economy and a thriving society. As we have seen, the culture in Silicon Valley combined intensive and vibrant transnational diversity with a counter-culture that valued technological work and innovation intrinsically rather than instrumentally in terms of

STIMSON PAGINATION (M2469).indd 100

20/12/2010 15:12

Diversity and endogeny in regional development

101

whether or how they would translate into economic gains (Saxenian, 1999, 2000, 2006; Florida, 2002). Of course, the opportunity to make money was not squandered, and it may even have been primary for some firms, but this was relatively rare in the organizational and underlying cultures, and when it did occur in organizations they failed in disproportionately high numbers (Baron and Hannan, 2002). In the core incubating culture spawning the phenomenon of Silicon Valley, money was a secondary rather than a primary driver for large segments of the population. Baron and Hannan (2002) showed that the Commitment blueprint – treating employees as persons rather than instrumentally as means – was the most successful within Silicon Valley. This model had the least likelihood of failure, while the autocratic model, with a central exchange relationship of working for money, had the greatest likelihood of failure. This ‘blueprint’, along with some of the others focused on work for the joy of it, can be seen as an extension to the organizational level of a counter-culture that refuses to value money as the end and persons as the means, but instead values relationships and authentic activities as valuable in their own terms – rather than instrumentally for their monetary potential. Again, this highlights the fallacy of trying to replicate Silicon Valley in order to create economic wealth: paradoxically, part of what made Silicon Valley economically successful may have been the very fact that many of the original innovators were not motivated because of money but instead as a result of intrinsic enjoyment of the technology and the company created around it. The Silicon Valley experience also challenges other framework assumptions of Western management. For example, traditional management prizes asymptotic levels of efficiency, and yet one of the most important sources of success in Silicon Valley was the willingness to fail and the acceptance of failure as part of the process. From a traditional management perspective, failure is ‘waste’, and something that should progressively be eliminated. Not only has the cult of efficiency in Western management been eroding personal lives and the ability of employees to bring their whole selves to bear in the workplace, but also it is, like global capital’s destruction of the regional and local, self-defeating. Traditional management theories also collude with the hierarchical structure of salaries based on power and control within the company. Management ideologies usually rationalize these modes of apportioning wealth with patently false (and irrelevant: Kohn, 1993) claims that pay is related to actual contribution to the organization. These claims have now been dramatically exposed by the US financial meltdown, where CEOs who ruined their companies continued to reap handsome bonus pay immediately before, during and even after their firms received millions of dollars in bailout funds from taxpayers or were bought as a distressed sale (as in

STIMSON PAGINATION (M2469).indd 101

20/12/2010 15:12

102

Endogenous regional development

Bank of America buying Merrill Lynch as the latter verged on collapse). Many of the early Silicon Valley firms at least partially and relatively contravened these hierarchical and extorting structures and mindsets by adopting a relatively ‘flat’ organizational structure and apportionment of gains. Again, however, this structure was rooted in the ecology of the region. When other regions adopt a stance of appreciative intelligence towards their own organizational structures, they will not merely import either the hierarchical or the flat models, but will consider (authentically, and not just based on the narrowly conceived self-interest of the powerful) the kinds of organizational structure that can best support their highest potential. At the same time, in most cases, when the interests of the powerful are not in the driver’s seat, the most effective structure is indeed most likely to be a relatively flat one in which everyone has the autonomy, power and, thereby, ability to be their best and contribute fully to the organization. In identifying the transferable lessons from Silicon Valley, this chapter raises a fundamental paradox: why are we seeking imitations in the first place – and what does that say about what we are choosing to value and the relationship between what we are valuing and what those we’re imitating were valuing? Imitations of Silicon Valley abound because economic growth has become the new global obsession, to the point that whole countries are losing sight of invaluable treasures – of social fabric, spiritual depth and resilience, emotional intelligence – in their own backyards. And yet Silicon Valley achieved its economic growth in part precisely because it was not focused on economic growth. It was focused on the joy of innovating for the sake of innovating. Thus, we have seen many examples of transferable lessons from Silicon Valley, including the caution that the act of seeking transferable lessons may already violate one of the key conditions that was important for Silicon Valley to be Silicon Valley (e.g., that it ‘wasn’t trying to be Silicon Valley’). And yet, it is only human to find things elsewhere that we want to incorporate into our lives and communities. Under these circumstances, what are some things we should keep in mind? When we analyze a phenomenon for ‘transferable lessons’, we create a ‘figure-ground’ distinction, to use the famous gestalt terminology (e.g., Rubin, 2000), in which we focus on the salient components of the model that got our attention in the first place (the ‘figure’), and we tend to deemphasize or overlook the interdependencies of those aspects with the other aspects of reality that have become part of the ‘ground’. Instead, we need to pay attention to the deep ecology impacting the functioning of the whole and parts, both in the original context as well as in the context into which we seek to import. It is also important to consider other collateral,

STIMSON PAGINATION (M2469).indd 102

20/12/2010 15:12

Diversity and endogeny in regional development

103

unintended consequences an import may have on interrelated aspects of the milieu into which we seek to place it. When we look outside and see something we value, how or when might we or might we not want to try to adopt it? We will illustrate the question of transferrability in Silicon Valley with an analogue that shows kinds of interdependencies and deep ecology that should be considered. Twenty-five years ago Karl Weick (1983) introduced the notion of ‘reverse simulation’ into management literature. In a typical management or organizational simulation the goal is to test the relationships between variables in the controlled condition of a laboratory or classroom, extract a few principles or theories and propose various ways these lessons learned may be applied to ‘real world’ organizations. Following fields such as engineering, Weick suggested the opposite: instead of trying to simulate reality, create new realities in the lab and transfer them to the real world. What happens when we create in a laboratory (simulated environment) new forms of organizational processes and structures that differ from what are available in our lives outside, and we would like to incorporate some of those possibilities in our real lives? In such situations, we often look too narrowly at the conditions that made the possibility come alive in the particular context in which it occurred, and we often similarly overlook the conditions in the new context that might alter the meaning or valence of the import in the new environment (Wimsatt, 2007). We should attend not just to the phenomenon and its structure, but also to the nature of the broader relationships and ‘being’ of the totality of the context that allows a phenomenon to occur as it did (Wimsatt, 1980, 1994, 2007; Thatchenkery et al., 1999). We should also keep in mind a healthy humility about our ability to see and understand these complexities or to anticipate or resolve unintended consequences. Often our ‘solution’ to our last problem ends up creating many more – and worse – problems. Consider a deliberately simplified example: after a few days of specially structured group interactions, participants in a group dynamics laboratory learned several key principles of open communication, giving and receiving feedback and influencing others, and so on. Participants learned in the laboratory that when they spoke with authenticity and from their hearts, others listened to them better than when they had related superficially. However, when they tried to apply the learning at work after a few days it did not work, leading them to blame the ‘too theoretical academic knowledge’ as the root cause. Telling the boss that ‘I feel intimidated when you speak to me like that’ did not yield the type of open response the participant had received in the group dynamics lab a few days earlier. How can we apply the lessons learned about open communication to

STIMSON PAGINATION (M2469).indd 103

20/12/2010 15:12

104

Endogenous regional development

the real world? If we are to apply the principle of reverse simulation with an understanding of the biases of figure and ground phenomena, we will focus more on the conditions that facilitated the emergence of the learning than on the content of the learning as such (Thatchenkery et al., 1999). For example, a certain level of trust had to develop in the small group lab before a member was willing to give or receive feedback. Therefore, what should really transfer to the outside world is not a technique of giving feedback, but the conditions that allow trust to develop in a group. If sufficient trust had been developed between the boss and the employee, the feedback cited earlier might have been received more positively. Behara et al. (2008) have applied Weick’s notion of reverse simulation in an organizational context. While designing and implementing two distinct product delivery processes for a large financial services company in the Organizational Learning Laboratory at George Mason University, which one of the authors had founded, a new methodology called Empathic Knowledge Management was developed. It used the appreciative inquiry framework (Cooperrider and Srivastva, 1987) and utilized tacit knowledge of participants for process redesign in a learning laboratory environment. The focus in the lab was to create conditions that would allow staff from dispersed units in the organization to come together and share knowledge so that the process redesign time could be cut by half. Once a prototype loan processing tool was designed in record time in the Organizational Learning Lab, the focus was not on applying the new tool in the rest of the organization, but on figuring out how to create and maintain in their financial services organization the unique learning climate that had flourished in the laboratory during their three weeks of stay. In any such endeavor, it is important to keep in mind not just the similarities but also the differences between the original and transferred context at every level from content to process to the ‘being’ of the facilitators, participants and organizational actors and cultures. It is also important to remember that, since human and social systems are far more interdependent and complex than we will be able to theorize in our (necessarily) simplified views and models of them, whatever we think we ‘know’ is but a small and fallible veneer on what we don’t know. When other regions see value in what Silicon Valley has done, they need to look not just at the superficial level of success factors or salient characteristics, but also at: ● ●

the deeper context of Silicon Valley that allowed those characteristics to blossom; the deeper contexts in their own region that are and are not resonant with what they are valuing from Silicon Valley;

STIMSON PAGINATION (M2469).indd 104

20/12/2010 15:12

Diversity and endogeny in regional development ●

105

what additional collateral, cascading effects might result from importing and valuing the Silicon Valley model within their own community and how that fits or doesn’t fit with the endogenous values and vibrancy of their own community.

Perhaps a region that has not become highly commercialized can recognize, celebrate and support the contexts and cultures in which people can continue to express themselves and be motivated intrinsically as opposed to extrinsically (e.g., in which people engage in activities fundamentally as ends in themselves vs. instrumentally as means to other ends such as making money; Kohn, 1993). Perhaps a city’s residents can reframe and examine their unique strengths and regional advantages to come up with new services and innovation that will further support the community? What does it look like to reframe the fear of failure into embracing failures as learning opportunities? Appreciative intelligence and the embeddedness of human social and cultural realities suggest that the real message and transferable lessons from Silicon Valley are the following: 1.

2.

3.

Each locale should appreciate, value and embrace its own unique character and beauty, and be fully, thoroughly and elegantly itself rather than trying to imitate models plucked from a distant and foreign ecology. That when communities see aspects of life they appreciate in other places, they should be circumspect and reflective in why and how they may adopt these elements and what cascading unintended effects adoption might have. Value is not unitary but has many and diverse forms, and thus economic riches are not necessarily even the ideal or goal for a community to seek.

While the venture capitalist John Doerr may announce to the world that Silicon Valley represents ‘the largest legal creation of wealth in the history of the planet’, we must not forget that it is only one form of wealth that he is noticing. If we extend our conception of wealth, making it more diversified and pluralistic to include other forms of wealth such as spiritual and relational wealth, there are a number of Silicon Valleys in the world that we are not celebrating. It is only in our modern Western discourse that the creation of material and technical wealth has become an obsession held out for all humanity to replicate, and that the pursuit of its accumulation and concomitant concentration in certain regions appears to be a goal worth replicating (Sachs, 1992; Escobar, 1995). Moreover, a

STIMSON PAGINATION (M2469).indd 105

20/12/2010 15:12

106

Endogenous regional development

focus on materialist wealth accumulation may in fact even be incompatible with, and at the expense of, authentic relational and spiritual wealth, since it usually involves treating ourselves and each other instrumentally as means, rather than as ends. Amid the scholarship extolling the success of Silicon Valley, a few dissenting voices also can be heard. Rogers and Larsen (1984) in their book Silicon Valley Fever have raised fundamental questions about the social and business conditions of that environment, lamenting the veneration of the workaholic high-tech achiever lifestyle, the stinginess of the region in supporting the arts, the ‘dark side’ of life in the less affluent South County and a variety of other downsides that are often overlooked or taken to be unimportant when we extol a model for its successful outcomes. Florida and Kenney (1990) also caution us against embracing Silicon Valley as an unmixed blessing, reminding us that neither Silicon Valley nor Route 128 have been able to rectify the fissures that have become increasingly apparent in the US economy and social structures, and that our infatuation with Silicon Valley stems from the fact that their image of freewheeling, high-technology entrepreneurship and quick-shooting venture capital fits in nicely with the free enterprise ideology. If there ever was an ‘Eden of Cooperation’ these authors suggest it has degenerated into a fiery pit of destructive cut-throat competition when confronted with huge potential sums unleashing equal levels of greed. When other regions look to Silicon Valley for inspiration to promote economic and technological breakthroughs, policy-makers must look not merely at the success but also at the degree to which the spread of a highly mobile venture capital mindset might erode age-old traditions and cultural norms that have constituted the wealth of these societies. By teaching us to seek happiness in an ever-receding horizon of material externals, rapid economic growth may indeed cause more human unhappiness than the stable, sustainable lifestyles usually disparaged as ‘stagnant’. In depicting the Silicon Valley as an attractive model worth emulating, Gary Hamel’s description in his award-winning article in the Harvard Business Review unintentionally captures this paradox – simultaneously great and terrible, depending on one’s vantage point – of Silicon Valley culture: The Valley is the distilled essence of entrepreneurial energy. Its ethos is simple: If it’s not new, it’s not cool; if it’s not cool, it’s not worth doing. If you don’t own shares, you’re getting screwed. If you’ve been in the same job for more than two years, your career is over. If you haven’t been through an IPO, you’re a virgin. This is where a $2 million house is a teardown. This is where a Porsche is just one more compact car and sushi’s just another fast food. Never has so much wealth been created in so little time by so few people. If the Valley’s residents pause to think about it for even a nanosecond, they know they’re as

STIMSON PAGINATION (M2469).indd 106

20/12/2010 15:12

Diversity and endogeny in regional development

107

blessed as those who lived in Italy during the Renaissance. Like the Florentines and Venetians, they’re building a new age – an age of virtual presence, of globally interconnected communities, of frictionless commerce, of instantly accessible knowledge and stunningly seductive media. (Hamel, 1999)

In sum, through applying appreciative intelligence we can discern that what made Silicon Valley the poster child for economic development ‘best practices’ manuals and the envy of the world was its own deep authenticity and celebration of itself. The paradox of Silicon Valley is that attempts by other regions to emulate it are therefore fraught with challenges and lead to problematic choices. As suggested by, among other examples, the reverse simulation illustration above, rather than trying to be another Silicon Valley, regions should look deep within, find the many different forms of value in which they are rich, and celebrate their own unique constellation of strengths and possibilities for realizing a better world.

REFERENCES Acs, Z.J. and Audretsch, D.B. (1987), ‘Innovation, Market Structure, and Firm Size’, The Review of Economics and Statistics, 69(4), 567–74. Adams, S.B. (2005), ‘Stanford and Silicon Valley: Lessons on Becoming a Hightech Region’, California Management Review, 48(1), 29–51. Alchian, A.A. (1950), ‘Uncertainty, Evolution, and Economic Theory’, The Journal of Political Economy, 58(3), 211–21. Alvares, C. (1992), ‘Science’, in W. Sachs (ed.), The Development Dictionary, Zed Books, London, pp. 219–32. Audia, P.G. and Rider, C.I. (2005), ‘A Garage and an Idea: What More Does an Entrepreneur Need?’, California Management Review, 48(1), 6–28. Audretsch, D.B. and Stephan, P.E. (1996), ‘Company–Scientist Locational Links: The Case of Biotechnology’, The American Economic Review, 86(3), 641–52. Baron, J.N. and Hannan, M.T. (2002), ‘Organizational Blueprints for Success in High-tech Start-ups: Lessons from the Stanford Project on Emerging Companies’, California Management Review, 44(3), 8–36. Begg, I. (1999), ‘Cities and Competitiveness’, Urban Studies, 36(5–6), 795–809. Behara, R., Thatchenkery, T. and Kenney, C. (2008), ‘Empathic Knowledge Management: Reverse Simulation Experiments in a Learning Laboratory’, International Journal of Information Technology and Management, 7(3), 283– 314. Beugelsdijk, S. (2003), Culture and Economic Development in Europe, Doctoral Dissertation, Tilburg University. Block, N.J. and Dworkin, G. (1976), The IQ Controversy: Critical Readings, Pantheon Books, New York. Blumenthal, H. (1955), ‘The Economic Base of the Metropolis’, Journal of the American Institute of Planners, 21(4), 114–32. Burrell, G. and Morgan, G. (1985), Sociological Paradigms and Organizational Analysis, Ashgate, Aldershot, UK.

STIMSON PAGINATION (M2469).indd 107

20/12/2010 15:12

108

Endogenous regional development

Camagni, R. (2002), ‘On the Concept of Territorial Competitiveness: Sound or Misleading?’, Urban Studies, 39(13), 2395–411. Cooperrider, D. and Srivastva, S. (1987), ‘Appreciative Inquiry in Organizational Life’, Research in Organizational Change and Development, 1, 129–69. Delbecq, A.L. and Weiss, J. (2000), ‘The Business Culture of Silicon Valley: A Turn-of-the-Century Reflection’, Journal of Management Inquiry, 9(1), 37–44. Elgin, B. (2005), ‘Managing Google’s Idea Factory’, Business Week (3 Oct.), 88–90. Escobar, A. (1995), Encountering Development: The Making and Unmaking of the Third World, Princeton University Press, Princeton, NJ. Farmer, P. (2004), ‘On Suffering and Structural Violence’, in P. Farmer (ed.), Pathologies of Power, University of California Press, Berkeley, CA. Ferrary, M. (2003), ‘The Gift Exchange in the Social Networks of Silicon Valley’, California Management Review, 45(4), 120–38. Fleming, L. and Marx, M. (2006), ‘Managing Creativity in Small Worlds’, California Management Review, 48(4), 6–27. Florida, R. (2002), The Rise of the Creative Class, Basic Books, New York. Florida, R. (2003), ‘Entrepreneurship, Creativity, and Regional Economic Growth’, in D. Hart (ed.), The Emergence of Entrepreneurship Policy, Governance, Start-ups, and Growth in the U.S. Knowledge Economy, Cambridge University Press, New York, pp. 39–53. Florida, R. and Kenney, M. (1990), ‘Silicon Valley and Route 128 Won’t Save Us’, California Management Review, 33(1), 68–88. Frankl, V. (1963), Man’s Search for Meaning, Washington Square Press, New York. Freire, P. (1996), Letters to Cristina, Routledge, New York. Galison, P. (1987), How Experiments End, The University of Chicago Press, Chicago. Gardner, H. (1993), Multiple Intelligences: The Theory in Practice, Basic Books, New York. Gergen, K. (1994), Realities and Relationships, Harvard University Press, Cambridge, MA. Gergen, K. (2009, An Invitation to Social Construction, Sage, Thousand Oaks, CA. Gergen, K. and Thatchenkery, T. (2004), ‘Organization Science as Social Construction: Postmodern Potentials’, Journal of Applied Behavioral Science, 40(2), 228–49. Gowdy, J. (2000), ‘Terms and Concepts in Ecological Economics’, Wildlife Society Bulletin, 28(1), 26–33. Hamel, G. (1999), ‘Bringing Silicon Valley Inside’, Harvard Business Review, 77(5), 70–84. Haraway, D. (1990), Primate Visions: Gender, Race and Nature in the World of Modern Science, Routledge, New York. Heineman, M. (1981), ‘The Obsolete Scientific Imperative in Social Work Research’, Social Service Review, 55(3), 371–92. Heineman-Pieper, Jessica (2009), Reclaiming Responsibility: New Foundations for a Science of and by Persons, Saarbrücken, VDM Verlag Dr. Muller. Hempel, Carl G. (1965), Aspects of Scientific Explanation, Free Press, New York. Henderson, J. Vernon (2007), ‘Understanding Knowledge Spillovers’, Regional Science and Urban Economics, 37(4), 497–508. Hoover, E.M. (1948), The Location of Economic Activity, McGraw-Hill, New York.

STIMSON PAGINATION (M2469).indd 108

20/12/2010 15:12

Diversity and endogeny in regional development

109

Hoover, E.M. (1971), ‘Basic Approaches to the Study of Demographic Aspects of Economic Development: Economic-Demographic Models’, Population Index, 37(2), 66–75. Hulsink, W., Manuel, D. and Bouwman, H. (2007), Clustering in ICT: From Route 128 to Silicon Valley, from DEC to Google, from Hardware to Content (30 Oct.), ERIM (Erasmus Research Institute of Management) Report Series, Reference #ERS-2007-064-ORG. Isard, W. (1949), ‘The General Theory of Location and Space-Economy’, The Quarterly Journal of Economics, 63(4), 476–506. Kohn, A. (1993), Punished by Rewards, Houghton Mifflin, Boston, MA. Krugman, P. (1991), ‘Increasing Returns and Economic Geography’, Journal of Political Economy, 99(3), 483–99. Kuhn, T. (1962), The Structure of Scientific Revolutions, The University of Chicago Press, Chicago. Lecuyer, C. (2007), Making Silicon Valley: Innovation and the Growth of High Tech, 1930–1970, The MIT Press, Cambridge, MA. Lee, C.-M., Miller, W.F, Hancock, M.G., and Rowen, H.S. (eds) (2000), The Silicon Valley Edge: A Habitat for Innovation and Entrepreneurship, Stanford University Press, Stanford, CA. Lewontin, R.C. (1994a), ‘Facts and the Factitious in the Natural Sciences’, in J. Chandler, A.I. Davidson and H. Harootunian (eds), Questions of Evidence: Proof, Practice and Persuasion Across the Disciplines, University of Chicago Press, Chicago, pp. 478–91. Lewontin, R.C. (1994b), ‘A Rejoinder to William Wimsatt’, in J. Chandler, A.I. Davidson and H. Harootunian (eds), Questions of Evidence: Proof, Practice and Persuasion Across the Disciplines, University of Chicago Press, Chicago, pp. 504–9. Marshall, A. (1890), Principles of Economics, Macmillan, London. Metcalfe, B. (1998), ‘Asian Tour Provides Useful Insight on Silicon Valley’s Worldwide Internet Edge’, Info World (2 March). O’Brien, J.M. (2007), ‘The Pay-Pal Mafia’, Fortune, 156(11), 96–106. Popper, Karl R. (2002), The Logic of Scientific Discovery, Routledge, New York. Porter, Michael E. (1998), ‘Clusters and the New Economics of Competition’, Harvard Business Review, 76(6), 77–90. Putnam, R.D. (2000), Bowling Alone: The Collapse and Revival of American Community, Simon and Schuster, New York. Rahnema, M. and Bawtree, V. (1997), The Post-development Reader, Zed Books, London. Rogers, E.M. and Larsen, J.K. (1984), Silicon Valley Fever, Basic Books, New York. Rosnow, R.L. and Rosenthal, R. (1997), People Studying People: Artifacts and Ethics in Behavioral Research, Freeman, New York. Rubin, E. (2000), ‘Figure and Ground’, in S. Yantis (ed.), Visual Perception, Psychology Press, Philadelphia. Sachs, W. (ed.) (1992), The Development Dictionary, Zed Books, New York. Saxenian, A. (1996), Regional Advantage: Culture and Competition in Silicon Valley and Route 12, Harvard University Press, Cambridge, MA. Saxenian, A. (1999), Silicon Valley’s New Immigrant Entrepreneurs, Public Policy Institute of California, San Francisco. Saxenian, A. (2000), ‘Networks of Immigrant Entrepreneurs’, in C.-M. Lee,

STIMSON PAGINATION (M2469).indd 109

20/12/2010 15:12

110

Endogenous regional development

W.F. Miller, M.G. Hancock and H.S. Rowen (eds), The Silicon Valley Edge: A Habitat for Innovation and Entrepreneurship, Stanford University Press, Stanford, CA. Saxenian, A. (2006), The New Argonauts: Regional Advantage in a Global Economy, Harvard University Press, Cambridge, MA. Shipley, C. (2006), ‘What Makes Silicon Valley So Different?’ (19 Oct.), Wisconsin Technology Network, available at: http://wistechnology.com/articles/3416, accessed 23 May 2010. Shiva, V. (2000), Stolen Harvest, India Research Press, New Delhi. Thatchenkery, T. (2006), ‘Organization Development in Asia: Globalization, Homogenization, and the End of Culture-specific Practices’, in B. Jones and M. Brazzel (eds), The NTL Handbook of Organization Development and Change: Principles, Practices, and Perspectives, Pfeiffer/Wiley, San Francisco, pp. 387–403. Thatchenkery, T. and Metzker, C. (2006), Appreciative Intelligence: Seeing the Mighty Oak in the Acorn, Berrett-Koehler, San Francisco. Thatchenkery, T., Behara, R. and Kenney, C. (1999), ‘Building Capabilities for Change through Laboratory Simulations’, Developments in Business Simulation and Experiential Learning, 26, 144–6. Weick, K. (1983), ‘Utilization as Reverse Simulation: Making the World More Like the Laboratory’, in R.H. Kilmann, K.W. Thomas, D.P. Slevin, R. Nath and S.L. Jerrell (eds), Producing Useful Knowledge for Organizations, Praeger, New York, pp. 494–520. Weiss, J. and Delbecq, A. (1987), ‘High-technology Cultures and Management: Silicon Valley and Route 128’, Group and Organization Studies, 12(1), 39–54. Wimsatt, W. (1976), ‘Reductionism, Levels Organization, and the Mind-Body Problem’, in G. Globus, G. Maxwell and I. Savodnik (eds), Consciousness and the Brain: A Scientific and Philosophical Inquiry, Plenum, New York, pp. 205–67. Wimsatt, W. (1980), ‘Reductionistic Research Strategies and Biases in the Units of Selection Controversy’, in T. Nickles (ed.), Scientific Discovery, Reidel, Boston, pp. 213–60. Wimsatt, W. (1994), ‘The Ontology of Complex Systems: Levels of Organization, Perspectives, and Causal Thickets’, Canadian Journal of Philosophy, 20(Suppl.), 207–74. Wimsatt, W. (2007), Philosophy for Limited Beings: Piecewise Approximations to Reality, Harvard University Press, Cambridge, MA. Zinn, H. (2003), A People’s History of the United States, Harper-Collins, New York.

STIMSON PAGINATION (M2469).indd 110

20/12/2010 15:12

7.

An exploratory approach to model determinants of endogenous regional growth performance Robert Stimson and Roger Stough*

INTRODUCTION Regional economic development may be viewed both as a process and a product (or outcome). It is a multi-dimensional phenomenon, involving many actors and being influenced by many factors. It seems to defy precise definition, incorporating both quantitative and qualitative dimensions. Further, the study of regional economic development incorporates a concern not only with analysis and modelling, but also a concern for policy and strategy that may facilitate the regional development process and facilitate regional change. The process of regional economic development is certainly dynamic. Over time, various theoretical approaches to theorizing about and modelling regional growth and development have evolved, and, during the last couple of decades, there has been an increasing emphasis on the role of endogenous factors. Stimson et al. (2003) have proposed the notion of the ‘virtuous circle’ as a path to achieving sustainable regional economic development (see Figure 7.1). They suggest that the ‘virtuous circle’ may be maintained through the mediating or intervening effects of factors such as effective leadership as it might be used to change and adjust institutions in order to adapt the structure, processes and infrastructure of a regional economy that is appropriate and needed to meet and anticipate changing circumstances, and to facilitate the optimal use of the region’s resource endowments and to assist industries to tap their full market potential. Stimson et al. (2003) put forward the proposition that: ●

strong leadership means a region will be proactive in initiating regional economic development strategy to monitor regional performance; 111

STIMSON PAGINATION (M2469).indd 111

20/12/2010 15:12

112

Endogenous regional development Strong proactive leadership Good use and tapping of potentials Sustainable development Market conditions

Resource endowments

Mechanisms for using and tapping

Vision for future development

Strategy, plans, processes

Facilitate institutional change to enhance regional capacity and capability Effective institutions and regional infrastructure

Source: Stimson et al. (2003).

Figure 7.1

● ●

The virtuous circle for sustainable regional development

leadership will set a vision for its future development; and the institutions in the region will implement processes and plans that will facilitate institutional change.

That, in turn, might enhance the capacity and capability of the region to: ● ● ●

positively adjust to changing circumstances; attain a good fit with market conditions; and harness its resource endowments,

in order to maintain and improve its performance and to achieve sustainable development as a learning region and to be one that is competitive. In this chapter we first provide a brief discussion of the evolution of approaches to regional economic development, emphasizing endogenous factors. We then outline an approach we have been developing to formulate and operationalize a new model framework to measure regional endogenous performance and model the determinants of spatial variations in regional endogenous growth performance. Third, we briefly summarize the results of those exploratory attempts to apply a model framework.

STIMSON PAGINATION (M2469).indd 112

20/12/2010 15:12

An exploratory approach to model determinants

113

THE EVOLUTION OF REGIONAL ECONOMIC DEVELOPMENT THEORY The importance of endogenous factors in regional economic development and growth (and decline) has long been recognized and is largely the basis of the so-called ‘new growth theory’ that has been popularized over the last couple of decades. Theories of endogenous growth place emphasis not only on regional resource endowments and human capital – which have always been viewed as important factors affecting economic development – but also on technology, entrepreneurship and institutional factors, including the role of leadership. However, there is neither a standard definition of nor a method of measuring endogenous growth, nor the consistent specification of the factors that determine variations in regional endogenous growth and performance. The evolution of theories on regional economic growth and development has seen a shift in the emphasis placed on exogenous factors to endogenous factors. Traditional regional economic development approaches were embedded in neoclassical economic growth theory, based largely on the Solow (1956, 2000) model. These have been replaced over the last couple of decades by a suite of models and arguments that are commonly known as the ‘new growth theory’ where the focus is directed towards endogenous factors and processes (see, for example, Johansson et al., 2001). Some of the key contributions to this shift in emphasis towards endogenous processes as drivers in regional growth and development are discussed in a recent review by Stimson et al. (2009a). This includes, inter alia, the following developments and contributions: 1.

2.

3. 4.

In the late 1970s, Norton and Rees (1979) and Rees (1979) proposed that technology was a prime driver in regional economic development. From that time, theorists such as Romer (1986, 1990), Lucas (1985), Barro (1990), Rebelo (1991), Grossman and Helpman (1991) and Arthur (1994) sought to explain technical progress as it generates economic development as an endogenous effect rather than accepting the neoclassical view of long-term growth being due to exogenous factors. Thomas (1975) and later Erickson (1994), among others, showed how technological change is related to the competitiveness of regions. Norton and Rees (1979) and Erickson and Leinbach (1979) showed how the product cycle, when incorporated into a spatial setting, may impact differentially on regions through three stages, namely an innovation stage, a growth stage and a standardization stage.

STIMSON PAGINATION (M2469).indd 113

20/12/2010 15:12

114

5.

6.

7.

8.

9.

10.

11.

12.

Endogenous regional development

Markusen (1985) extended the product cycle theory of regional development by articulating how profit cycles and oligopoly in various types of industrial organization and corporate development can magnify regional development differentials. The concept of innovative milieu was formulated to explain the ‘how, when and why’ of new technology generation. That notion linked back to the importance of agglomeration economies and localization economies that may lead to the development of new industrial spaces (Scott, 1988; Porter, 1990; Krugman, 1991). Some theorists, such as Fukuyama (1995), have suggested that it is not just economic but also value and cultural factors – including social capital and trust – that are important in the rise of technology agglomerations as seen in the Silicon Valley phenomenon. The effect of human capital skills and income in explaining differentials in levels of regional economic performance has received a lot of attention (see, for example, Hanushek and Kimko, 2000; Goetz and Rapasingla, 2001). The effect of industrial structure on regional stability and growth has been emphasized, with one argument being that industrial diversity and a trend towards diversification of industry sector employment enhances opportunities for growth and development (see, for example, Henderson et al., 1995; Gordon and McCann, 2000), although some researchers claim that there is scant empirical evidence on that proposition (Kaufman, 1993; Lande, 1994; Productivity Commission, 1998). The power of urban scale and agglomeration on regional performance is also emphasized (Taylor et al., 2002), with a study by Duranton and Puga (2000) reviewing national urban systems suggesting that larger cities are more diversified, that individual city-size rankings and individual city specialization tend to be stable over time, and that specialized and diversified cities co-exist across a national urban system. Rees (2001) points out that technology-based theories of regional economic development need to incorporate the role of entrepreneurship and leadership, particularly as factors in the endogenous growth of regions’ market circumstances. There now seems to be an almost universal realization of what Garlick et al. (2006) refer to as the ‘institutional embeddedness’ of endogenous processes and factors in regional development.

The term ‘new growth theory’ has been coined to describe the transitions such as those referred to above in the evolution of thinking on

STIMSON PAGINATION (M2469).indd 114

20/12/2010 15:12

An exploratory approach to model determinants

115

regional economic development over the last three to four decades, with the increasing emphasis being placed on the role of and the importance of, endogenous factors. Importantly, in the new growth theory models that have emerged, allowance is made for both agglomeration effects (economies of scale and externalities) and market imperfections, with the price mechanism not necessarily generating an optimal outcome through efficient allocation of resources. Also, the processes of capital accumulation and free trade do not necessarily lead to convergence between regions, with positive agglomeration effects concentrating activity in one or a few regions through self-enforcing effects that attract new investment. Most importantly, the new growth theory allows for both concentration and divergence. As discussed by Stimson, Stough and Roberts (2006), these evolutionary developments in regional economic growth and development theory are welcome for regional economic development analysts and planners because, among other things, they explicitly introduce a spatial dimension into economic growth theory, a dimension that was ignored in neoclassical economic development theory. That evolution is particularly important as the role of regions in national economies has changed significantly since the 1970s. This has been a result to some degree of the effects of globalization and structural change and adjustment. Understanding those processes of change is crucial for analysing and understanding differentials in the patterns of regional economic performance and for formulating and implementing regional economic development planning strategy. But just as important is the need to carefully address endogenous factors and processes to enhance the capacity and capability of a region to proceed along a path of achieving sustainable development and growth.

AN APPROACH TO MEASURING AND MODELLING ENDOGENOUS REGIONAL GROWTH A New Model Framework The authors and their collaborators have been seeking to develop and operationalize a new model framework (see Figure 7.2) to measure and investigate the determinants of spatial variations in endogenous regional growth performance. In particular, existing theories of regional economic development tend to underplay the significance of leadership and other institutional factors. Stimson, Stough and Salazar (2003, 2005) have proposed a model

STIMSON PAGINATION (M2469).indd 115

20/12/2010 15:12

116

Endogenous regional development

Quasi-independent Variables

Intervening Variables

Dependent Variable(s)

The dynamic interrelationships that act to create the catalysis for regional development (I) Institutions (E) Entrepreneurship

Resource endowments and market conditions (RE, M)

OUTCOME A region that is: competitive entrepreneurial sustainable (RED)

(L) Leadership

Direct effects Indirect effects

Measure and evaluation change over time Benchmark performance (e.g., regional shift component in shift-share analysis)

Source: Stimson, Stough and Salazar (2003).

Figure 7.2

A new model framework for the regional economic development process

framework conceptualizing regional economic development (RED) and to explain spatial variations in regional endogenous growth, which specifies three key intervening or mediating factors:1 ● ● ●

leadership (L); institutional factors (I); entrepreneurship (E);

which are seen as creating both catalysts and vehicles for better utilization of a region’s resource endowments (RE) and its market fit (M). The model may be specified as follows: RED= f [(RE, M) mediated by (L, I, E)] A Definition of Regional Endogenous Growth: A Proxy Measure for the Dependent Variable Despite the evolving focus on endogenous factors in the regional economic development and growth literature, it is somewhat surprising that there

STIMSON PAGINATION (M2469).indd 116

20/12/2010 15:12

An exploratory approach to model determinants

117

is no standard definition of endogenous regional economic growth (or decline) in terms of the specification of an agreed variable that measures it. Thus, a key issue is: what is an appropriate proxy measure of a region’s endogenous growth? Furthermore, it is also surprising that there is a lack of operational models to measure the affect of factors such as those discussed above on explaining spatial variations in regional endogenous growth performance in a nation or state. In general economic analysis, most studies derive and measure a variable for endogenous growth by using Ordinary Least Squares,2 or more recently, panel data analysis. However, the data required to do so is often not readily available at the disaggregated regional level. As such, many economic geographers and regional economists have turned to other techniques. Approaches used to measure regional endogenous economic growth/ development have been discussed by Stimson, Stough and Salazar (2003, 2005). It might take the form of a proxy or surrogate measure such as: (1) the aggregated (across all industry sectors) regional differential shift component value in a shift-share analysis, a common technique in analysing regional differential performance used by economic geographers and regional economists; or (2) an employment scale weighted location quotient change over time, standardized by the size of the region’s labour force. Such measures can then be used as the dependent variable in a model of regional endogenous economic growth/development. In the exploratory model applications discussed later in this chapter, there are attempts to combine a range of endogenous factors to reflect regional economic growth (which is measured by employment, given the data constraints on measuring production, at the regional level) rather than one measure, such as technology or level of savings. This is done by using shift-share analysis as proposed (see above) by Stimson, Stough and Salazar (2003b, 2005). The reason for adopting this approach is pragmatic as secondary data tends to be readily available in most countries to perform a shift-share analysis, and typically that may be achieved using census data for industry employment in regions. The regional shift component is a reasonable surrogate measure of the degree to which employment growth or decline in a region is due to endogenous or ‘within-region’ processes after accounting for exogenous factors, such as national shift and the industrymix shift effects. Indeed, that is what the regional shift component is purported to measure. While the authors recognize that using the differential/ regional shift component derived from a shift-share analysis is not an ideal method of measuring endogenous regional economic growth (or decline), it is nonetheless considered to be the most optimal given the lack of data at the regional level to operationalize other measurement approaches.

STIMSON PAGINATION (M2469).indd 117

20/12/2010 15:12

118

Endogenous regional development

Potential Independent and Intervening Variables The potential sets of variables proposed by Stimson, Stough and Salazar (2003b, 2005) that might be appropriate as measures of the independent and the mediating factors in the model3 are the following: 1.

RE = resource endowments, which might be measured by a set of variables such as: – – – – – – – – –

– – – 2.

area size of the region; climate; topography; agglomeration of industry key sectors (measured by location quotients for employment in industry sectors); population size and rate of growth/decline; education levels (a derived index of human capital) and literacy; per capita income, income distribution and income distribution change over time; housing ownership; investment in industrial and commercial construction, benchmarked to the region’s national share vis-à-vis its national share of population; infrastructure investment (per capita), such as on roads, schools, hospitals, and so on; industrial structure and change in industrial structure (measured by an industrial diversity index); regional organizational slack.

M = market fit, which might be measured by a set of variables such as: – – – –

basic economic activity in major industry sectors (measured by location quotients for employment in industry sectors); airline connections with other regions/cities; road freight in/out movements; volume and value of exports in key products and services.

It would also be useful to use variables that measure the degree to which the region’s products fit with changing demand and related markets, to ascertain the degree to which supply fits the local market, and to evaluate the extent to which the local infrastructure provides the necessary linkages to export markets. 3.

L = leadership, which might be measured by a set of variables such as:

STIMSON PAGINATION (M2469).indd 118

20/12/2010 15:12

An exploratory approach to model determinants

– – – – –

4.

119

the degree of change/stability in local political leadership; expert assessment of leadership quality; corporate headquarters located in the region; density of business and community organizations per 10 000 population; size of regional organizations (public and non-profit) budgets/ employment.

I = institutions, which might be measured by a set of variables such as: –

institutional thickness (corporate and community organizations per 10 000 population); – layers of government/government fragmentation; – formal institutions of governance, measured by number of public agencies per 10 000 population; – number of headquarters of major corporations (e.g., Fortune 1000 firms); – value foundation capitalization per 10 000 population; – government fragmentation; – level of regional organizations (number and budget level); – an index of social capital. 5.

E = entrepreneurship, which might be measured by sets of variables such as: – – – – –

churn rate or business start-up rate; venture capital activity; corporate venturing activity; patents issued per 10 000 workers; location quotient of employment in ‘symbolic analyst’ occupations.

Stimson, Stough and Salazar (2005) argue that RED is positively related to RE, M, L, I and E, but that there are likely to be lead and lag effects in the short to intermediate run, and perhaps cyclical effects in the longer run. Thus: REDt = REt−1 + Mt−1 + (It−1 to It−10/10) + Lt−2 + Et−2 + e Exploratory Model Applications So far there have been two exploratory attempts to apply this model framework to analyse spatial differentials in regional endogenous growth

STIMSON PAGINATION (M2469).indd 119

20/12/2010 15:12

120

Endogenous regional development

performance and identify factors that may explain those variable patterns in regional endogenous growth performance: (1) In the first application, Stimson, Robson and Shyy (2005, 2006, 2009b) have conducted an exploratory analysis using the OLS technique to estimate a model of regional endogenous growth performance across non-metropolitan regions in each of the five mainland states of Australia. (2) In the second application, Stough et al. (2007) have conducted an exploratory study of variations in endogenous regional growth across US Metropolitan Statistical Areas (MSAs). In both of these model applications, the dependent variable measure of regional endogenous growth is the regional or differential shift component across all industry sectors derived from a shift-share analysis of employment change over a specified period of time, standardized by size of the regional labour force. Both studies employ a standard OLS technique to estimate the model of regional endogenous growth. This dependent variable is regressed on a set of endogenously developed regional variables on demography and economy, and, where possible leadership, institutions and entrepreneurship, that are selected as explanatory variables. Both static point-in-time values around the base year and dynamic change-over-time values over the period of time being studied are used to model these explanatory variables, which are assumed a priori to be the possible determinants of regional endogenous growth.4 While stepwise regression is the core of the analytical methodology, OLS is used to estimate the ‘fit’ at each iteration as variables are stepped out of the initial or general starting model. In that modelling, consideration is given to testing for spatial autocorrelation to address spatial proximity/spillover effects. In the sections that follow we provide a summary discussion of the results of these two exploratory model applications.

THE AUSTRALIAN APPLICATION The purpose of the Stimson, Robson and Shyy (2005, 2006, 2009b) exploratory model application was to empirically examine and model patterns of regional endogenous employment growth in Local Government Areas (LGAs) across non-metropolitan areas of the five mainland states of Australia and to identify which factors might explain the spatial variations in the level of endogenous growth performance over the decade 1991 to 2001. It was restricted to using secondary data on regional demographic and economic characteristics derived from the census.

STIMSON PAGINATION (M2469).indd 120

20/12/2010 15:12

An exploratory approach to model determinants

121

The five mainland states were modelled separately in order to generate comparative data.5 Model Variables The modelling used a set of 27 independent or explanatory variables (see Table 7.1) – both static and dynamic – representing measures relating to factors that are hypothesized to play a significant role in affecting regional economic development and growth as suggested in the international and local Australian literature appraised by the researcher. Those factors relate to aspects of: ● ● ● ● ●

regional industrial structure and specialization/diversification; population size and growth; labour force participation and unemployment; human capital and income; occupational structure; the effects of proximity to the state capital city (that is the metropolitan area and proximity to the coast.

Those 27 variables were included in an OLS general model, initially a convention to estimate the model of regional endogenous growth across regional LGAs of the five mainland states of Australia. Following that, a stepwise approach was used to determine a specific model.6 Patterns of Endogenous Regional Growth Performance Stimson, Robson and Shyy (2005) mapped and analysed the spatial patterns of variation in the regional endogenous employment growth (or decline) performance over the decade 1991 to 2001 of the nonmetropolitan regional LGAs in the five mainland states of Australia on the REG_SHIFT dependent variable of endogenous growth. It was found that: ●

●

●

in New South Wales, overall 45 of the regional LGAs experienced positive endogenous growth, while 90 had experienced negative growth; in Victoria, only eight of the regional LGAs experienced positive endogenous growth, while 31 LGAs displayed negative endogenous growth performance; in Queensland, 39 of regional LGAs had experienced endogenous growth, while 79 LGAs displayed negative growth;

STIMSON PAGINATION (M2469).indd 121

20/12/2010 15:12

122

Table 7.1

Endogenous regional development

Definition of variables used in the non-metropolitan regions model for the five mainland states of Australia

Variables Dependent variable REG_SHIFT Independent variables SPEC_91 SPEC_CH SCI_91-01 SCI_CH L_INC_01 UNEMP_91 UNEMP_CH L_POP_91 POP_CH LQMAN_91 LQMAN_CH LQPBS_91 LQPBS_CH LQPER_91 LQPER_CH UNIQUALS_91 UNIQUALS_CH TECHQUALS_91 TECHQUALS_CH ROUTW_91 INPERS_91 SYMBA_91

STIMSON PAGINATION (M2469).indd 122

Definition of Variables Regional Shift (from 1991 to 2001)/Labour Force (1991) (i.e., employed + unemployed) Specialization Index for 1991 across 17 industry sectors Change in Specialization Index from 1991 to 2001 Structural Change Index for 1991 to 2001 Change in the Structure Change Index from 1991–1996 to 1996–2001 Log (Average Annual Income for 2001) Unemployment Rate in 1991 (for all persons) Unemployment Rate Change from 1991 to 2001 (for all persons) Log (Population for all persons) in 1991 Percentage Point Change in Population from 1991 to 2001 Location Quotient for Manufacturing Industry in 1991 Change in Location Quotient for Manufacturing Industry from 1991 to 2001 Location Quotient for Property and Business Services Industry in 1991 Change in Location Quotient for Property and Business Services Industry from 1991 to 2001 Location Quotient for Personal and Other Services Industry in 1991 Change in Location Quotient for Personal and Other Services Industry from 1991 to 2001 Proportion of Population with a Bachelor Degree or higher in 1991 Change in Proportion of Population with a Bachelor Degree or higher from 1991 to 2001 Proportion of Population with a Technical Qualification in 1991 Change in Proportion of Population with a Technical Qualification from 1991 to 2001 Proportion of Total Occupations (all persons) as Routine Production Workers for 1991 Proportion of Total Occupations (all persons) as Inperson Service Workers for 1991 Proportion of Total Occupations (all persons) as Symbolic Analysts for 1991

20/12/2010 15:12

An exploratory approach to model determinants

Table 7.1

123

(continued)

Variables

Definition of Variables

ROUTW_CH

Change in Proportion of Total Occupations (all persons) as Routine Production Workers from 1991 to 2001 Change in Proportion of Total Occupations (all persons) as In-person Service Workers from 1991 to 2001 Change in Proportion of Total Occupations (all persons) as Symbolic Analysts from 1991 to 2001 Border is adjacent to coastline (dummy variable 0 = NO, 1 = YES) Is adjacent to the State Capital Statistical District (SD) (dummy variable 0 = NO, 1 = YES)

INPERS_CH SYMBA_CH A_COAST P_METRO

Source: Stimson, Robson and Shyy (2006). ● ●

in South Australia, 26 of the regional LGAs experienced positive endogenous growth performance, while it is negative for 23 LGAs; in Western Australia, there was positive endogenous growth in 32 of the regional LGAs while 80 of them displayed negative growth.

The mapping approach developed by Stimson, Robson and Shyy 2006, 2009b) is novel, using GIS to visualize patterns of variations in regional endogenous employment growth/decline performance according to the population size category of the region. An example for the state of New South Wales (the state with the highest population) is provided in Figure 7.3. Model Results The OLS general model run for all five of the mainland states produced high R-squared values ranging from top values of 0.98 for South Australia, 0.96 for Victoria and 0.92 for Western Australia, to the slightly lower values of 0.88 for Queensland and 0.85 for New South Wales. The general model thus explains much of the variance in dependent variables on regional endogenous growth measured by the regional shift in employment change over the decade 1991–2001 across the non-metropolitan LGAs. The results from the general model indicated not only some commonality but also a degree of variability in magnitude and direction of the effects of the independent variables in explaining the spatial variation in levels of endogenous employment growth/decline across non-metropolitan regional LGAs in the five mainland states analysed.

STIMSON PAGINATION (M2469).indd 123

20/12/2010 15:12

124

Endogenous regional development

Source: Stimson, Robson and Shyy (2006).

Figure 7.3

Patterns of endogenous growth performance: New South Wales

For the application of the specific model, the adjusted R-squared values derived are high, explaining as much as 98 per cent of the variance for South Australia, 96 per cent for Victoria and 92 per cent for Western Australia, 88 per cent of the variance for Queensland and 86 per cent for New South Wales. The test conducted to investigate the degree of spatial dependence showed that only in South Australia is any significant degree of spatial dependency apparent with the Moran’s I statistic being 0.154, which is well below the maximum value of 1. For the other states the Moran’s I statistic varies from −0.012 for Victoria to 0.051 for Western Australia. Not surprisingly there are differences between the five mainland states in the number of significant variables and the mix of independent variables

STIMSON PAGINATION (M2469).indd 124

20/12/2010 15:12

125

STIMSON PAGINATION (M2469).indd 125

20/12/2010 15:12

UNIQUALS_91 − UNIQUALS_CH + TECHQUALS_91 − TECHQUALS_CH +

LQMAN_CH+ LQPBS_91 + LQPBS_CH + LQPER_91 +

POP_CH +

UNEMP_91 − UNEMP_CH −

New South Wales

ROUTW_91 − ROURW_CH −

UNIQUALS_91 − UNIQUALQ_CH+

INPERS_CH +

ROUTW_CH +

UNIQUALS_91 − UNIQUALS_CH + TECHQUALS-91 −

INPERS_91 −

ROUTW_91 −

TECHQUALS-91 +

ROUTW_CH + INPERS_91 +

TECHQUALS_CH +

LQPER_91 + LQPER_CH +

UNEMP_CH − L_POP_91 + POP_CH +

SPEC_ 91 + SPEC_CH + SCI_91-01 + SCI_91-01 +

Western Australia

LQPER_91 +

LQMAN_CH+

L_POP_91 + POP_CH +

SPEC_91 + SPEC_CH +

South Australia

LQPBS_91 + LQPER_91 +

UNEMP_CH − L_POP_91 + POP_ CH +

L_INC_01 +

SPEC_91 + SPEC_CH +

Queensland

LQPBS_91 +

L_POP_91 − POP_CH +

SCI_91-01 +

Victoria

Summary of specific model results: statistically significant independent variables explaining spatial differences in regional endogenous employment performance (1991–2001) in the five mainland states, Australia (direction of influence, positive/negative)

SPEC_91 SPEC_CH SCI_91-01 SCI_CH L_INC_01 UNEMP_91 UNEMP_CH L_POP_91 POP_CH LQMAN_91 LQMAN_CH LQPBS_91 LQPBS_CH LQPER_91 LQPER_CH UNIQUALS_91 UNIQUALS_CH TECHQUALS_91 TECHQUALS_CH ROUTW_91 ROUTW_CH INPERS_91 INPERES_CH

Independent variables

Table 7.2

126

STIMSON PAGINATION (M2469).indd 126

20/12/2010 15:12

Source:

New South Wales

(continued)

Stimson, Robson and Shyy (2006).

SYBBA_91 SYMBA_CH D_COAST D_METRO

Independent variables

Table 7.2

SYMBA_91 − SYMBA_CH −

Victoria SYMBA_91 +

Queensland

D_METRO −

South Australia

D_COAST − D_METRO −

Western Australia

An exploratory approach to model determinants

127

0.04

Residuals

Tweed

0.03

Snowy River Muswellbrook

Residuals

0.02

Tamworth

Cobar

Broken Hill Bega Valley

0.01

Shoalhaven Armidale Dumaresq

Queanbeyan

Dubbo

0

Wagga Wagga

Albury 20

0

40 Bathurst

60

80

Deniliquin 100 Orange

120 Griffith

Wollongong

140 Grafton

Glen Innes

–0.01

Ballina

–0.02

Greater Taree Newcastle

Coffs Harbour

Narrabri

LGAs

Source:

Stimson, Robson and Shyy (2006).

Figure 7.4

Order of regional LGA residuals for the specific model: New South Wales

in the specific model. Table 7.2 provides a summary of those significant explanatory factors. Stimson, Robson and Shyy (2005, 2009a) plotted on a graph the order of residual scores of LGAs from the line of best fit derived from the specific model (that is, the stepwise regression). And they also mapped the spatial patterns of the residual scores for LGAs. Here we only reproduce the plot for New South Wales (Figure 7.4), which shows there are no discernable outliers, with only two LGAs being above the +0.03 deviation. The large majority of LGAs fall within the range +/−0.01. From this exploratory model application to the five mainland states of Australia, it is evident that the specific model gives a relatively good fit, especially in Victoria and South Australia and to a slightly lesser degree in New South Wales, while in Queensland and Western Australia there is a greater degree of deviation in from the line of best fit derived from the model. The modelling conducted by Stimson, Robson and Shyy (2005, 2006, 2009a, 2009b) lends support to the importance of variables relating to industrial structure and industry specialization, population size and to population change, levels of human capital and aspects of job concentration in industry sectors and occupation categories that have been hypothesized as being important in explaining differences in regional endogenous growth performance. But there are considerable differences between the five mainland states of Australia in the degree to which the individual independent variables are significant independent (explanatory) factors accounting for variations between the regional LGAs in their regional

STIMSON PAGINATION (M2469).indd 127

20/12/2010 15:12

128

Endogenous regional development

endogenous growth performance over the decade 1991–2001. Further, there are also some variations between those states in the direction of the impact of some of the variables on regional endogenous growth. From the specific model results, the following conclusions are drawn by Stimson, Robson and Shyy (2006, 2009b): 1.

2.

3. 4. 5.

6.

7.

8.

Population growth emerges as a particular factor with a strong positive impact in all five mainland states, and population size at the beginning of the decade has an important positive impact in all states except New South Wales. The level of industry sector specialization, and the change in it over time, both have an important positive effect in all states except New South Wales, and the structural change index and the dynamics in it over time are important positive factors in Queensland, South Australia and Western Australia. The level of income at the end of the decade 1991 to 2001 is a significant factor only in Queensland. Only in New South Wales do the unemployment variables have an impact. There is a mixed picture regarding the impacts of the degree of concentration of jobs in broad industry sectors, and the effects are most apparent in New South Wales and in Western Australia. But there is neither a consistent directional effect for these variables, nor for the change over time in their location quotients. There is a particularly marked impact from the incidence of people with university and technical qualifications in New South Wales and Queensland, but to a lesser degree in Victoria and the other states. The incidence of people with university qualifications at the beginning of the decade has a negative impact, but the change in that incidence over the decade is a positive impact factor in New South Wales and Western Australia, where the change in incidence of people with technical qualifications also has a positive impact. There is no really discernable pattern in the impact of the incidence of jobs in the Reich (1991) broad occupation categories, except in Victoria where the effects are negative, while in South Australia, Western Australia and Queensland the effects are positive when such a variable is significant. Only in Western Australia and South Australia is it evident that the proxy variables relating to location on or near the coast or proximity to the metropolitan area have a significant effect, and that effect is negative.

STIMSON PAGINATION (M2469).indd 128

20/12/2010 15:12

An exploratory approach to model determinants

129

THE US APPLICATION The purpose of the Stough et al. (2007) exploratory model was to empirically examine and model the sources of regional endogenous growth across the US MSAs for the period 1999–2002. Specifically, the modelling attempts to uncover or examine which factors assumed to be endogenous to a region are actually affecting changes in the regional endogenous employment growth of 245 MSAs between 1999 and 2002.7 Method The US study followed the methodology employed in the Stimson, Robson and Shyy (2005, 2006, 2009b) exploratory study in Australia discussed above. It uses the same set of variables relating to demographic and economic characteristics of a region, but in this US application Stough et al. (2007) added to the set of independent or explanatory data variables a selection of variables that are surrogate measures (that is, indicators) of factors relating to leadership, institutions and entrepreneurship. Thus, this model application to MSAs in the US is an exploratory attempt to use the full model framework for modelling regional endogenous growth proposed by Stimson, Stough and Salazar (2003, 2005) and as set out earlier in Figure 7.1. The model uses secondary data available from multiple data sources, including the US Census Bureau, Bureau of Economic Analysis (BEA) and Bureau of Labor Statistics (BLS). Spatial Patterns of Endogenous Employment Growth Performance The map in Figure 7.5 shows the pattern of MSA scores on the dependent variable measuring endogenous employment growth/decline for the period 1999–2002. On the map the MSAs are differentiated according to whether their score on the dependent variable was weak or strong, positive or negative, using the ‘cut-off points’ method of classification. MSAs are differentiated according to their population size category. It is evident that MSAs with different levels of population experienced different levels of regional endogenous growth. Generally MSAs with small population size experience relatively higher regional endogenous economic growth than those with larger populations. However, the variance for the small-size MSAs is much greater than for the larger-sized MSAs. Specifically, the regional endogenous growth rate of the mediumsized MSAs (population size between 0.2 million and 1 million) is lower than that of the MSAs with a smaller population (less than 0.2 million) by

STIMSON PAGINATION (M2469).indd 129

20/12/2010 15:12

130

STIMSON PAGINATION (M2469).indd 130

20/12/2010 15:12

Pattern of MSA scores on the dependent variable standardized endogenous employment growth, 1999–2002, by population size category

The authors.

Figure 7.5

Source:

An exploratory approach to model determinants

131

9.53 per cent. Further, the rate for larger MSAs (population size between 1 million and 6 million) is even lower than that of the MSAs with population size less than 0.2 million by 12.52 per cent. It is interesting that the small-size MSAs dominate in the 20 top-ranking MSAs with the highest positive scores on the endogenous employment growth dependent variable, with only four medium-size MSAs with populations between 200 000 and 1 million being in the top 20 ranked MSAs. Not one of the large MSAs with populations more than 1 million is in the top 20 ranked MSAs. In contrast, nine of the large MSAs are in the bottom 20 ranked MSAs with the greatest negative scores on the endogenous employment growth dependent variable, and only one of the small MSAs was in that list of 20 bottom-ranked MSAs. General Model Results For the application of the general model, the R-squared value for the final general regression model was 0.57, with an adjusted R-squared value of 0.55, meaning that about 55 per cent of the variance in the dependent variable is explained by the final general OLS regression model, which includes 11 significant explanatory variables. Table 7.3 lists the variables used in the model, and Table 7.4 gives parameter estimation and statistical fit information for the model. From Table 7.4, some underlying factors capable of affecting regional endogenous economic growth may be specified and explained. A summary of the results follows. Stough et al. (2007) show that four sets of variables measuring resource endowments (POP_CH, BACH_00, DOCT_00), market fit (LQGOV_98, INPERS_99, INPERS_CH), institutions (LGOVEM_CH) and entrepreneurship (EM1_4_CH, EM5_9_CH) have differing levels of effect on regional endogenous growth for the US MSAs over the study period. For example, the following outcomes were identified: 1.

2.

Regional educational attainment level has different effects on regional endogenous growth. A 1 per cent increase in the percentage of population over 25 with a Bachelor Degree in 2000 is associated with a 0.88 per cent decrease in regional endogenous growth, while a 1 per cent increase in the percentage of population over 25 with a Doctoral Degree in 2000 is associated with an increase of 3.61 per cent in regional endogenous growth. Regional government plays an important role in metropolitan regional economic growth. A 1 per cent increase in the regional location quotient for government and government enterprises in 1998 is associated

STIMSON PAGINATION (M2469).indd 131

20/12/2010 15:12

132

STIMSON PAGINATION (M2469).indd 132

20/12/2010 15:12

Regional Shift (from 1999 to 2002)/employment (1999)

Dependent variable REG_SHIFT

Explanatory variables II: market fit LQMAN_98 Location Quotient for Manufacturing Industry in 1998 LQSER_98 Location Quotient for Services Industry in 1998 LQGOV_98 Location Quotient for Government and Government Enterprises in 1998 LQMAN_CH Change in Location Quotient for Manufacturing Industry from 1998 to 2002 LQSER_CH Change in Location Quotient for Services Industry from 1998 to 2002 LQGOV_CH Change in Location Quotient for Government and Government Enterprises from 1998 to 2002 ROUTW_99 Percentage of Total Occupations (all persons) as Routine Production Workers for 1999 INPERS_99 Percentage of Total Occupations (all persons) as In-person Service Workers for 1999 SYMBA_99 Percentage of Total Occupations (all persons) as Symbolic analysts for 1999

Explanatory variables I: resource endowments L_POP_99 Log (Population for all persons – 1999) POP_CH Percentage change in Population from 1999 to 2002 L_INC_02 Log Per Capita Personal Income for 2002 UNEMP_00 Unemployment Rate in 2000 (for all persons) UNEMP_CH Unemployment Rate Change from 2000 to 2002 (for all persons) BACH_00 Percentage of Population over 25 with a Bachelor Degree in 2000 EDUMP_00 Percentage of Population over 25 with a Master Degree or Professional Degree in 2000 DOCT_00 Percentage of Population over 25 with a Doctoral Degree in 2000

Definition

Definition of variables used in the US MSAs model application

Variable

Table 7.3

BLS BLS BLS

BEA BEA BEA BEA BEA BEA

BEA BEA BEA BLS BLS CENSUS CENSUS CENSUS

BEA

Source

133

STIMSON PAGINATION (M2469).indd 133

20/12/2010 15:12

Explanatory variables V: regional entrepreneurship FIRM0_98 Percentage of Self-employed Firms (do not employ any employee) in the total number of firms in 1998 EM1_4_98 Percentage of the Employment in firms with 1–4 employees in the total employment of an MSA in 1998 EM5_9_98 Percentage of the Employment in firms with 5–9 employees in the total employment of an MSA in 1998 EM10_19_98 Percentage of the Employment in firms with 10–19 employees in the total employment of an MSA in 1998 FIRM0_CH Percentage Change of Self-employed firms (do not employ any employee) in the total number of firms between 1998 and 2002

SBA

SBA

SBA

SBA

SBA

CENSUS CENSUS CENSUS CENSUS DDB

Explanatory variables IV: regional institutions LGOVEX_97 Log Local Government Expenditures in 1997 LGOVEM_97 Log Local Government Employment in 1997 LGOVEX_CH Percentage Change in Local Government Expenditures between 1997 and 2002 LGOVEM_CH Percentage Change in Local Government Employments between 1997 and 2002 SOCIAL_CAP Index of Social Capital Averaged during five years from 1993–98

BLS

BLS

BLS

FORTUNE

Percentage Change of Total Occupations (all persons) as Routine Production Workers from 1999 to 2002 Percentage Change of Total Occupations (all persons) as In-person Service Workers from 1999 to 2002 Percentage Change of Total Occupations (all persons) as Symbolic Analysts from 1999 to 2002

Explanatory variables III: regional leadership FORTU1000 Corporate Headquarters of Fortune 1000 located in an MSA, 2005

SYMBA_CH

INPERS_CH

ROUTW_CH

134

STIMSON PAGINATION (M2469).indd 134

20/12/2010 15:12

Percentage Change of the Employment in firms with 1–4 employees in the total employment of an MSA between 1998 and 2002 Percentage Change of the Employment in firms with 5–9 employees in the total employment of an MSA between 1998 and 2002 Percentage Change of the Employment in firms with 10–19 employees in the total employment of an MSA between 1998 and 2002

EM1_4_CH

Source: Stough et al. (2007).

Explanatory variables VI: population dummies POPDUM_1 Equals 1 if population<0.2 million and 0 otherwise POPDUM_2 Equals 1 if 0.2 million<=population<1 million and 0 otherwise POPDUM_3 Equals 1 if 1 million<=population<6 million and 0 otherwise

EM10_19_CH

EM5_9_CH

Definition

Variable

Table 7.3 (continued)

SBA

SBA

SBA

Source

An exploratory approach to model determinants

Table 7.4

OLS general model results for the US MSAs model application

Intercept POP_CH BACH_00 DOCT_00 LQGOV_98 INPERS_99 INPERS_CH LGOVEM_CH EM1_4_CH EM5_9_CH POPDUM_2 POPDUM_3 Source:

3.

4.

135

Coef. Value

Std. Err.

t-value

P > |t|

−3.237 0.603 −0.881 3.612 2.031 0.410 −0.323 9.236 −7.043 4.142 −9.538 −12.522

5.600 0.131 0.172 0.763 1.225 0.118 0.178 2.534 1.900 1.419 1.136 1.623

−0.58 4.61 −5.14 4.74 1.66 3.47 −1.82 3.64 −3.71 2.92 −8.40 −7.72

0.564 0.000 0.000 0.000 0.099 0.001 0.071 0.000 0.000 0.004 0.000 0.000

Stough et al. (2007).

with an increase of 2.03 per cent in regional endogenous economic growth. Meanwhile, a 1 per cent increase in local government employment is associated with an increase of 9.23 per cent in regional endogenous economic growth. The new occupational structure of routine production workers, inperson service workers and symbolic analysts created by Reich (1991) has differential effects. Compared with the percentage of total number of occupations for routine production workers, a 1 per cent increase in the percentage of the total number of occupations for in-person service workers can increase regional economic growth by 0.41 per cent. However, a 1 per cent increase in the percentage change of total number of occupations for in-person service workers decreases regional endogenous growth by 0.32 per cent. Regional entrepreneurship capital, measured by the percentage of and percentage change of the employment in small businesses in the total employment of a region, positively affects the level of regional endogenous growth. Meanwhile, the pools of small businesses of different sizes in a region may have different effects on regional economic growth. A 1 per cent increase in the percentage change of the employment in firms with 1–4 employees in the total employment is associated with a 7.04 per cent decrease in regional endogenous growth, while a 1 per cent increase in the percentage change of employment in firms with 5–9 employees in the total employment of an MSA during

STIMSON PAGINATION (M2469).indd 135

20/12/2010 15:12

136

Endogenous regional development

the same period is associated with a 4.10 per cent increase in regional endogenous growth. Specific Model Results In conducting the specific model, a number of issues were considered by Stough et al. (2007): 1.

First, since MSAs of different population sizes seem to behave differently in terms of regional endogenous growth during the period of 1999 to 2002, it is necessary to examine how the regional endogenous growth of each category of MSA segmented into population size groups (small, medium and large) is endogenously determined by regional factors and how these regional factors may differentiate one another in terms of MSAs with different population size. Thus, the original sample with 245 observations was divided into three population size categories: –

small-size MSAs with the population dummy POPDUM_1 equalling 1 (less than 0.2 million, 95 observations); – medium-size MSAs with the population dummy POPDUM_2 equalling 1 (between 0.2 and 1 million, 106 observations); – large-size MSAs with the dummy POPDUM_3 equalling 1 (between 1 and 6 million, 44 observations). 2.

3.

Variance Inflation Factors (VIFs) were examined for each sub-group and explanatory variables with VIF values over ten were excluded from each regression model. Finally, ROUTW_99 and ROUTW_CH, EM5_9_98 were excluded from the small-size MSAs model; EM5_9_98 and EDUMP_00 were excluded from the medium-size MSAs model; and EM5_9_98, L_ POP_99, SYMBA_99, LQGOV_98, DOCT_00, FIRM0_98 and INPERS_99 were excluded from the large-size MSAs model.

The specific model (backward removal stepwise approach) was then applied to each of the three sub-groups with three specific models obtained. The R-squared values for the final regression models for the small-, medium- and large-size MSAs are 0.40, 0.60 and 0.82 respectively. The adjusted R-squared values are 0.33, 0.57 and 0.77, indicating that the endogenous growth in small MSAs from 1999 to 2002 are determined more by such potential regional factors as resource endowments, market

STIMSON PAGINATION (M2469).indd 136

20/12/2010 15:12

An exploratory approach to model determinants

137

fit, leadership, institutions and entrepreneurship than those in medium and large MSAs. The results of the three final specific models of the three size categories of MSAs are not presented here, but they are both interesting, in some cases provocative and in some cases even perplexing. For example, consider the following: 1.

2.

3.

While the population factor has significant effects on the regional economic growth at all three levels of the US MSAs, the direction of that effect varies. Population change has a greater impact on the regional endogenous economic growth of medium-sized MSAs than for largesize MSAs, while small population size negatively affects the regional endogenous growth of small-size MSAs. This implies agglomeration economies may be an important element in the generation of endogenous growth, which is consistent with the literature. Educational attainment level affects regional economic growth with the highest level of education attainment (measured by the percentage of population over 25 with a Doctoral Degree in an MSA) positively affecting regional endogenous economic growth and the lower education level (measured by the percentage of population over 25 with a Bachelor Degree in an MSA) negatively affecting regional endogenous economic growth. Regional entrepreneurship capital affects regional endogenous growth of MSAs, but is manifested differently depending on the measure of entrepreneurship capital used. For small MSAs, the stock value of the entrepreneurship capital measured by the percentage of employment in firms with 1–4 employees positively affects regional economic growth, while the stock value of entrepreneurship capital measured by the percentage of employment in firms with 10–19 employees negatively affects regional economic growth. Moreover, for small MSAs the change in entrepreneurship capital also affects regional endogenous growth: the percentage change of the employment in firms with 5–9 employees in the total employment has a positive impact on regional endogenous growth over 1999 to 2002, while the percentage change of employment in firms with 10–19 employees in the total employment of an MSA between 1998 to 2002 has a negative impact on regional endogenous growth. Change of regional entrepreneurship capital negatively affects regional endogenous growth for both medium and large MSAs compared with the small MSAs.

STIMSON PAGINATION (M2469).indd 137

20/12/2010 15:12

138

Endogenous regional development

CONCLUSION In this chapter we have outlined the development – by the authors and their collaborators – of a new model framework to measure and investigate the determinants of spatial variations in regional endogenous growth performance, and in particular to identify the roles of leadership, institutional factors and entrepreneurship. As exploratory applications of the model, the sources of regional endogenous growth across the non-metropolitan regions of the five mainland states of Australia over the decade 1991 to 2001 and across the US MSAs between 1999 and 2002 were empirically modelled and then identified by a sequenced OLS regression and a stepwise regression approach. By constructing two OLS endogenous growth regression models with the regional share of the employment change for a region as the dependent variable and some variables for measuring regional resource endowments and market fit in the case of both exploratory model applications, plus the effects of variables relating to leadership, institutions and entrepreneurship characteristics of the region in the case of the latter exploratory model application, as explanatory variables, we find that some of the regional endogenous factors do play important roles in affecting the regional-share employment changes. Of course much remains to be done in the development of these operational models that seek to identify the roles of endogenous and nonendogenous factors in explaining spatial variability of levels of regional growth/decline performance across the spatial units of a space economy. That includes the need to more effectively and directly address the issue of spatial autocorrelation and incorporate weightings in the data matrix to account for spatial spillover/proximity effects. Further, the problem of directly measuring the mediating variables of entrepreneurship, leadership and institutions leaves much to be desired and begs for significant improvements in their measurement. It is interesting in the US case that metropolitan regional size is found to be an important factor in regional endogenous growth. The significance of this is that it strongly suggests the importance of agglomeration forces in contemporary regional growth and thus further contributes to the confirmation of the validity of the ‘new economic geography’. This in turn casts additional concern on the importance of the problem of stimulating and inducing economic growth in general and endogenous economic growth in rural and non-urban areas. Finally, the modelling approach might be enhanced by adopting a pathanalytic approach where the intervening/mediating effects of the leadership, institutional and entrepreneurship variables on regional endogenous growth performance are explicitly tested.

STIMSON PAGINATION (M2469).indd 138

20/12/2010 15:12

An exploratory approach to model determinants

139

NOTES *

1.

2. 3. 4.

5. 6.

7.

The authors wish to acknowledge the financial support of the Australian Research Council Linkage International Scheme (grant #LX0346785) and the George Mason University Foundation for the research project ‘Regional Economic Development and Performance: Roles of Leadership and Institutional Factors in Endogenous Growth’ from which this chapter draws. We also acknowledge the contribution to that research by Maria Salazar, Alistair Robson, Tung-Kai Shyy, Chunpu Song, Scott Jackson, Jiamin Wang and Heifung Qian. Finally, we have gained considerably from the comments of two anonymous reviewers. It is important to note that the new model framework does not negate the importance of exogenous factors and in fact recognizes these traditional sources of economic growth; rather it emphasizes the largely missed importance of such mediating factors as entrepreneurship, leadership and more generally institutions in the development process. It is important to note that OLS regression models need to be modified to incorporate the effects of spatial spillover/proximity effects and to give consideration to spatial autocorrelation as regional economic development and growth models use spatial data. The interested reader is invited to review the work by Plummer and Sheppard (2006), which illustrates both institutional and endogenous theory foundations for many of the variables suggested here. Some regions may have a high point-in-time value at the start of period of time but have little change over that period; others may have a low point-in-time value at the start but have a large change over time. The dynamic change-over-time variables are added to the analysis to capture the effect of change during a period of time. The decision to model the five mainland states separately does present a problem in that the interaction effects between regions across state borders is lost. However, in policy terms there is often a focus on state-level analysis in Australia. Each step involved withdrawing one independent variable from the model. That deleted variable had the highest probability that its absolute t-value was equal to or greater than 0.05. The model was then regressed, and new estimates of the model were obtained. This process was repeated until the variable with the highest probability that its absolute t-value was less than 0.05 was identified. The period between 1999 and 2002 was selected for the availability of data; in particular, the most recent data for some variables related to endogenous growth are only available in this period. Also, the US economy went into a recession in 2002, hence research based on this time period may partly reveal the effect and importance of endogenous growth factors when the national economy is in relative decline. However, this is likely minimal because the recession that began in 2002 was by most standards a mild one and its duration quite time limited.

REFERENCES Arthur, W.B. (1994), Increasing Returns and Path Dependency in the Economy, University of Michigan Press, Ann Arbor. Barro, R.J. (1990), ‘Endogenous Technological Change’, Journal of Political Economy, 98(5), S71–S102. Duranton, G. and Puga, D. (2000), ‘Diversity and Specialization in Cities: Why, Where and When Does it Matter?’, Urban Studies, 37(3), 533–55. Erickson, R.A. (1994), ‘Technology, Industrial Restructuring and Regional Development’, Growth and Change, 25(3), 353–79. Erickson, R.A. and Leinbach, T. (1979), ‘Characteristics of Branch Plants

STIMSON PAGINATION (M2469).indd 139

20/12/2010 15:12

140

Endogenous regional development

Attracted to Non Metropolitan Areas’, in R. Lonsdale and H.L. Seyter (eds), Non Metropolitan Industrialization, Winston, Washington, DC. Fukuyama, F. (1995), Trust: The Social Virtues and Creation of Prosperity, Free Press, New York. Garlick, S., Taylor, M. and Plummer, P. (2006), An Enterprising Approach to Regional Growth: The Role of VET in Regional Development, NCVER (National Council for Vocational Education Research), Australian National Training Authority, Adelaide. Goetz, S.J. and Rapasingla, A. (2001), ‘The Returns to Higher Education: Estimates for the Contiguous States’, paper to Regional Science Association International, North American Annual Meeting, Charleston, November. Gordon, I.R. and McCann, P. (2000), ‘Industrial Clusters, Complexes, Agglomeration and/or Social Networks, Urban Studies, 37(3), 513–32. Grossman, G.M. and Helpman, E. (1991), Innovation and Growth in the Global Economy, MIT Press, Cambridge, MA. Hanushek, E.A. and Kimko, D.D. (2000), ‘Schooling, Labour-force Quality, and the Growth of Nations’, American Economic Review, 90(5), 1184–208. Henderson, J.V., Kuncoro, A. and Turner, M. (1995), ‘Industrial Development in Cities’, Journal of Political Economy, 103(5), 1067–90. Johansson, B., Karlsson, Ch. and Stough, R.R. (2001), Theories and Endogenous Growth: Lessons for Regional Policy, Springer-Verlag, Heidelberg. Kaufman, R. (1993), ‘An Empirical Exploration of the Relation Among Diversity, Stability and Performance in Economic Systems’, Structural Change and Economic Dynamics, 4(2), 299–313. Krugman, P. (1991), Geography and Trade, MIT Press, Cambridge, MA. Lande, P. (1994), ‘Regional Industrial Structure and Economic Growth and Stability’, Journal of Regional Science, 34(3), 343–60. Lucas, R.E. (1985), Models of Business Cycles, Basil Blackwell, Oxford. Markusen, A. (1985), Profit Cycles, Oligopoly and Regional Development, MIT Press, Cambridge, MA. Norton, R.D. and Rees, J. (1979), ‘The Product Cycle and the Decentralization of North American Manufacturing’, Regional Studies, 13(2), 141–51. Plummer, P. and Sheppard, E. (2006), ‘Geography Matters: Agency, Structure and Dynamics at the Intersection of Economics and Geography’, Journal of Economic Geography, 6(5), 619–37. Porter, M.E. (1990), The Competitive Advantage of Nations, Macmillan, New York. Productivity Commission (1998), Aspects of Structural Change in Australia, Research Report, Ausinfo, Canberra. Rebelo, S. (1991), ‘Long Run Policy Analysis and Long Run Growth’, Journal of Political Economy, 98(5), S71–S102. Rees, J. (1979), ‘State Technology Programs and Industry Experience in the USA’, Review of Urban and Regional Development Studies, 3, 39–59. Rees, J. (2001), ‘Technology and Regional Development: Theory Revisited’, in B. Johansson, Ch. Karlsson and R.R. Stough (eds), Theories of Endogenous Regional Growth, Springer-Verlag, Heidelberg, pp. 94–110. Reich, R. (1991), The Work of Nations: Preparing Ourselves for 21st Century Capitalism, Vintage Books, New York. Romer, P. (1986), ‘Increasing Returns and Long Run Growth’, Journal of Political Economy, 94(5), 1002–37.

STIMSON PAGINATION (M2469).indd 140

20/12/2010 15:12

An exploratory approach to model determinants

141

Romer, P.M. (1990), ‘Endogenous Technological Change’, Journal of Political Economy, 98(5), S71–S102. Scott, A.J. (1988), New Industrial Spaces: Flexible Production Organization and Regional Development in North America and Western Europe, Pion, London. Solow, R.M. (1956), ‘A Contribution to the Theory of Economic Growth’, Quarterly Journal of Economics, 70(284), 65–94. Solow, R.M. (2000), Growth Theory: An Exposition, Oxford University Press, New York. Stimson, R.J., Robson, A. and Shyy, T.-K. (2005), ‘Modeling Determinants of Spatial Variations in Regional Endogenous Growth: Non-metropolitan Regions in the Mainland States of Australia’, 45th Congress of the European Regional Science Association, Amsterdam, August. Stimson, R.J., Robson, A. and Shyy, T.-K. (2006), ‘Modeling Regional Endogenous Growth: An Application to the Non Metropolitan Regions of Australia’, International Workshop on Creativity and Smart Policies as Signposts for Innovative Development, Amsterdam: Tinbergen Institute and Free University, May. Stimson, R.J., Robson, A. and Shyy, T.-K. (2009a), ‘Measuring Regional Endogenous Growth’, in R. Capello and P.J. Nijkamp (eds), Regional Growth and Development Theories in the XXI Century: Theoretical Achievements and Future Challenges, Edward Elgar, Cheltenham, UK and Northampton, MA, USA, pp. 354–72. Stimson, R.J., Robson, A. and Shyy, T.-K. (2009b), ‘Modeling Regional Endogenous Growth: An Application to the Non-metropolitan Regions of Australia’, Annals of Regional Science, 43(2), 379–98. Stimson, R.J., Stough, R.R. and Roberts, B.H. (2006), Regional Economic Development: Analysis and Planning Strategy, Springer, Berlin. Stimson, R., Stough, R. and Salazar, M. (2003), ‘Leadership and Institutional Factors in Endogenous Regional Economic Development’, paper presented to The North American Regional Science Association 50th Annual Meeting, Philadelphia, November. Stimson, R., Stough, R. and Salazar, M. (2005), ‘Leadership and Institutional Factors in Endogenous Regional Economic Development’, Investigaciones Regionales, 7, 23–52. Stimson, R., Robson, A.P., Stough, R. and Salazar, M. (2003), ‘Leadership, Institutions and Regional Economic Development: A New Conceptual Framework’, paper presented to The Pacific Regional Science Conference Organization, Acapulco, Mexico, July. Stough, R.R., Song, C., Wang, J. and Qian, H. (2007), ‘Modeling Endogenous Growth in U.S. Metropolitan Regions’, paper presented at the Annual Meeting of the Western Regional Science Association, Newport Beach, CA, February. Taylor, P.J., Catalano, G. and Gane, N. (2002), ‘A Geography of Global Change: Services and Cities 2000–01’, GaWC Research Bulletin, Globalisation and World Cities Study Group and Network, 77, 1–9. Thomas, M.D. (1975), ‘Growth Pole Theory, Technological Change and Regional Economic Growth’, Papers of the Regional Science Association, 34(1), 3–25.

STIMSON PAGINATION (M2469).indd 141

20/12/2010 15:12

8.

A theory of entrepreneurial rents in endogenous growth: implications for regional innovation policies Zoltan Acs and Mark Sanders

INTRODUCTION This chapter is about the role of rents and endogenous entry in economic growth – two issues we feel are not getting the attention they deserve in the modern literature on endogenous growth. When Romer (1990) presented his seminal article, his key contribution was to provide a private incentive for doing R&D. A positive knowledge spillover (in the tradition of Arrow, 1962) from current to future R&D then provided the mechanism for endogenous growth. But in his model he attributed the full rents of new entry to a specialized R&D sector that auctions off blueprints for innovations to entrepreneurs. Competition among the latter leaves no rents for them and their role is reduced to merely commercializing whatever the R&D sector comes up with. Aghion and Howitt (1992) on the other hand, put entry central in their Schumpeterian model of growth. Entrepreneurs appropriate the full rents to entry in their model, but Aghion and Howitt also have them do the R&D that generates the knowledge in a tournament. There they clearly deviate from Schumpeter who sees no role for the entrepreneur in opportunity creation (Schumpeter [1911] 1934). In that way Aghion and Howitt leave no rents to motivate R&D by firms who do not aim to enter the market with new varieties or higher-quality products. In fact they make a strong point to explain why incumbent firms do not engage in R&D. Both models are a huge improvement over earlier growth models. Yet in the data we observe that the vast bulk of rents from innovation accrue to the entrepreneur introducing the innovation in the market. And we can see incumbent firms do the vast bulk of R&D in any market economy.1 It has long been recognized that rents are what motivates an entrepreneur to enter. Schumpeter (ibid.) claimed that entry was the key source 142

STIMSON PAGINATION (M2469).indd 142

20/12/2010 15:12

A theory of entrepreneurial rents

143

of innovation. In his ‘model’ of economic growth there is no R&D and knowledge creation and entrepreneurial opportunities are falling like manna from heaven. A large body of evidence suggests that entrepreneurial firms do in fact capture rents from entry. For example, the continued entry of new firms and the going public of these entities suggest that the entrepreneurial sector captures the lion’s share of rents in the economy. This implies that these rents cannot be financing knowledge generation because entry is not generating any knowledge from commercializing ideas. In this chapter we present a model in which incumbent firms finance R&D for a clear profit motive but entrepreneurs capture the rents of commercializing the opportunities that this R&D generates through regional intra-temporal knowledge spillovers (Acs et al., 2009). In doing so we answer the question why incumbent firms do not commercialize all the knowledge their R&D generates and give entrepreneurs their proper function in generating economic growth, which is to recognize and exploit such opportunities. It is also shown that these seemingly marginal changes have some seriously different implications. Our model predicts that in a decentralized equilibrium R&D will be under-funded and there is room for welfare gains through policy. However, a delicate balance now needs to be found when R&D and entrepreneurship compete over the same resources. As more R&D implies less entrepreneurship to pick up the opportunities that R&D generates, a trade-off has to be made. The social optimum is to stimulate both R&D and entrepreneurship and invest resources in facilitating the knowledge spillover to entrepreneurs. The purpose of this chapter is to develop a model that turns the original Romer incentive structure on its head. Knowledge is generated inside the R&D labs of incumbent firms. We develop a two-sector model where incumbent firms pursue R&D investments to increase their flow of rents through process innovations leading to efficiency gains. The incentive structure for incumbent firms is such that they do not pursue new varieties for fear of cannibalizing their own rents. While incumbent firms do not fully exploit new ideas knowledge spills over to potential entrants. This is both an intra-temporal spillover and a regional activity. The rents of commercialization finance, motivate and endogenously attract the entrants. Commercialization is driven by the pursuit of rents by entrepreneurs who set up new ventures. In what follows we review the literature, outline the model and present its propositions, then draw some conclusions.

STIMSON PAGINATION (M2469).indd 143

20/12/2010 15:12

144

Endogenous regional development

REVIEWING THE LITERATURE Policy-makers are interested in promoting economic growth. This is true at the national and the regional level. And politicians look to academia to help them understand the process of economic development and inform their decisions. Academics have, long ago, identified technical change through innovation as a key process for generating long-term stable economic growth. But that begs the question – what causes innovation in a region or economy? Building on the Jaffe-Feldman-Varga model of regional knowledge spillovers, Acs and Varga (2002) suggest a ‘spatialized’ theoretical framework of technology-led economic growth that explains three fundamental issues.2 1.

First, it provides an explanation of why knowledge-related economic activities may concentrate in certain regions, leaving others relatively underdeveloped. Second, it answers the questions of how technological advance occurs and what are the key processes and institutions involved, with a particular focus on the geographic dimension. Third, it presents an analytical framework where the role of technological change in regional and national economic growth is clearly explained.

2.

3.

In order to explore these three issues Acs and Varga (ibid.) examine three separate and distinct literatures: ● ● ●

the new economic geography; the new growth theory; the new economics of innovation.

The three literatures focus on different aspects but at the same time are also complements of each other. The ‘new’ theories of growth endogenize technological change and as such link knowledge creation and spillovers to technological change and macroeconomic growth at the aggregate level. New growth theory has identified knowledge generation (Romer, 1986, 1990; Aghion and Howitt, 1992) through R&D and human capital accumulation (Lucas, 1988) as the ultimate sources of economic growth. The empirical evidence in support of this view is mounting (Temple, 1999). The evidence clearly shows that more knowledge generation through R&D and human capital accumulation is correlated with economic performance, both across countries and regions and over time, and there is hardly

STIMSON PAGINATION (M2469).indd 144

20/12/2010 15:12

A theory of entrepreneurial rents

145

any region that boasts periods of strong economic performance without strong knowledge generation. But the opposite is not true. There are several countries and regions in the world where R&D expenditures and education are high by any standard and have been so for quite some time and yet economic performance is low. The exemplary case here is Sweden, which is leading the charts in educational attainment and quality as well as in R&D expenditures per capita. This is one piece of empirical evidence that does not fit well into the endogenous growth framework and has been referred to as the Swedish or European Paradox. After over a decade of poor economic performance, Japan may now also be put in the list of countries where knowledge creation clearly turned out to be no guarantee for growth. Therefore, despite the great improvement that new growth models represent over the earlier neoclassical growth model, new growth theory fails to capture the full complexity of the innovation process. As we will argue below, the common assumption that knowledge creation and commercialization are effectively a simultaneous process is particularly harmful to this line of research. Relaxing that assumption may help explain the above-mentioned European Paradox as well as provide useful points of entry for national and regional innovation policy. In contrast to new growth theory, the ‘systems of innovation’ frameworks are very detailed with respect to the different stages and agents involved in the innovation process. The idea behind the innovation systems approach is quite simple but extremely appealing. According to this approach, innovation is a result of a complex process but this process is shaped in a systemic manner. The elements of the system are innovating firms and firms in related and connected industries (suppliers, buyers), private and public research laboratories, universities, supporting business services (like legal or technical services), financial institutions (especially venture capital) and the government. These elements are interconnected by innovation-related linkages where these linkages represent knowledge flows among them. Linkages can be informal in nature (occasional meetings in conferences, social events etc.) or they can also be definitely formal (contracted research, collaborative product development etc.). The effectiveness (i.e., productivity in terms of number of innovations) of the system is determined by both the knowledge already accumulated by the actors and the level of their interconnectedness (i.e., the intensity of knowledge flows). But behavior in these frameworks is typically assumed to be predetermined and mechanical. The ability and motivations for interactions are shaped largely by traditions, social norms, values and the countries’ legal systems and do not respond much to economic incentives and/or such behavioral responses are modeled as simple heuristics rather

STIMSON PAGINATION (M2469).indd 145

20/12/2010 15:12

146

Endogenous regional development

than rational behavior. Moreover, the systems-of-innovation frameworks have little to say about aggregate innovation rates and macroeconomic growth. Both new growth theory and the systems of innovation approach typically also ignore the spatial dimension, although it arguably has been introduced into the systems of innovation frameworks in recently developed ‘regional innovation systems’ studies (Braczyk et al., 1998). Finally, new economic geography models investigate general equilibrium in a spatial setting (Krugman, 1991). This means that they provide explanations not only for the determination of equilibrium prices, incomes and quantities in each market but also the development of the particular geographical structure of the economy. In other words, new economic geography derives economic and spatial equilibrium simultaneously (Fujita et al., 1999; Fujita and Thisse, 2002). Spatial equilibrium arises as an outcome of the balance between centripetal forces working towards agglomeration (such as increasing returns to scale, industrial demand and localized knowledge spillovers) and centrifugal forces promoting dispersion (such as transportation costs). Until recently, new economic geography models did not consider the spatial aspects of economic growth. However, even the recent models of technological change follow the pattern as endogenous growth models and abstract from the complexity inherent in innovation systems studies. As emphasized by Acs and Varga (2002), each one of the above three approaches has its strengths and weaknesses but they could serve to create the building blocks of an explanatory framework of technology-led economic growth at the regional level. In this chapter we outline a model in the endogenous growth modeling tradition (in fact, an adaptation of the Romer, 1990 model) that adds an essential element in the complexity of the innovative process by separating the decision to create knowledge from the decision to commercialize an opportunity. This is what we feel provides the missing link between the three literatures described above: entrepreneurship. We define entrepreneurship as the act of commercializing new knowledge.3 And we show that separating knowledge creation from knowledge commercialization implies that knowledge creation alone is a necessary but not a sufficient condition to generate growth. Our model gets a strong regional flavor when we realize that the knowledge spillover from creators to commercializers necessarily involves intense communication, as a lot of the knowledge is new, uncodifiable and tacit. The costs of communicating such knowledge over large geographical distances are large. In our model the knowledge spillover structure implies that entrepreneurs will localize in regions with high levels of R&D and knowledge creation in related industries. The vast empirical literature on

STIMSON PAGINATION (M2469).indd 146

20/12/2010 15:12

A theory of entrepreneurial rents

147

geographical clusters can be read to support the notion that such knowledge spillovers are generally localized. The combination of these two adaptations to the original Romer model implies that regions with strong knowledge creation will experience high growth only when there are enough entrepreneurial resources around to get the new knowledge to the market. Our model then generates a testable hypothesis. Growth should arise in those regions and countries that have strong knowledge creation and entrepreneurship, whereas knowledge creation without entrepreneurship and entrepreneurship without knowledge creation generate less growth for the same levels of effort. Our model predicts that both R&D and entrepreneurial activity will positively affect growth in a region, but the marginal impact of either depends strongly on the presence of the other. Space and time do not permit us to test our hypotheses empirically in this chapter, but the evidence in the literature is suggestive. The empirical evidence on the importance of entrepreneurship in explaining patterns of regional growth and development has recently been surveyed in Versloot and van Praag (2007) and Acs and Varga (2005) present results for European country data and also find a positive and statistically significant interaction effect between R&D and entrepreneurship. Audretsch and Keilbach (2005) ran a test on German regions showing that indeed a positive interaction exists at the regional level. Based on the above-presented empirical results in the literature, however, we feel it is worthwhile to dwell on the policy implications of our analysis. There is a case to be made for R&D subsidies, as knowledge generation has positive externalities, but given the current set of innovation policies in place in most countries and regions, policy-makers should carefully identify the bottleneck in the innovation chain from the original knowledge creation to the final marketing of new products and services before committing more public resources to the generation of more new knowledge. To assume that new knowledge will create economic growth automatically and costlessly is similar to dropping seeds on rock. Only a fraction of the thus developed valuable knowledge will actually take root. In some regions more knowledge investments may be needed, but we would argue that in most regions more entrepreneurial resources would improve the effectiveness of existing knowledge creation and generate more economic growth for the tax dollar. We now proceed with a general outline of a growth model, where we zoom in on the innovative process and present the formal economic decision problems for knowledge creators and commercializers. From that analysis we can derive our main propositions in an a-spatial context. After that we discuss the implications of taking into account spatial considerations in the model.

STIMSON PAGINATION (M2469).indd 147

20/12/2010 15:12

148

Endogenous regional development

OUTLINING A MODEL OF ENTREPRENEURIAL RENTS AND GROWTH Consider a three-sector, two-factor economy in which consumers consume, save to accumulate raw capital and supply their labor exogenously. Final goods producers produce consumption goods, using labor, intermediate capital goods and invest in R&D (labor) to improve their productivity and reduce production costs. And finally, intermediate goods producers supply them with an expanding variety of intermediate goods that are produced with raw capital, obtained in capital markets and introducing a new variety requires the input of labor resources. The financial flows in our model can then be illustrated in Figure 8.1, where the arrows represent real money flows in terms of the final good that is the numeraire. Below we introduce our notation and give the exact definition of the arrows. Then we shortly discuss the actors and discuss what problem they solve under what constraints. Finally we discuss how the markets in the model equilibrate. Consumers With numbers referring to the arrows in the Figure 8.1, consumers have two outgoing and two incoming flows: (1) (2)

consumption of C (at price P=1); savings Y−C, which are invested in bonds, B, yielding interest rate r;

Consumers of final good C

(1)

(4)

Producers of final good C (5)

Labor Market

(8)

(9)

(6)

(7) (13)

(3)

(10) Capital Market

Producers of n intermediate goods

Entry

(11)

(2) (12)

Source: The authors.

Figure 8.1

Financial flows in the model

STIMSON PAGINATION (M2469).indd 148

20/12/2010 15:12

A theory of entrepreneurial rents

(3) (4)

149

interest income rB; labor income wL* where total labor is supplied inelastically and normalized to 1.

Consumers in our model receive interest and labor income every period and spend their income on consumption and the purchase of new bonds. They maximize a standard log-linear utility function and face a standard budget constraint: max :U 5 Ct

e0`log [ Ct ] dt

# s.t.: Bt 5 wtL* 1 rtBt 2 Ct where U is the utility index and a dot over the variable signifies a time derivative. Consumers then choose consumption (and implicitly savings) in every period following the Ramsey rule, such that a constant fraction of income is saved when the interest rate is constant and exceeds the rate of time preference: # Ct/Ct 5 rt 2 r Final Goods Producers Total consumption equals the sales and production of final goods producers as we assume the market for final goods clears. The next four arrows in Figure 8.1 then relate to the behavior of final goods producers, who are price takers in factor and output markets: (5) (6) (7) (8) (9)

production wages wLP where LP is labor employed in production; n costs of n intermediate goods e0 c (i) x (i) di bought at price c(i) and in quantity x(i); R&D wages wLR where LR is labor employed in R&D; investment in R&D (equal to labor costs) financed by issuing new bonds; interest payments on issued bonds;

For the final goods producers to have an incentive to do R&D we introduce the firm-specific factor ‘knowledge’, A, into their production function and specify the process by which they can increase that knowledge stock. Firms then choose the optimal levels of production and R&D employment and their use of intermediate goods at every point in time. The knowledge stock makes their problem an inter-temporal one that is given by:

STIMSON PAGINATION (M2469).indd 149

20/12/2010 15:12

150

Endogenous regional development `

`

n

max : 3 e2rtP 5 3 e2rt aC 2 w (LP 1 LR) 2 3 c (i) x (i) dib

LP, x (i) , LR

0

0

0 n

s.t.: C 5 Y 5 AaLbP 3 x (i) 1 2a 2b 0

# A 5 yAgn1 2gLR

where we have dropped the time arguments to save on notation and y is a scaling parameter. Note that we have assumed that R&D in the final goods sector receives a positive knowledge spillover from past R&D, A, and the variety in intermediates, n. This reflects the assumption that it is easier to do R&D from an already large knowledge base and it is easier to increase productivity in the final goods sector when a lot of different specialized intermediates are available. By the assumed symmetry and constant returns to scale specification we can study the behavior of a representative firm and solve the above problem.4 The demand for production labor and individual intermediate variety i are given by: LDP 5

bY wP

21

c (i) a 1b xD (i) 5

n

a c (i)

1 2a 2b 2 a 1b

(1 2 a 2 b) Y

i50

And the final goods sector will employ R&D workers as long as the wage is below:5 wR 5

aYy (A/n) 2g (r 2 w# R /wR 1 gn# /n)

Intermediate Goods Producers (10) (11)

rental costs of the raw capital used in producing intermediate goods and financed with bonds, rK; dividends on ownership shares and/or interest on loans equal to the n expected value of rents E0 [ e0 p (i) di ] at entry, to finance start-up investments that equal the wages (forgone) by the entrepreneur.

STIMSON PAGINATION (M2469).indd 150

20/12/2010 15:12

A theory of entrepreneurial rents

(12) (13)

151

investment in entry (equal to labor costs) financed by issuing stocks and/or bonds by entrants; labor costs of entry in intermediate sector financed by issuing stock or bonds, wLE.

Intermediate producers are assumed to be monopolists and set prices to maximize their profits. A simple production technology that converts one unit of raw capital into a unit of the intermediate variety, completes the problem for the intermediate producer: max: p (i) 5 c (i) x (i) 2 rK (i) c (i )

s.t.: x (i) 5 K (i) 5 xD (i) The producer will set his price as a mark-up over marginal costs: c (i) 5

r 12a2b

And as marginal costs are equal for all varieties, all varieties are priced and consequently used at the same level. Given that producing a new variety yields positive profits, new entrants have an incentive to commercialize ideas for new varieties. We assume that this commercialization process is costly in terms of labor and specify the entry process as: # n 5 ␾ALE where we have assumed that new variety creation is proportional to the stock of accumulated R&D knowledge in final goods production. This reflects our assumption that R&D in final goods production generates a lot of knowledge spillovers in the form of new ideas and opportunities for intermediate goods. Below we will discuss why such spillovers are generally local and will create the emergence of clusters of related economic activity. In the model we have no geography and entry takes place as long as the wage is below: wE 5

(a 1 b) (1 2 a 2 b) ␾ (A/n) Y # # r 1 n/n 2 Y/Y

To finance the labor costs of entry, new entrants issue stock or bonds and in equilibrium the profits from production are exactly equal to the interest payments on bonds plus the returns on stock.

STIMSON PAGINATION (M2469).indd 151

20/12/2010 15:12

152

Endogenous regional development

Equilibrium and Propositions The model equilibrates when all flows into and out of the boxes add up to 0 (which means agents solve their maximization problems given their constraints and do not leave any resources idle) and prices equilibrate the supply and demand on the two factor markets. It can be shown that: Proposition 1: In equilibrium, growth in income can only come from productivity gains in final goods production and variety expansion in intermediates. Proposition 1A: Optimal growth requires the stimulation of both R&D and entrepreneurship. Proposition 1B: The positive external effect that would justify a subsidy on R&D, only materializes when entrepreneurs are present and able to commercialize that knowledge, so any policy that helps R&D but hurts entrepreneurship is therefore counterproductive.6 Proposition 1C: The model predicts that R&D and entrepreneurship both contribute to economic growth. The latter mainly operates through a positive interaction effect between downstream R&D and upstream product innovation through entry. The former operates through a direct effect of R&D on process innovation. Without going through the mathematical details we can derive these main propositions by considering first the motives for innovative activity of entrepreneurs and R&D workers in the model. Recall that intermediate producers use a simple one-for-one technology to create their intermediates from raw, homogenous capital. As in Romer (1990) the Constant Elasticity of Substitution (CES) production function at the final goods production stage implies that the intermediate varieties are imperfect substitutes in final goods production and thus a latent demand for all new varieties exists. The monopolists in the intermediate sector earn monopoly rents, creating an incentive for entry. Patent protection on existing intermediates might be assumed to prevent entry in the existing intermediate markets and leave entry with a new intermediate variety as the only alternative.7 Instead, one might also assume that the entrepreneur has and retains exclusive knowledge regarding his venture and competitors can never enter with perfect substitutes and drive profits to 0. 8 This implies that entry can only take place with new varieties that are imperfect substitutes. Romer (1990) then assumes that a specialized R&D sector generates the blueprints for a new intermediate good and auctions them off to a competitive fringe of potential entrants. The downstream rents thus motivate and finance an R&D sector that generates new ideas. The only barrier

STIMSON PAGINATION (M2469).indd 152

20/12/2010 15:12

A theory of entrepreneurial rents

153

to entry in Romer is the possession of a blueprint and therefore the R&D firm, the knowledge creator, appropriates the full discounted rents in equilibrium. In Romer the entrepreneurial opportunity is created as a private property and commercialization is costless and automatic. We have assumed instead that new firm entry is costly and risky and we follow Schumpeter ([1911] 1934)) in assuming that entrepreneurial opportunities are pure and costless spillovers. Therefore, the associated rents are fully appropriated by the entrepreneur, leaving the one that commercializes knowledge as the residual claimant to the monopoly rents.9 It implies that the rents from commercialization are no longer available to finance knowledge creation and no independent R&D sector as in Romer (1990) can exist. However, knowledge needs to be created somewhere and for a clear economic purpose if we wish to avoid a return to the neoclassical ‘manna from heaven’ growth models. Large amounts of investment in corporate R&D also suggest that knowledge creation is somehow profitable to the firms undertaking it. We would argue, however, that the improvement and more efficient production of existing products is the stated aim of the corporate R&D labs we see in the world today. Not the generation and subsequent auctioning off of blueprints for new (intermediate) goods as in Romer (1990). To make the generation of knowledge profitable to final goods producers in our model, they cannot operate under perfect competition with constant returns to scale in intermediates and labor as in Romer (1990). We assumed instead that the production function has constant returns to three factors of production: labor; an aggregate of intermediates, that is, capital; and a stock of private production knowledge. Price taking on the demand side in labor and intermediate markets then implies that all firms have operating profits as wage and intermediates costs do not exhaust sales. This profit is the return to the firm-specific knowledge stock. We assumed that it needs to be accumulated prior to production so a new entrant must first accumulate one for himself. Free entry will therefore not eliminate the operating profits. The stock of production knowledge can be augmented every period by doing R&D. Profit-maximizing firms then choose a level of R&D labor that equates the discounted future value of additional operating profits to the marginal wage costs of their R&D workers. It can be shown that in equilibrium all firms will have the same level of production knowledge and R&D investment.10 Our structure makes the intended outcome of R&D, efficiency gains to the firm, a pure private good of which the returns can be fully appropriated.11 However, we also assumed that the R&D generates an accidental by-product; knowledge and opportunities that the final goods-producing firm does not find

STIMSON PAGINATION (M2469).indd 153

20/12/2010 15:12

154

Endogenous regional development

profitable to commercialize. And we assume that firms cannot prevent the spillover of that knowledge.12 The ideas for new intermediate goods are therefore a costless knowledge spillover from incumbent firms’ R&D. Ongoing R&D in incumbent firms generates a flow of ideas, some of which are shelved by the incumbents for whatever reasons but then commercialized through new firm entry. The markets for labor and capital connect all activities and close our model. So far we have ignored the spatial dimension of our analysis and provide a space-free aggregate macroeconomic model of innovationdriven growth. Starting from this basic framework one can, however, also develop the proposition that innovative activities will tend to cluster spatially. To do so we consider the introduction of limited labor mobility, transport costs and communication costs in our framework.

REGIONAL KNOWLEDGE SPILLOVERS In our model, R&D generates an externality in the form of knowledge spillovers to the entrepreneurial sector. For example, Acs and Varga (2005) found empirical support for the claim that knowledge spillovers between knowledge creators and knowledge commercializers are geographically localized. There are several ways in which our model might be amended to explain that result. A first and obvious one is the mobility of labor. If labor is more mobile between R&D and entrepreneurship than it is among regions, then the knowledge an individual has developed in the R&D lab presents an opportunity to leave the lab and start an intermediate firm. When geographic labor mobility is low it is likely that such a venture is set up in the same region. The mechanisms underlying this channel for knowledge spillovers, the phenomena of spin-off and spin-out, have been studied intensively by Klepper (2006), for example. He has found this conduit for knowledge transfer to be highly relevant in, for example, the early automobile industry and the tire industry around Detroit. We formulate Proposition 2: Proposition 2: If we assume that knowledge spillovers require labor mobility between entrepreneurship and R&D, we can derive that knowledge spillovers between R&D labs and entrants operate in particular within regions with high labor mobility and are less important over longer geographical distances. Hence innovative activities will cluster geographically. Second, we may introduce an agglomeration effect in the knowledge commercialization function. In the presence of related activities, a region’s

STIMSON PAGINATION (M2469).indd 154

20/12/2010 15:12

A theory of entrepreneurial rents

155

infrastructure is more likely to be adapted to the needs of a particular new venture. It is also reasonable to assume that a new intermediate firm is more likely to succeed when it is located in close geographical proximity to its customers. This reduces transport costs and the costs of communicating with the customer. In our model the customer is the final goods producer that was the origin of the opportunity. This centripetal force is, of course, stronger for more specialized intermediates that demand a more specialized and expensive infrastructure and for higher transport and transaction costs. Proposition 3: If we assume that infrastructural, transport and transaction costs are large and increasing in the geographical distance between intermediate firms and final goods producers, then intermediate firms will locate close to each other and the final goods producer that generated the opportunities on which the new ventures are founded. Hence innovative activities will cluster geographically. Third, we might introduce a parameter in our model to reflect the costs of communicating the knowledge between creators and commercializers. This cost parameter would reflect language barriers, cultural differences, lack of appropriate expertise to appraise the opportunity and so on, but would also capture institutional barriers to knowledge spillover. Geographical proximity and agglomeration variables will then proxy in empirical specifications for the importance of such knowledge transfer costs, as, for example, in Varga (2006). This extension to the model, as opposed to the ones above, also explains political border effects. Borders between countries, ethnic communities, cultural identities but also scientific disciplines will reduce the spillover and hence the introduction of such communication costs explains the agglomeration and clustering of innovative activities within countries, jurisdictions and cultural communities. Proposition 4: If we assume that communication is costly and increasingly so when we cross cultural boundaries, political borders, legal jurisdictions, language barriers, scientific disciplines and common knowledge pools, then the knowledge spillover theory of entrepreneurship predicts the clustering of innovative activities. To the extent that these socio-cultural distances are proxied by geographical distance, the model predicts geographical clustering of innovative activities. The three propositions presented in this section all predict geographical clustering of innovative activity but identify very different channels through which knowledge spillovers would cause that clustering. And this

STIMSON PAGINATION (M2469).indd 155

20/12/2010 15:12

156

Endogenous regional development

list of channels is by no means claimed to be exhaustive. Understanding the (relative) importance of the various channels is essential in formulating a suitable innovation policy at the aggregate, the regional and the local level.

POLICY IMPLICATIONS AND CONCLUDING REMARKS Our model formalized the ‘knowledge spillover theory of entrepreneurship’ (Acs et al., 2009) and placed the entrepreneur as the ‘missing link between knowledge creation and economic growth’ (Branuerhjelm et al., 2010) in an integrated general equilibrium framework and combined modeling techniques developed in the growth literature with insights from the more descriptive and empirical literature on innovation and entrepreneurship. Moreover, in accordance with the evidence, in our model the entrepreneur is the agent with whom the buck stops. The creation of the knowledge he commercializes is not (necessarily) motivated by the rents that the entrepreneur receives for commercialization. Such rents reward the act of commercialization and as such should not be destroyed to enhance static efficiency (a result we share with all innovation-driven growth models). But the claim to these rents should also not be transferred to the generators of knowledge that may have had no intent of commercializing and/ or require no incentive to create such knowledge in the first place.13 Not the generation but the implementation of new knowledge is what is valuable to society at large and what generates economic growth. Knowledge creation, of course, is a necessary but insufficient condition for innovation and growth, and creation without implementation is clearly a waste of resources. We argue, therefore, that policy-makers should stop and think about where the bottlenecks in the innovative process are, before committing large amounts of public money and/or entitlements to profits and rent to the (formal) knowledge generation process. These results also carry over to the regional level if we consider the impact of limited geographical labor mobility, transport costs and communication costs. As the knowledge spillovers that drive economic growth are likely to be regionalized, regional policies should aim to smoothen the spillover. A first requirement is that sufficient resources are available for both knowledge creation and knowledge commercialization. And as entrepreneurial talent is a key resource in the innovation chain, regional policies should aim to develop it. Moreover, the impediments to knowledge spillovers from creators to commercializers deserve attention. Legal

STIMSON PAGINATION (M2469).indd 156

20/12/2010 15:12

A theory of entrepreneurial rents

157

impediments such as non-competition clauses in labor contracts can easily be abandoned. But by investing in physical and communication infrastructures and by stimulating or enabling the exchange of knowledge, local and regional governments can support the entire innovation chain. Direct support to new entrants or R&D should be given only as long as that does not reduce the incentives to create or commercialize new knowledge. The model outlined in this chapter has important policy implications at the aggregate and regional level but also raises important questions to be addressed. The presence of knowledge spillovers is well documented in the literature. But the exact channels through which such spillovers arise are a challenging area for further research. Our three propositions predicting regional clustering are empirically indistinguishable in most studies, due to data availability issues. The detailed case studies by Klepper (1996) provide support for the first channel that was identified. The work by Florida (2002a and 2002b) presents evidence in support of the third channel we have discussed, but to our knowledge studies that try to distinguish between them have not yet been done. Also at the aggregate level, our theory and its underlying assumptions require further empirical scrutiny. We feel the available evidence supports our claim that knowledge spillovers are important for (regional) economic growth but a lot more can be done to test the model predictions. This empirical research agenda will hopefully inspire other researchers to join us in pursuing it.

NOTES 1. 2. 3 4. 5 6. 7.

8.

The part that is not done by incumbent firms is done by publicly funded institutions such as universities or specialized R&D firms as Romer (1990) envisioned them. See Acs (2009) for a detailed and referenced exposition of this framework. There are many definitions of entrepreneurship in the literature. We follow the Schumpeterian tradition. See, for example, Braunerhjelm (2008) for a recent overview of this literature. Even if the knowledge stock is allowed to differ among final goods producers it can be shown that only those that have A=Amax will employ R&D workers and increase their A such that the firms with lower knowledge stocks will diminish. There is a horizontal demand curve for R&D labor due to the assumed linearity in R&D labor in the innovation function. See Acs and Sanders (2008) for an application to the role of patent protection in this context. So far we copied the Romer (1990) structure. In Aghion and Howitt (1992) these entrants drive out incumbents with a higher-quality version of existing varieties and entry leads to average-quality improvement not to variety expansion. To keep the model manageable we follow Romer (1990) here and focus on variety expansion but we also feel that entrepreneurial entry may introduce improved versions of existing products. Patent protection is problematic in this model as we assume that the knowledge creator is not the same agent as the knowledge commercializer. Patents are generally awarded

STIMSON PAGINATION (M2469).indd 157

20/12/2010 15:12

158

9.

10. 11. 12. 13.

Endogenous regional development to the knowledge creator. There is a large literature (Acs, 2008) that stresses the importance of the individual entrepreneur for the success of new ventures. His unique combination of cultural background, skills, knowledge, access to finance and other key resources and not least important, luck, makes it unlikely that any other entrant could enter the same market and drive down profits to 0 by copying the incumbent. Although in expectation terms the profits flow back to the consumers in Figure 8.1, the entrepreneur, more often than not, is that specific consumer and the profits are his expected returns on forgoing labor earnings during the entry stage. We have modeled this as the entrepreneur issuing stock and bonds to finance his wage costs to reflect the fact that they take such opportunity costs into account and desire a market-determined return on their investment. This follows intuitively from the assumption that all final goods-producing firms are equal, face the same maximization problem, production possibilities, final demand curve and set of input prices. We discuss the precise set of assumptions we need to make for this result in Acs and Sanders (2008). And if there are such impediments, that would be a first target for policy. Braunerhjelm et al. (2010) refer to the knowledge filter when they discuss the physical, cultural, political and institutional barriers to such knowledge spillovers. As, we have argued in Acs and Sanders (2008), may well be the effect of stronger patent and IPR protection.

REFERENCES Acs, Z.J. (2008), ‘Foundations of High Impact Entrepreneurship’, Foundations and Trends, 4(6), 535–620. Acs, Z. (2009), ‘Jaffe-Feldman-Varga: The Search for Knowledge Spillovers’, in A. Varga (ed.), Universities, Knowledge Transfer and Regional Development: Geography, Entrepreneurship and Policy, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Acs, Z.J. and Varga, A. (2002), ‘Geography, Endogenous Growth and Innovation’, International Regional Science Review, 25(1), 132–48. Acs, Z.J. and Varga, A. (2005), ‘Entrepreneurship, Agglomeration and Technological Change’, Small Business Economics, 24(3), 323–34. Acs, Z.J. and Sanders, M. (2008), ‘Intellectual Property Rights and the Knowledge Spillover Theory of Entrepreneurship’, Working Papers No. 08-23, Utrecht School of Economics. Acs, Z.J., Braunerhjelm, P., Audretsch, D. and Carlsson, B. (2009), ‘The Knowledge Spillover Theory of Entrepreneurship’, Small Business Economics, 32(1), 15–30. Aghion, P. and Howitt, P. (1992), ‘A Model of Growth through Creative Destruction’, Econometrica, 60(2), 323–51. Arrow, K.J. (1962), ‘Economic Welfare and the Allocation of Resources for Invention’, in R.R. Nelson (ed.), The Rate and Direction of Inventive Activity, Princeton, NJ: Princeton University Press, pp. 609–26. Audretsch, D.B. and Keilbach, H. (2005), ‘Entrepreneurship capital and regional growth’, The Annals of Regional Science, 39(1), 457–69. Audretsch, D. and Lehmann, E. (2006), ‘Does the Knowledge Spillover Theory of Entrepreneurship hold for Regions’, Research Policy, 34(8), 1191–2002. Braczyk, H., Cooke, P. and Heidenreich, M. (1998), Regional Innovation Systems: The Role of Governances in a Globalized World, London: UCL Press.

STIMSON PAGINATION (M2469).indd 158

20/12/2010 15:12

A theory of entrepreneurial rents

159

Braunerhjelm, P. (2008), ‘Entrepreneurship, Knowledge, and Economic Growth’, Foundations and Trends® in Entrepreneurship, 4(5), 451–533. Braunerhjelm, P., Acs, Z.J., Audretsch, D.B. and Carlsson, B. (2010), ‘The Missing Link: Knowledge Diffusion and Entrepreneurship in Endogenous Growth’, Small Business Economics, 34(2), 105–25. Florida, R. (2002a), ‘Bohemia and Economic Geography’, Journal of Economic Geograhy, 2(1), 55–71. Florida, R. (2002b), ‘The Economic Geography of Talent’, Annals of the Association of American Geographers, 92(4), 743–55. Fujita, M. and Thisse, J. (2002), Economics of Agglomeration. Cities, Industrial Location, and Regional Growth, Cambridge, MA and London: Cambridge University Press. Fujita, M., Krugman, P. and Venables, A. (1999), The Spatial Economy, Cambridge, MA: MIT Press. Klepper, S. (1996), ‘Entry, Exit, Growth and Innovation over the Product Life Cycle’, American Economic Review, 86(3), 562–83. Krugman, P. (1991), ‘Increasing Returns and Economic Geography’, Journal of Political Economy, 99(3), 483–99. Lucas, R. (1988). ‘On the Mechanics of Economic Development, Journal of Monetary Economics, 22(1), 3–42. Romer, P. (1986), ‘Increasing Returns and Economic Growth’, Journal of Political Economy, 94(5), 1002–37. Romer, P. (1990), ‘Endogenous Technological Change’, Journal of Political Economy, 98(5), S71–S102. Schumpeter, J.A. [1911] (1934), The Theory of Economic Development, Cambridge, MA: Harvard University Press. Temple, J. (1999), ‘The New Growth Evidence’, Journal of Economic Literature, 37(1), 112–56. Varga, A. (2006), ‘The Spatial Dimension of Innovation and Growth: Empirical Research Methodology and Policy Analysis’, European Planning Studies, 14(9), 1171–86. Versloot, P.H. and Van Praag, C.M. (2007), ‘What is the value of entrepreneurship? A review of recent research’, Small Business Economics, 29(4), 351–82.

STIMSON PAGINATION (M2469).indd 159

20/12/2010 15:12

9.

Foreign direct investment, knowledge assets and the economic geography of growth in the Asian BRIICS countries Tomokazu Arita, Chie Iguchi and Philip McCann

INTRODUCTION In the context of globalization, our objective in this chapter is to highlight the role foreign direct investment plays in the growth of three of the world’s largest newly-industrializing countries, and to examine the links between these investment inflows and economic geography. The three countries concerned are China, India and Indonesia. The major features of the current phase of globalization that are common to all regions of the world are: ● ● ●

increasing institutional integration between countries; rapidly improving information and communication technologies; rapidly increasing global investment activities (McCann, 2008).

The combination of these features produces complex interrelationships between economic geography, economies of scale and global trade and investment. These interrelationships have been discussed in detail elsewhere (McCann, 2008, 2009; World Bank, 2009) so are not discussed further here. However, one of the other major features of the current era of globalization is the role played by the expansion of the global labour force. This rapid expansion is dominated by six major newly-industrializing economies, the so-called BRIICS countries, comprising Brazil, Russia, India, Indonesia, China and South Africa (McCann, 2009). The opening-up of these countries between 1989 and 1993 transformed the global labour market in that they brought almost 1.5 billion new workers into the global labour force. 160

STIMSON PAGINATION (M2469).indd 160

20/12/2010 15:12

Growth in the Asian BRIICS countries

161

Asia is often seen as the focus of the driving forces of this period of globalization. The reason is that the three largest of these enormous newly industrializing countries, namely China, India and Indonesia, are within the South and East Asian economic arena. However, the performance of these different countries in terms of global and regional output and the role played by these economies is a complex picture. Figure 9.1 helps us to visualize these differences. In order to understand the relative scale effects of the Asian economies, we also include the two Australasian economies of Australia and New Zealand as comparator cases. If we consider the growth performance of these Asian countries during the current phase of globalization, as we see from Figure 9.1 it is clear that China and India are currently the two fastest-growing large economies in the world, with 2005 growth rates of 10.2 per cent and 9.2 per cent respectively. China’s gross national income in 2005 was US$2269.7 billion, which ranked it as the world’s fifth largest economy in 2005, and almost identical in size to the UK economy; India’s economy in 2005 was some US$804.4bn, and ranked as the tenth largest economy in the world, while Indonesia, with an economy of some US$282.2bn in 2005 is ranked at 23rd. Yet scale and affluence are different things. If we consider per capita income rather than the gross income, China, with a per capita income of US$1740 is ranked 128th in the world, Indonesia with a per capita income of US$1280 is ranked 139th, while India with a per capita income of US$730 is ranked 158th in the world. As such, the potential for growth in each of these countries is still enormous. Economic growth in each of each of these countries has been associated with deregulation and liberalization, and the massive internal reallocation of production factors between sectors and regions. However, the link between these internal growth processes and the broader processes of globalization that are taking place is via the enormous inflows of foreign direct investment (FDI) from the advanced economies into these countries. Yet, inward FDI exhibits different levels of relative importance in different host economies. Amongst developing countries, the transnationality index of openness, which indicates the scale of inward multinational investment in terms of FDI inflows, stocks, value-added and employment, relative to total GDP, ranks China 32nd, India 36th, and Indonesia 38th. In general, across all developing economies, the overall relative trans-nationality openness of countries to FDI tends to be higher in small countries and lower in the larger economies, although Indonesia is relatively more closed than its scale might suggest, while China is relatively more open than its scale would suggest. FDI will, however, continue to flow in very large quantities from developed countries into these countries over the foreseeable future. Recent

STIMSON PAGINATION (M2469).indd 161

20/12/2010 15:12

162

Endogenous regional development

Source: The authors.

Figure 9.1

The size, wealth and growth of the Asian and Australasian economies in 2006

STIMSON PAGINATION (M2469).indd 162

20/12/2010 15:12

Growth in the Asian BRIICS countries

163

UNCTAD survey evidence of multinational executives regarding the most attractive locations for FDI over the coming years, finds that China is ranked as the number 1 country in the world and India is ranked number 2. The investment inflows resulting from these perceptions will ensure that these newly-industrializing countries will become increasingly open and integrated into the global economic system. Yet, although the relative levels of FDI currently appear to be small in these countries, multinational inward investment plays an extremely important role in the economic growth of these countries. In particular, as this chapter will demonstrate, inward investment plays a crucial role in determining both the technological and geographical distribution of activities in each of these countries during the current stages of economic growth. The rest of the chapter is organized as follows. In the next section we discuss the nature and scale of inward foreign direct investment in these three countries, with a particular focus on the rise of knowledge-related investments. In what follows we examine the spatial patterns of inward investment in these countries and we relate these emerging patterns to the economic geography of each country, and in particular, to the role of agglomeration economies in each society. The final section provides some brief conclusions. Unless otherwise stated, the empirical evidence in the following sections comes from UNCTAD (2005, 2007) and the World Bank (2009).

INWARD FOREIGN DIRECT INVESTMENT AND R&D IN THE ASIAN BRIICS ECONOMIES The dominant modes of FDI entry in developed economies are mergers and acquisitions. In contrast, in newly-industrializing countries such as China, India and Indonesia, the dominant modes of FDI entry are socalled ‘greenfield’ projects. Greenfield FDI projects are the foreign affiliate investments whereby a brand new establishment is constructed on a new site, and this mode of FDI obviously represents a very different form of FDI from those undertaken by M&As. The number of such greenfield FDI projects increased globally by 13 per cent to some 11 800 projects in 2005 (McCann, 2009). Manufacturing accounted for 54 per cent of these projects, with the service sector accounting for 42 per cent and primary industries accounting for 4 per cent. In terms of broad regions, South, East and South East Asia accounted for 3515, or some 30 per cent of these greenfield projects. Figure 9.2 gives us a sense of the relative order of magnitude of these inward greenfield FDI projects in Asian economies in 2005 in comparison to advanced economies. China alone accounted for

STIMSON PAGINATION (M2469).indd 163

20/12/2010 15:12

164

Endogenous regional development

Source: The authors.

Figure 9.2

The number of greenfield inward FDI projects by country in 2005

1378 greenfield FDI projects, or 11.6 per cent of the global total. India accounted for 981 greenfield projects, representing 8.3 per cent of the global total, while Indonesia accounted for 93 projects (UNCTAD, 2007). In order to give a sense of the relative scale of these numbers, in the same year, as we see in Figure 9.2, the number of greenfield inward FDI projects in the US was 723, UK 669, France 582. Asian economies, and in particular China and India, are by far the most important locations in the world for greenfield FDI projects (McCann, 2009). Yet as we see in Figure 9.2, apart from China, the order of magnitude of these greenfield project numbers is still roughly comparable to many of the large advanced economies. China now has by far the largest number of domestically located multinational foreign establishments with 42 753 foreign affiliates in 2004 with some 24 million employees. These affiliate establishments are heavily concentrated in manufacturing. The 24 million employed in China in foreign affiliates represents one-third of the global total workforce currently employed in foreign affiliates. This number has increased fivefold from less than five million in 1991, a number that is equivalent to the current total level of domestic US employment in foreign affiliates (McCann, 2009).

STIMSON PAGINATION (M2469).indd 164

20/12/2010 15:12

Growth in the Asian BRIICS countries

165

In the case of Indonesia, a substantial phase of deregulation started in 1986, and from the mid-1980s onward employment in foreign-owned manufacturing increased rapidly. These increases were concentrated in the machinery industries and in multinational firms with large foreign ownership shares (Takii and Ramstetter, 2005). Inward foreign direct investment arrived primarily from North East Asia (ibid.), with Japan being the dominant source of inward FDI (Syamwil and Tanimura, 2000). This was just at the time that Indonesia needed to restructure its manufacturing sector to be more labour-intensive and export-orientated (Timmer, 2004). The initial 1986 deregulations regarding foreign investment were then further bolstered by more fundamental reforms in 1989, 1992 and 1994. Rapid growth and industrialization were both causes and results of even larger increases in inflows of FDI during the economic boom of the late 1980s and early 1990s (Takii and Ramstetter, 2005), and then beyond the economic crisis of 1997 (Sjoholm, 1997). The FDI policies in Indonesia are now as liberal as they were in the late 1960s, and are largely in line with most other countries. Although official estimates suggest that there was an abrupt reversal of these flows in 1998–99 through to 2001, on the basis of industrial and employment survey data, however, Takii and Ramstetter (2005) find that the presence of multinational firms in Indonesia continued through the crisis of 1997–98 and beyond at a modest rate both in terms of employment and value-added. Aggregate employment in multinational firms in Indonesia increased by some 20 per cent between 1995–97 and 2000–01, with the multinational firms’ shares accounted for by large plants also increasing from 19 per cent to 21 per cent during the same period (ibid.). The knowledge base of the Asian BRIICS countries is also steadily increasing, and the possibilities for upgrading R&D and innovation in these countries over the long term are obvious given the sheer scale of the countries. For example, in 2001, China and India together accounted for close to one-third of the total global number of tertiary educated technical people (McCann, 2009), while the Bangalore high-technology industries alone have 35 000 people who are US educated or trained. Yet, of all developing economies, it is China’s growth in its R&D capacity that has been the most remarkable (ibid.). Between 1996 and 2003 China increased its domestic R&D expenditure by over 3.8 times, such that by 2002 China became the only newly-industrializing economy that is amongst the world’s top ten R&D expenditure countries. China is now ranked number one for both total R&D and also business R&D expenditure, and India is number four. After the US and UK, China is now third in the world for the total number R&D-related foreign affiliates located there. In 1996 it was outside of the top ten, but by 2002 it was ranked sixth in the world in

STIMSON PAGINATION (M2469).indd 165

20/12/2010 15:12

166

Endogenous regional development

terms of total R&D expenditure and seventh in the world in terms of business R&D expenditure (ibid.). Amongst developing regions, it is the countries in South and East Asia that are the major locations for multinational R&D investment. During 2002–04, of the 1773 inward FDI projects involving an R&D component, 1095 (62 per cent) were undertaken in developing or transition economies, of which 861 projects (49 per cent) were undertaken in developing Asia alone. In the case of developing Asia, the share of R&D accounted for by the foreign affiliates of US multinational firms increased from 3 per cent of their total foreign located R&D in 1994 to 10 per cent in 2002. Similar trends are also observable for multinational firms from other developed economies that are locating R&D-related investments in Asia (ibid.). Within developing Asia itself, it is China in particular that dominates inflows of multinational R&D investment, and the impact on China of these inflows has also been the most marked. Between 1998 and 2002, the share of total domestic business R&D in China accounted for by foreign affiliates located there increased from 18 per cent to 22 per cent. The R&D expenditure associated with this R&D-related inward FDI now accounts for 13.5 per cent of China’s total domestic public sector plus private sector R&D expenditure (ibid.). To get a sense of how important these multinational R&D investments are to China, we can observe that the 42 000 foreign affiliates located in China are currently employing some 24 million Chinese, and this still only represents 3.1 per cent of the total employed workforce in China. As such, the relative importance of multinational R&D expenditure to China’s knowledge-related activities is four times greater than the relative importance of multinational firms to China’s overall employment (ibid.). In contrast, in the case of India, the most recent reliable estimates indicate that in 1999, multinational R&D expenditure accounted for just 3.4 per cent of domestic private sector R&D. The types of R&D undertaken in different countries and regions therefore appear to vary. In India, over three-quarters of R&D expenditure is on services, and primarily on software development, whereas in China, most multinational R&D focuses on adaptive innovations for the Chinese market, but there is increasing evidence across Asia that innovative R&D is growing. It is clear that the scale of China’s and India’s growth in both domestic R&D and also inward FDI-related R&D growth is very significant, and in both cases these are very much greater than the equivalent growth rates for Indonesia. However, from Figure 9.3 we can get a sense of the relative global and regional contribution of China’s knowledge sectors by considering other indicators of innovation such as patents and trademarks. From the perspective of such knowledge indicators, the relative role

STIMSON PAGINATION (M2469).indd 166

20/12/2010 15:12

Growth in the Asian BRIICS countries

167

Source: The authors.

Figure 9.3

The knowledge outputs of the Asian economies 2001–03

STIMSON PAGINATION (M2469).indd 167

20/12/2010 15:12

168

Endogenous regional development

played by China and India is very much less than the output growth and FDI figures imply. The total 2002 R&D expenditure of developing countries in South, East and South East Asia including both China and India is only 14 per cent of the value for Japan and 6.7 per cent of US R&D expenditure (ibid.). Similarly, if we consider the number of US patents and trademarks granted to the residents of particular countries during 2001–03, we see that China had 1543, while India had 1022, and Indonesia 108. For comparison, the respective figures for Taiwan and Korea are 20 414 and 12 195. Similarly, in terms of US patents and trademarks granted to firms or organizations of particular countries during 2001–03 we see that India had 558, China 475 and Indonesia 31. Once again, for comparison purposes the equivalent figures for Taiwan and Korea are 12 606 and 11 152, respectively (ibid.). Therefore, although amongst the newly-industrializing countries China and India are the leading knowledge generators, their relative contributions are not large in comparison to other advanced Asian economies. In total, the share of global R&D expenditure accounted for by all the BRIICS countries combined is only 4 per cent, of which China accounts for more than half of this level. In contrast, the world’s developed economies account for 94.7 per cent of global R&D expenditure (ibid.). As such, it will still take a long time for the knowledge base of the newlyindustrializing countries to catch up with that of the advanced economies.

THE CHANGING ECONOMIC GEOGRAPHY OF THE ASIAN BRIICS COUNTRIES The enormous transformations that are currently taking place within the newly-industrializing economies also have a very explicit economic geography logic to them that is more or less common across each of the countries. In each case, national economic growth is inherently about regional restructuring and the role of agglomeration economies is central to these growth processes. China Of all developing or transition economies, the most remarkable transformation has been that of China (McCann, 2009). Between 1980 and 2000 China increased its share of global exports from 0.9 per cent to 6 per cent, its share of global imports from 1.1 per cent to 4.1 per cent and its share of global GDP from 2.9 per cent to 3.4 per cent (Fujita, 2007). The result of this increasing trade and openness was that between the 1980s and

STIMSON PAGINATION (M2469).indd 168

20/12/2010 15:12

Growth in the Asian BRIICS countries

169

2000, China’s GDP per capita increased by ten-fold. However, the initial impetus for China’s restructuring and growth came from fairly modest reforms. China began with the introduction of a rudimentary system of property rights in order to create incentives and only recently gave constitutional recognition to private property (World Bank, 2005). However, recognizing the need to access global capital, technology and knowledge assets via inward multinational investment, China has subsequently also liberalized many rules regarding services and manufacturing industry ownership. These changes now allow for greater levels of overseas ownership in many advanced sectors and have been instituted because China is aiming to attract both a broader range and a higher quality of inward FDI. In particular, the National Economy and Social Development Plan 2005 emphasized the need to improve the quality of FDI by encouraging it in high-technology industries, advanced manufacturing, modern services, agriculture and environmental protection. The plan encourages the establishment of R&D centres, regional headquarters and bases of advanced manufacturing. It also welcomed the role of FDI in the reform of stateowned enterprises (McCann, 2009). The growth in R&D-related FDI investments in China began in 1993 and reached some 700 projects by 2004 amounting to some $4bn in inward FDI. Most projects were implemented after China’s accession to WTO in December 2001 (ibid.). These R&D investments are mainly focused on technology-intensive industries such as ICT, automotive and chemicals, and there is clear economic geography logic to these investments, which are concentrated in a small number of locations. As seen in Figure 9.4, in 2004, Beijing had 189 foreign-owned R&D centres of which 60 per cent were in ICT, Shanghai had 140 foreign-owned R&D centres of which 91 are in Pudong, and Guangdong and Jiangsu provinces in the south close to Hong Kong are home to a combined number of over 100 R&D centres (ibid.). The fact that multinational R&D centres were being located in Shanghai, Beijing, Guangdong and Jiangsu provinces displays a clear logic. These locations are the core knowledge regions that are growing quickly, and form the major locations for all types of international investment. Although they cannot yet be described fully as global cities (McCann, 2008), both Beijing and Shanghai exhibit some world city characteristics. At the same time, in 2003 the south eastern provinces of Guangdong and Jiangsu individually accounted for 28 per cent and 19 per cent of FDI, respectively. The reason that FDI in general, and knowledge-related FDI in particular, was being located in these cities and regions, is because these are the major growth regions of China (McCann, 2009). During the 1970s and 1980s, inequality between provinces and also

STIMSON PAGINATION (M2469).indd 169

20/12/2010 15:12

170

Endogenous regional development

Source: The authors.

Figure 9.4

The spatial distribution of R&D centres in China in 2004

between urban and rural areas in China fell consistently (Golley, 2007). Until the mid-1980s, the growth in per capita productivity and expenditure per capita was higher in rural than in urban areas, which suggested a slow process of rural–urban convergence (Angang et al., 2005). Between 1978 and 1985, the ratio of per capita disposable income between urban and rural residents had fallen from 2.57 to 1.85, and ratio of per capita consumption had fallen to just over 2.1 (ibid.). However, from the mid1980s onwards this urban–rural ratio has been reversed. By 1990 the ratio of both per capita disposable income and consumption had risen to above 2.0. Since the economic reforms started in earnest, as expected on the basis

STIMSON PAGINATION (M2469).indd 170

20/12/2010 15:12

Growth in the Asian BRIICS countries

171

Source: The authors.

Figure 9.5

Relative urban to rural incomes in China in 2004

of the earlier arguments, inequality between provinces in China has risen continuously since 1991 (Golley, 2007). By 1997, the urban–rural ratios of both per capita disposable income and consumption had increased to approximately 2.5 (Angang et al., 2005). By 2001 the urban–rural ratio of both per capita disposable income and consumption had risen to approximately 3:1, while the ratio of per capita income had risen to 2.1:1 (ibid.). The result of this urban rural divergence was that, as we see in Figure 9.5, by 2004 the largest city-regions in China exhibit the highest per capita incomes, with the ratio of per capita province GDP to national per capita GDP being highest for Shanghai 5.0, then for Beijing 3.5, then Tianjin 2.7.

STIMSON PAGINATION (M2469).indd 171

20/12/2010 15:12

172

Endogenous regional development

The next highest ratio areas are the regions of Zhejiang 2.1 and Jiangsu 1.8, which are the regions close to Shanghai, followed by the regions close to Hong Kong or Guangdong 1.8, and then the coastal regions close to the dominant cities of Fujian 1.6, Liaoning 1.6 and Shandong 1.4 (Golley, 2007). These ratios imply that the dominant cities currently exhibit GDP per capita levels that are approximately 13 times that of the lowest regions (Fujita, 2007). If the three major city-regions of Shanghai, Beijing and Tianjin are removed then the increase in regional inequality across China is noticeably reduced. On the other hand, however, if we group together all of the coastal regions including the dominant city-regions, then regional inequality between the coastal and interior regions of China increases even more dramatically (Golley, 2007). This demonstrates the role played by particular city-regions in the dramatic growth of China over the last two decades. In 2000, the coastal region between Beijing and Hong Kong as a whole produced 71 per cent of China’s total industrial output. This enormous output accounted for more than 60 per cent of output in all but two sectors and at least 80 per cent of output in close to half of the industry sectors, including 97 per cent of China’s cultural, educational and sports outputs (ibid.). The growth of China is a coastal phenomenon. However, even within the coastal region of China there is a core region, which consists of the south east regions adjacent or close to the major cities, and represents broadly an arc of regions bounded by Shanghai and Hong Kong. These core region provinces grew by more than the coastal region as a whole in almost all of the sectors in which the coastal region grew (ibid.). As expected on the basis of earlier arguments, increasing inter-regional inequality is now a general phenomenon in China. However, in terms of economic geography, the escalating growth and wealth of certain regions is also highly associated with the increasing agglomeration of activities in these regions. Once again, this is predicted by economic geography arguments. The core regions of the south east are not only the fastest per capita growth regions, but also they are the regions of the most rapidly increasing agglomeration. Golley (2007) calculates that that between 1989 and 2000, 26 out of 28 major manufacturing and industrial sectors have become more spatially concentrated, as reflected by increasing spatial Gini coefficients. As such, the general trend towards increasing intranational inequality across many countries is clearly very evident in China (McCann, 2009). Regional economic restructuring in China has meant that poverty reduction since the mid-1980s has been most dramatic in the eastern regions, followed by the central regions, with poverty increasing in the western regions (Angang et al., 2005). However, this is not just an urban

STIMSON PAGINATION (M2469).indd 172

20/12/2010 15:12

Growth in the Asian BRIICS countries

173

phenomenon. The ratio of per capita farming incomes in the east and centre regions relative to the west region have also increased between 1980 and 2000, from 1.27 and 1.05, to 1.92 and 1.30, respectively (ibid.). Part of the reason is human capital. The areas of highest growth are broadly the regions with highest rates of literacy (ibid.). In addition, disparities in Chinese income per capita are also exacerbated by a fiscal tax and transfer system that significantly benefits urban residents (ibid.). More generally, however, the competition and wealth effects associated with buoyant regional growth across a range of local sectors tend to spill over to other local sectors, and agriculture in such buoyant regions also benefits from this. India Many of these same regional economic and economic geography phenomena evident in China are apparent in countries such as India and Indonesia (McCann, 2009). Like China, India began its economic restructuring with initially modest reforms by reducing trade barriers and distortions within the economy. In 1991 the average tariff was 83 per cent, and only 13 per cent of goods were importable without a licence. By 1998 tariffs had been reduced to 30 per cent, and the range of goods importable without a licence was 57 per cent. Since then, its GDP per capita has increased fourfold between 1980 and 2002 (World Bank, 2005). As in the case of China, recognizing the need to access global capital, technology and knowledge assets via inward multinational investment, India has also moved to increase its attractiveness as a location for FDI. The Indian Investment Commission aims at drawing FDI as well as domestic investment, while the Foreign Investment Board is intended to act a one-stop service centre and facilitator for FDI. In 2004 foreign equity ceilings in Indian aviation services, private banks, non-news print publications, and the petroleum industry were all adjusted upwards in order to attract more international investment. As with China, the process of deregulation has also been accompanied by enormous internal rural to urban migration flows, and the expansion of many of the already immense cities. As we see in Figure 9.6, the major growth centres within India are the mega-cities of Mumbai, Calcutta and Delhi, as well as the high-technology centre of Bangalore. Yet, even though China and India are now often compared, there are actually fundamental differences between the two economies and their responses to globalization (McCann, 2009). First, one obvious difference is simply sheer scale. The Chinese economy is almost three times the size of the Indian economy, with per capita incomes of well over twice those of India. Second, there are major trade performance differences between the

STIMSON PAGINATION (M2469).indd 173

20/12/2010 15:12

174

Endogenous regional development

Source: The authors.

Figure 9.6

The major economic growth city-regions in India

two countries. In 1950, China’s share of global trade was 1 per cent while that of India’s was 2.2 per cent, whereas by 2002, China’s share of global trade had increased to 4.8 per cent while India’s had actually declined to 0.8 per cent (Lardy, 2005). In part, these trade performance differences are because China’s rapid growth began slightly earlier than India’s growth, and also because it has been more dramatic than India’s, particularly in opening trade. Third, there are also major differences between China and India in terms of their structure of GDP composition (McCann, 2009). The structure of GDP composition by industry in India is still rather different from China; there is a much greater emphasis in India on services than China, where manufacturing is still relatively more dominant (Panagariya, 2005). In 1980 in China, agriculture accounted for 30.1 per cent of GDP, industry for 48.5 per cent (of which manufacturing alone

STIMSON PAGINATION (M2469).indd 174

20/12/2010 15:12

Growth in the Asian BRIICS countries

175

accounted for 40.5 per cent) and services for 21.4 per cent of GDP. On the other hand, for India, agriculture accounted for 38.6 per cent of GDP, industry for 24.2 per cent (of which manufacturing accounted for only 16.3 per cent) and services for 37.2 per cent. Even though both countries have undergone enormous changes during the last three decades, the legacy of these inherited structures still remains. In 2002 in China, agriculture accounted for 15.9 per cent of GDP, industry for 50.9 per cent (of which manufacturing alone accounted for 34.5 per cent) and services accounted for 33.2 per cent of GDP, whereas for India, agriculture accounted for 24.9 per cent of GDP, industry for 26.9 per cent (of which manufacturing accounted for only 15.8 per cent) and services for 48.2 per cent of 2002 GDP (ibid.). As a result of its different industrial structure and also its English language advantages, the growth of FDI in India, and particularly the growth of off-shoring FDI, has been dominated by a range of service industries, rather than by manufacturing, which has been the case for China (McCann, 2009). Yet, many aspects of the dynamic growth industries of India have similar features to China. First, most of the trade of the Indian and Chinese economies is still in the form of re-exports of finished or semi-finished products or services produced by multinational firms that are based in Europe or the US. Second, many of the key growth centres are dominated by external links with multinational companies. In the Indian IT industry, which is dominated by the Bangalore region, two-thirds of all sales are accounted for by foreign-owned multinational affiliates located there (Scheve and Slaughter, 2007). Third, as India undergoes continuing regional economic restructuring, firms located in the regions with large home markets earn higher profits (Kambhampati and McCann, 2007). This implies that in terms of economic geography, the large home market effects associated with agglomeration are driving the internal economic growth and restructuring within the Indian economy. Indonesia In terms of economic geography and the regional aspects of economic growth, a similar picture emerges in the case of Indonesia. Indonesia has a population of between 220 and 230 million, spread over 900 islands spanning 5500km in an east–west direction (Amiti and Cameron, 2006), with almost two-thirds of the population living on the island of Java. As with both India and China, deregulation has been followed by enormous rural to urban migration flows, and the major economic growth areas are the immense cities of Java. Excluding the oil and gas sectors, the majority of overseas investments in Indonesia have occurred within

STIMSON PAGINATION (M2469).indd 175

20/12/2010 15:12

176

Endogenous regional development

Source: The authors.

Figure 9.7

The major economic growth city-regions in Indonesia

the non-oil-related manufacturing sectors in these same regions. These are the very same sectors that are not only key to the trade growth of the Indonesian economy as a whole, but are also concentrated in the expanding urban regions of Java (see Figure 9.7). The reasons why FDI inflows exhibit such an ‘economic geography’ logic is partly due to the fact that there are major regional disparities across Indonesia in terms of access to international transportation infrastructure. Jakarta is the national hub for the international air-transport system, and many parts of Indonesia are only accessible to the outside world via Jakarta. Similarly, there is also great disparity across Indonesia in terms of crucial communications infrastructure. While national fixedline density is only around 3.7 per cent, telecommunications infrastructure density for metropolitan Jakarta is well above 25 per cent, with other major cities such as Bandung and Surabaya having an average telecommunications infrastructure density of 11 to 20 per cent (Lee and Findlay, 2005). In contrast, average telecommunications infrastructure density for

STIMSON PAGINATION (M2469).indd 176

20/12/2010 15:12

Growth in the Asian BRIICS countries

177

rural Indonesia is only 0.2 per cent and more than 64 per cent of rural villages are entirely without telecommunications (ibid.). Like China, GDP per capita in Indonesia is also closely related to city size, with the dominant city of Java exhibiting GDP per capital levels that are approximately 13 times that of the lowest regions (Fujita, 2007). The opening up of the country is likely to accentuate these differences, except for the case where multinationals are engaged in primary sector activities located in other outlying regions. These spatial disparities in terms of population and infrastructure are also manifested in terms of firm investment behaviour. Since the emergence of the Indonesian manufacturing industry from the 1970s onwards, the role and importance of the major urban areas of Java has increased. Moreover, as deregulation has continued apace, the relative advantage of these urban regions has increased even further. Notwithstanding the oneoff shock of the 1997–98 crisis, over the long term more and more firms have been investing ever-increasing amounts in the urban centres of Java, the results of which will tend to be increasing regional inequality within Indonesia. Yet, why firms do not generally relocate from urban Java to other parts of Indonesia in order to take advantage of the very large variations in labour prices that exist across Indonesia’s regions, and even within individual islands such as Java, is a moot point (Amiti and Cameron, 2007). Although trade protectionism probably played a role in the initial concentration of production and investment in these regions, the reasons why such an extreme spatial concentration of investment within Indonesia continues appears to be related primarily to the existence of agglomeration economies. Indonesian manufacturing exhibits extreme spatial concentration in a small number of regions within Java (Syamwil and Tanimura, 2000). Over the last three decades these Javan regions have consistently accounted for over 85 per cent of total national manufacturing output in large and medium-sized industries (ibid.), and these geographical patterns of investment and employment are similarly exhibited for both domestic and foreign-owned manufacturing firms. As such, increasing international trade and investment openness appears to be associated with greater disparities among Indonesia’s regions, as inward investing firms seek out the pecuniary advantages of associated agglomeration. This is particularly the case in the apparel and textiles industries, which is the sector in Indonesia that has exhibited the greatest revealed comparative advantage since the 1980s (James, 2007). If inward technology spillover effects also operate in these sectors, these are likely to exacerbate the existing advantages of agglomeration in the dominant urban regions of Java (Amiti and Cameron, 2007).

STIMSON PAGINATION (M2469).indd 177

20/12/2010 15:12

178

Endogenous regional development

On the basis of data from the 1980s and 1990s Amiti and Cameron (ibid.) estimate the agglomeration advantages that arise from vertical linkages between firms in Indonesia. Taking account of the location of input suppliers to estimate cost linkages and the location of customers to estimate demand linkages, Amiti and Cameron (ibid.) show that the externalities that arise from demand and cost linkages in Indonesia are quantitatively very important and highly localized. They find that an increase in either cost or demand linkages from the 10th to the 90th percentile increases wages by more than 20 per cent. As such, firms benefit greatly from proximity to a large supply of inputs and good market access, and firms with the best supply or market access can afford to pay more than 20 per cent higher wages than those with the poorest access (Amiti and Cameron, ibid.). Moreover, the benefits of these vertical linkages are extremely highly localized. Only 10 per cent of the market access benefit spreads beyond 108km and only 10 per cent of the supply access benefit spreads beyond 262km (Amiti and Cameron, ibid.). At the same time, firms located in the major urban areas tend to be more vertically disintegrated and to exhibit strong local linkages with a variety of backward input supply sources than firms in rural areas (Amiti and Cameron, ibid.). There are two crucial aspects to the findings of Amiti and Cameron (ibid.). First, such extreme economic advantages associated with geographical localization imply that Indonesian agglomeration effects operate almost entirely within Java. As agglomeration effects are not only associated with spatial clustering but also with economies of scale, market access and infrastructure (Syamwil and Tanimura, 2000) these findings suggest that Indonesian economic growth will continue to become systematically more oriented towards the urban areas of Java. For a country that spans 5500km across 900 islands, the prospects for other regions within Indonesia would appear to be very much more limited because firms located in Indonesia’s outer islands are too far away to benefit from the agglomeration advantages associated with industries on the main island of Java (Amiti and Cameron, 2007). The logic of these agglomeration effects therefore means that central government efforts to promote economic growth in geographically peripheral and low wage regions face enormous challenges. The second crucial aspect of the findings of Amiti and Cameron (ibid.) is regarding the nature of the agglomeration effects in Java. They find that labour pooling is quantitatively much less important than the existence of localized vertical linkages and there also appear to be little or no real technology spillover effects. Regarding these local spillover effects, most multinational firms in Indonesia are export-oriented and generally do not

STIMSON PAGINATION (M2469).indd 178

20/12/2010 15:12

Growth in the Asian BRIICS countries

179

supply Indonesian customers (Blalock and Gertler, 2007). As such, any pecuniary or technology spillover effects that are found to operate are likely to be upstream spillovers affecting the domestic supplier industries. Even among Asian countries, however, Indonesia is a relative latecomer in terms of its industrial sophistication, technological capabilities, R&D effort and export orientation (Kumar et al., 1999). This could be due to a lack of absorptive capacity, given the very low levels of domestic Indonesian R&D. Although they are very limited, technology spillovers from FDI are found primarily in sectors with a high degree of competition (Sjoholm, 1997; Blalock and Veloso, 2007), and the degree of pecuniary spillovers depends on the extent of the technology gap between domestic and foreign firms, and the level of domestic market concentration. At the same time, the very low observed levels of R&D underline the importance of inward FDI as a potential external source of technology and knowledge for Indonesia’s future growth.

CONCLUSIONS The growth experience of China, India and Indonesia all share some common elements. Most notably, the economic growth in each of these countries is dominated by certain regions, in which agglomeration economies appear to be crucial. As such, major cities are the focus for economic growth in all three societies. Both labour and capital are increasingly mobile and the location of activity is shifting dramatically into the major urban centres. The result of this is that each society is becoming more unequal, and the gap between the urban and rural areas is increasing rapidly. In addition, the gap between the dominant urban areas and other smaller urban areas is also increasing. On the other hand, there is something of a difference between the countries in the role played by localized technology spillovers. The positive effects of these local technology spillovers appears to be more marked in China than in India, and in turn the effect of these in India is far more marked than in Indonesia, where the evidence for these is very limited indeed.

REFERENCES Amiti, M. and Cameron, L. (2007), ‘Economic Geography and Wages’, Review of Economics and Statistics, 89(1), 15–29. Angang, H., Linlin, H. and Zhixiao, C. (2005), ‘China’s Economic Growth and Poverty Reduction’, in W. Tseng and D. Cowen (eds), India and China’s

STIMSON PAGINATION (M2469).indd 179

20/12/2010 15:12

180

Endogenous regional development

Recent Experience with Reform and Growth, International Monetary Fund and Palgrave, Basingstoke. Blalock, G. and Gertler, P.J. (2007), ‘Welfare Gains from Foreign Direct Investment through Technology Transfer to Local Suppliers’, Journal of International Economics, 74(2), 402–21. Blalock, G. and Veloso, F.M. (2007), ‘Imports, Productivity Growth, and Supply Chain Learning’, World Development, 35(7), 1134–51. Fujita, M. (2007), ‘Development of East Asian Regional Economies: A View from Spatial Economics’, in M. Fujita, Regional Integration in East Asia from the Viewpoint of Spatial Economics, Palgrave, Basingstoke, UK. Golley, J. (2007), The Dynamics of Chinese Regional Development, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. James, W.E. (2007), ‘Comparative Advantage in Thailand and Indonesia and Potential Free Trade Agreements: Implications for Trade Diversion’, in M.A.B. Siddique (ed.), Regionalism, Trade and Economic Development in the AsiaPacific Region, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Kambhampati, U.S. and McCann, P. (2007), ‘The Regional Performance and Characteristics of Indian Manufacturing Industry’, Regional Studies, 41(3), 281–94. Kumar, V., Kumar, U. and Persaud, A. (1999), ‘Building Technological Capability Through Importing Technology: The Case of Indonesian Manufacturing Industry’, Journal of Technology Transfer, 24(1), 81–96. Lardy, N. (2005), ‘Trade Liberalization and its Role in Chinese Economic Growth’, in W. Tseng and D. Cowen (eds), India and China’s Recent Experience with Reform and Growth, International Monetary Fund and Palgrave, Basingstoke. Lee, R.C. and Findlay, C. (2005), ‘Telecommunications Reform in Indonesia: Achievements and Challenges’, Bulletin of Indonesian Economic Studies, 41(3), 341–65. McCann, P. (2008), ‘Globalization and Economic Geography: The World is Curved, Not Flat’, Cambridge Journal of Regions, Economy and Society, 1(3), 351–70. McCann, P. (2009), ‘Globalisation, Multinationals and the BRIICS Countries’, in R. Lattimore and R. Safadi (eds) (2009), Globalisation and Emerging Economies, OECD, Paris. Panagariya, A. (2005), ‘India in the 1980s and the 1990s: A Triumph of Reforms’, in W. Tseng and D. Cowen (eds), India and China’s Recent Experience with Reform and Growth, International Monetary Fund and Palgrave, Basingstoke. Scheve, K.V. and Slaughter, M.J. (2007), ‘A New Deal for Globalization’, Foreign Affairs, 34–47, July–August. Sjoholm, F. (1997), ‘Technology Gap, Competition and Spillovers from Foreign Direct Investment: Evidence from Establishment Data’, Working Paper No. 38, WIJS The European Institute of Japanese Studies, Stockholm. Syamwil, I.B. and Tanimura, P.H. (2000), ‘The Spatial Distribution of Japanese Manufacturing Industries in Indonesia’, Review of Urban and Regional Development Studies, 12(2), 121–35. Takii, S. and Ramstetter, E. (2005), ‘Multinational Presence and Labour Productivity Differentials in Indonesian Manufacturing’, Bulletin of Indonesian Economic Studies, 41(2), 221–42. Timmer, C.P. (2004), ‘The Road to Pro-poor Growth: The Indonesian Experience in Regional Perspective’, Bulletin of Indonesian Economic Studies, 37(1), 177–207.

STIMSON PAGINATION (M2469).indd 180

20/12/2010 15:12

Growth in the Asian BRIICS countries

181

UNCTAD (2005), World Investment Report: Transnational Corporations and the Internationalization of R&D, United Nations Conference on Trade and Development, United Nations, New York and Geneva. UNCTAD (2007), World Investment Report: Transnational Corporations, Extractive Industries and Development, United Nations Conference on Trade and Development, United Nations, New York and Geneva. World Bank (2005), World Development Report 2005: A Better Investment Climate for Everyone, Washington DC. World Bank (2009), Reshaping Economic Geography: World Development Report, Washington DC.

STIMSON PAGINATION (M2469).indd 181

20/12/2010 15:12

10.

Implications of European Union structural assistance to new member states on regional disparities: the question of absorption capacity Daniela Constantin, Zizi Goschin and Gabriela Dragan

INTRODUCTION In May 2004 the largest enlargement of the European Union (EU) took place with ten new member states (NMS), mostly from Eastern and Central Europe, joining it. They were then followed by Romania and Bulgaria in January 2007. That expansion of the EU entailed an unprecedentedly high amplitude of inter-regional disparities (Eurostat, 2007), and was of special significance for the cohesion policy in the 2007–13 programming period. In this chapter we discuss the implications of the EU structural assistance to the NMS on regional disparities, special emphasis being placed on the capacity of those states to absorb the allocated funds. First we provide a brief outline of the EU’s economic and social cohesion policy. We then look at the influence of the EU enlargement on that cohesion policy in the 2007–13 programming period, revealing the close links between cohesion and the EU’s regional policy. Next the regional disparities in the NMS within European context are discussed in connection with the contribution of the structural instruments to reducing these disparities. Finally the absorption capacity of the EU funds in the NMS is examined as a precondition for achieving this goal. A particular emphasis is placed on the administrative absorption capacity, considering that the institutional framework created in the NMS for the administration of the EU funds is expected to play a decisive role for their successful and complete integration in the EU structures. That is, the focus is on considering the nature of institutional issues as either enhancing or detracting endogenous factors.

182

STIMSON PAGINATION (M2469).indd 182

20/12/2010 15:12

Implications of European Union structural assistance

183

THE EU ECONOMIC AND SOCIAL COHESION POLICY: ADDRESSING REGIONAL DISPARITIES Addressing regional disparities through interventions that have the ultimate objective of achieving convergence is a hallmark of regional policy in the EU. The EU’s economic and social cohesion may be considered in terms of its two inter-related components, namely: (1) vertical cohesion, referring to the alleviation of social disparities and the solidarity with the disadvantaged social groups; (2) horizontal cohesion, which concentrates on the regional disparities’ decrease and the solidarity with the lagging regions’ population. Based on this approach, the EU’s regional policy is closely related to the horizontal dimension of the cohesion policy and is supported by the EU funds via allocations for convergence – competitiveness and employment – European territorial cooperation objectives of the current financial exercise. During the programming period 2007–13, the cohesion policy ranks first in terms of expenditure and coverage, with cohesion surpassing agriculture for the first time as the largest area of expenditure undertaken by the EU (Leonardi, 2006). In a total budget of approximately 862.4 billion euros, cohesion policy accounts for 307.6 billion euros (35.6 per cent); that is an average annual expenditure of 44 billion euros, compared with 41.8 billion euros allocated to market-related expenditure and direct payments to agriculture. As it was conceived from the very beginning as a necessary complement to the Single Market and Single Currency programmes, the cohesion policy will be of a great importance to the objectives of economic convergence – real and nominal – between the EU-15 and the NMS: (1) real convergence refers to the narrowing of development gaps: similarity of per capita GDP, nominal wage levels, equilibrium of real exchange rates and related to this, price levels and tradable/non-tradable price ratios; (2) nominal convergence aims at the narrowing and finally closing of the gaps in macroeconomic stability currently existing between NMS and incumbents. It focuses on the Maastricht criteria on inflation, interest rates, fiscal variables and exchange rate stability (Kasman et al., 2005). Almost 82 per cent of the total budget for cohesion policy is allocated to the objective of convergence of the member states and the regions. Its key aim is to promote growth-enhancing conditions and factors leading to real convergence within the EU. This objective covers the member states and the regions whose development is lagging behind. The targeted regions are those NUTS 2 regions with a GDP per capita less than 75 per cent of the EU average. They are funded from the European Regional Development Fund and the European Social Fund. At the same time the member states

STIMSON PAGINATION (M2469).indd 183

20/12/2010 15:12

184

Endogenous regional development

whose GNI is less than 90 per cent of the EU average benefit from the Cohesion Fund. In this context it is obvious that the NMS are the main beneficiaries of the renewed cohesion policy: all of them receive allocations from the Cohesion Fund, while 51 regions out of 55 NUTS 2 regions in the NMS are funded under the convergence objective. Hence, the big challenge these countries have to face in the current financial circumstances of the EU is whether they will be able to use the large amount of allocated funds. And further on, the question is whether they will be able to promote adequate economic policies and economic behaviour so as to generate high rates of endogenous growth and, thus, ensure an effective use of these funds. The answer to the first question is usually addressed in terms of the so-called ‘absorption capacity’ pertaining to the EU cohesion policy. That defines the degree to which a country is able to effectively and efficiently spend the financial resources allocated via European Funds. In other words, it expresses the ability of an EU member state to ‘digest and consume’ the funds in order to foster its development and thus to improve its economic and social performance (NEI, 2002; Horvat, 2004). The absorption capacity can be addressed from the perspective of the institutional system created in each member state in order to manage the funds (the supply side) as well as from the perspective of the beneficiaries of these funds (the demand side). The demand side mainly expresses the ability of the potential beneficiaries – public or private – to generate appropriate and acceptable projects (possible to be financed). The supply side is determined by three main factors, leading to three components of the absorption capacity, namely macroeconomic, administrative and financial absorption capacity: 1.

2.

Macroeconomic absorption capacity indicates the rate of the EU funding in terms of the GDP of the recipient member state. The European Summit in Berlin (1999) and then the results of the Copenhagen negotiations (2002) on ‘financial chapters’1 indicate an upper limit for the Structural and Cohesion Funds set at 4 per cent of the GDP of the respective member state. The capacity to absorb the macroeconomic effects generated by the inflow of the supplementary investments is also related to the macroeconomic absorption capacity. Administrative absorption capacity represents the ability and skills of central, regional and local authorities to prepare acceptable plans, programmes and projects in due time, to decide on programmes and projects, to arrange coordination among principal partners, to cope with the vast amount of administrative and reporting work required by the European Commission and to finance and supervise the

STIMSON PAGINATION (M2469).indd 184

20/12/2010 15:12

Implications of European Union structural assistance

3.

185

implementation properly, avoiding fraud as far as possible (Horvat, 2004). Financial absorption capacity refers to the ability to co-finance EU-supported programmes and projects, to plan and guarantee the national contributions in multi-annual budgets and to collect these contributions from several partners (public and private), interested in a programme or project. The national co-financing is needed, since in order to increase the incentive for using the funds efficiently the EU structural assistance finances only a part of the costs of a programme or project.

These components can be analysed by means of the EU programming documents and various evaluation studies in both EU-15 member states and NMS, although until now comprehensive studies regarding all three components have not been carried out. Moreover, for the NMS only macroeconomic and administrative absorption capacity can be evaluated so far. The financial absorption capacity can be evaluated only ex post, from 2009 onwards. According to the n+2 rule, each year’s ‘tranche’ that involves a programme co-financed by the EU funds must be used up before the end of the second year following the commitment. Nevertheless, useful information for the absorption capacity in the NMS can be found in the studies focusing on the pre-accession funds as well as on the preparedness of the (former now) candidate countries to absorb the EU funds after accession.

IMPLICATIONS OF THE EU ENLARGEMENT FOR THE COHESION POLICY 2007–13 Economic development within an integrated area might be considered as the result of two complex components: (1) competitiveness, which aims at the most efficient use of resources and factors; and (2) cohesion, which addresses mainly the reduction of discrepancies among regions and countries. However, for the EU, social and economic cohesion remains the determinant element of this equation since the main goal of the entire European construction – namely the promotion of security, stability and economic growth in the region – cannot be attained as long as significant disparities remain among different parts of this integrated area and obstruct the competitiveness of the whole structure of the EU. In terms of the number of newcomers, the latest round of enlargement might be considered the EU’s biggest ever, given that 12 NMS (ten in 2004 and two in 2007) joined the Union. The EU population has increased thus

STIMSON PAGINATION (M2469).indd 185

20/12/2010 15:12

186

Table 10.1

Endogenous regional development

Impact of successive enlargements of the EU

Enlargement EU-9 (73) EU-10 (81) EU-12 (86) EU-15 (95) EU-25 (04) EU-27 (07)

Population

Surface

+33.4 +3.7 +17.5 +6.2 +19.6 +6.5

+25.4 +7.9 +33.4 +34.9 +18.0 +8.5

GDP +32.2 +2.34 +11.3 +6.5 +8.9 +2.0

GDP/Capita −0.9 −1.3 −5.5 +0.2 −8.9 −4.0

Source: Compiled by the authors using data available from Eurostat, 2007.

by more than one-quarter, and its physical area by more than one-third. Now with almost 500 million citizens, the EU generates approximately 31 per cent of the world’s nominal Gross Domestic Product (GDP) in 2007. However, the combined GDP of all new member states has added only 11 per cent to the GDP of the EU-15, whereas the GDP per capita is 13 per cent lower than before enlargement (see Table 10.1). Compared with the EU average, the GDP per capita is 35 per cent per cent higher in the US and 15 per cent higher in Japan. If population and GDP per capita are considered, it is noticeable that the NMS have brought about an important burden for the EU: their populations represent approximately 21 per cent of the total population of the EU, whereas in all cases the GDP per capita is below the EU average (see Table 10.2). The worst situation is recorded by Bulgaria and Romania with less than 40 per cent of the EU average. By contrast, GDP per capita was 142.55 per cent in Ireland and 131.91 per cent in the Netherlands (Luxembourg apart). The changes of priorities in the new financial exercise for the period 2007–13 have been determined by changes in the overall background of the EU under internal and external pressures. The ‘new paradigm’ of the 2007–13 cohesion policy, as expressed by the European Commissioner, Danuta Hubner (2007, p. 1), should be the creation of new ‘opportunities for the future . . . rather than a compensation for the past’. In fact this new paradigm reflects the position of the most member states concerning the syntagma ‘competitiveness cohesion’, which no longer represents an antinomy – competitiveness, versus cohesion, but a tandem of interdependent objectives. Objectives According to the Community Strategic Guidelines for 2007–13 and the EU Budget, cohesion policy is considered as the main instrument at the

STIMSON PAGINATION (M2469).indd 186

20/12/2010 15:12

Implications of European Union structural assistance

Table 10.2

Population and GDP per capita in the NMS, 2006a

Country

Population (Million GDP Per Capita Inhabitants) (Euros, PPP)

Bulgaria Czech Republic Cyprus Estonia Hungary Latvia Lithuania Malta Poland Romania Slovakia Slovenia EU-27 Note:

a

187

7.70 10.30 1.00 1.30 10.10 2.3 3.40 0.40 38.10 21.58 5.40 2.00 493.58

8 700 18 600 21 900 15 900 15 300 13 100 13 500 17 700 12 400 8 800 14 700 20 800 23 500

GDP Per Capita: Percentage of the EU Average 37.02 79.15 93.19 67.66 65.10 55.74 57.45 75.31 52.76 37.45 62.55 88.51 100.00

EU-10 plus Romania and Bulgaria.

Source: Compiled by the authors using data available from Eurostat, 2008.

EU level for the accomplishment of the Lisbon Strategy. The three objectives are as follows: 1.

2.

Convergence Objective. This objective concentrates on those regions with a GDP per capita less than 75 per cent of the EU average. It envisages 100 regions, which account for 35 per cent of the EU-27 population. The purpose is to accelerate the economic convergence of less developed regions by improving conditions for growth and employment as a result of investments in human and physical capital, innovation and development of knowledge society, protection of environment and/or improving the administrative capacity. From the total amount of 264 billion euros allocated for this objective, the distribution is as follows: 67.34 per cent go to regions whose GDP per capita is below 75 per cent of the average; 8.38 per cent goes to regions under ‘statistical effect’;2 23.86 per cent goes to cohesion countries; and 0.42 per cent goes to the outermost regions3 (Euractiv, 2007). For the NMS, the Convergence Objective will play a more significant role given that the disparities among regions and states after the latest enlargement are more important. Regional Competitiveness and Employment Objective. This applies

STIMSON PAGINATION (M2469).indd 187

20/12/2010 15:12

188

3.

Endogenous regional development

to the rest of the EU, which means another 155 regions, including 61 per cent of the EU-27 population. It accounts for 15.8 per cent of the funds allocated to the cohesion policy. These regions have relatively high GDP levels, even if both growth and employment rates remain weak in many regions. The regional development programmes will strengthen regional competitiveness by supporting economic and social innovation, knowledge society, entrepreneurship, protection of environment and risk prevention. European Territorial Cooperation Objective. This objective aims at reinforcing cooperation at the cross-border, transnational and interregional levels. It involves 2.44 per cent of funds. The objective is complementary with the other two objectives, with it being possible for eligible regions to be funded under both of those previous objectives. The aim of this objective is to promote common solutions for authorities of different countries in the domain of urban, rural and coastal development, and to the development of economic relations and the setting up of small and medium-sized enterprises (SMEs).

According to the EU, the total share of the convergence regions in EU-27 GDP in 2002 was only 12.5 per cent, which compares with a 35 per cent population share (European Commission, 2006). GDP levels also indicate widely differing regional situations. The GDP per inhabitant in 2004 ranged from 24 per cent of the EU-27 average in the North-East region of Romania to 303 per cent in Inner London (Table 10.3). Among the 100 convergence regions, Romania and Bulgaria accounted for 12 of the 15 least prosperous regions (seven regions in Romania and five in Bulgaria). Structural Assistance The structural assistance for 2007–13 allocated to all member states represents 35 per cent of the EU budget (308 billion euros out of 862 billion euros total). For the EU-8 (the former communist countries) plus Romania and Bulgaria, the total amount allocated is 175 billion euros, representing more than one-half of the entire budget funds allocated for the cohesion (Table 10.4). Due to rule capping the structural assistance to a maximum of 4 per cent of the GDP of each country,4 the less developed countries – Romania and Bulgaria – have also the lowest allocation per capita. Assuming full absorption of the ceiling for the payment appropriations, the EU considers that under the new financial framework for 2007–13, the NMS will be net beneficiaries. However, the ability of the NMS to fully benefit from EU

STIMSON PAGINATION (M2469).indd 188

20/12/2010 15:12

Implications of European Union structural assistance

Table 10.3

Regional GDP per capita in the EU-27 in 2004 (in PPS, EU27 = 100)

The 15 Highest Regions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

189

Inner London (UK) Luxembourg (LU) Bruxelles Cap (BE) Hamburg (DE) Wien (AT) Ile de France (FR) Berkshire, Buckinghamshire, Oxfordshire (UK) Oberbayern (DE) Stockholm (SE) Utrecht (NL) Darmstadt (DE) Prague (CZ) Southern & Eastern (IE) Bremen (DE) North Eastern Scotland (UK)

The 15 Lowest Regions 303 251 248 195 180 175 174 169 166 158 157 157 157 156 154

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Nord-Est (RO) Severozapaden (BG) Yuzhen tsentralen (BG) Severen tsentralen (BG) Sud-Muntenia (RO) Sud-Vest Oltenia (RO) Severoiztochen (BG) Yugoiztochen (BG) Sud-Est (RO) Nord-Vest (RO) Lubelskie (PL) Podkapackie (PL) Centru (RO) Podlaskie (PL) Vest (RO)

24 26 26 26 28 29 29 30 31 33 35 35 35 38 39

Source: Compiled by the authors using data available from Eurostat (news release, 23/2007, 19 February 2007).

transfers is limited due to certain agreed reforms and transitional arrangements. As different authors have noted (see, e.g., Oprescu et al., 2005), while the old member states had to create and adapt their administrative structures to the requirements of Structural Funds gradually, for the NMS all institutional and operational requirements had to be accomplished in a very short period of time.

IMPLICATIONS OF REGIONAL DISPARITIES IN THE NMS FOR STRUCTURAL ASSISTANCE The Disparities As already mentioned, the addition to the EU of the ten NMS in 2004 and a further two in 2007 has substantially increased economic regional disparities. In EU-27, the GDP per capita in PPS is almost five times higher in the top 10 per cent of regions than in the bottom 10 per cent of regions, compared with a difference of less than three times in the EU-15 (Eurostat, 2007). In the enlarged EU, the ratio between GDP per capita in

STIMSON PAGINATION (M2469).indd 189

20/12/2010 15:12

190

Table 10.4 Country

Czech Republic Estonia Hungary Latvia Lithuania Poland Slovakia Slovenia Bulgaria Romania Total

Endogenous regional development

Total assistance allocated to the EU-10 for 2007–13 Conver- Competi- Territorial Assistance Percentage Total of GDP Per Assistance gence tiveness Cooperation (%) (%) (%) Capita (Billion (Euros) Euro), of Which: 26.69

97.0

1.6

1.4

2 627

3.5

3.39 25.31 4.01 6.78 67.28 11.51 4.10 6.67 19.67 175.40

98.5 90.0 98.0 98.4 98.9 94.2 97.5 97.3 97.7 96.8

0 8.0 0 0 0 3.9 0 0 0 1.6

1.5 2.0 2.0 1.6 1.1 1.9 2.5 2.7 2.3 1.6

2 555 2 561 1 751 2 041 1 773 2 102 2 082 901 911 1 930

4.1 3.9 3.9 4.2 3.6 3.9 2.0 4.0 3.2 3.6

Source: Compiled by the authors using data available from The Economist Intelligence Unit.

the top and bottom 25 per cent of regions also increased, from a ratio of 2:1 in EU-15 to a ratio of 3:1 in EU-27, and the average level of GDP per capita was reduced by almost 12 per cent (2004 data). The regional distribution of wealth among the 268 NUTS 2 regions of the EU-27 shows that the regional GDP per capita (in PPS) relative to the EU-27 average ranges from 23.58 per cent in North-East Romania to 302.9 per cent in the UK capital region of Inner London. GDP per capita is substantially lower in the NMS, where this indicator is below 50 per cent of the EU-27 average in most regions (31 out of a total number of 55 regions). A notable exception is Prague (Czech Republic), which is the region with the highest GDP per capita in the NMS (157 per cent of the EU-27 average). Dynamics of Regional Development In the NMS, the growth rate was especially high in the three Baltic States, all having an average annual real GDP growth over 6 per cent (generating an overall growth of 70 per cent in each country), as well as in Poland, Slovakia, Hungary, Slovenia and Cyprus. The latest members – Bulgaria and Romania – experienced long periods of economic decline during the

STIMSON PAGINATION (M2469).indd 190

20/12/2010 15:12

Implications of European Union structural assistance

191

1990s, but recovered afterwards to have significant growth rates, especially in the capital regions that concentrate most of their economic activities. Many of the regions with a low GDP per capita in the NMS are catching up fast. This trend is confirmed by a Eurostat statistical analysis, indicating regional convergence at the EU level based on the values of the Gini coefficient and the coefficient of variation, both weighted by population (Eurostat, 2007). But our own calculations of the Gini Inequality Index clearly show that economic disparities are bigger within the 12 NMS for which this indicator mounts to 0.2286, as compared with the regions in the EU-15 countries, which have an overall Gini Index of only 0.1478. Because of the higher differentials entailed by the last two enlargements of the EU, the Gini Inequality Index has increased substantially reaching a value of 0.2083 for EU-27 (2004 data). Smaller inequalities are found within each individual country (see Table 10.5, column 3). Following significant above-average growth rates in most of the NMS, economic convergence between the regions of the EU-27, significantly improved, with the difference in the ratio of GDP per capita in the most developed region (Inner London) and the least prosperous region (NorthEast Romania) declining from 13.9:1 in 2002 to 12.8:1 both in 2003 and 2004. This downward trend is encouraging although the differential is still substantial. The number of regions with GDP per capita values below 40 per cent of the EU-27 average also decreased from 23 regions in 2002 to 21 regions in 2003 and to 17 in 2004. The NMS are catching up with the EU-27 average at a rate of 0.8 percentage points every year (Eurostat, 2006). A closer look reveals that not all regions have such good evolutions. Although many less developed regions in NMS attained growth rates above the EU-27 average, there are still 15 of the 55 regions in the NMS having disappointingly low dynamics, of less than 2 per cent annually, which is the EU-27 mean growth. All 15 of these regions belong to three NMS, namely Romania, Czech Republic and Bulgaria. Within-country Disparities Deep regional disparities exist even within the countries themselves, as Table 10.5 clearly shows. In 2003, the highest value of regional GDP per capita was more than double compared with its lowest value in 12 EU countries, the broadest inter-regional differences being in the United Kingdom and Belgium, where the ratio between the two extreme values was 3.7:1 and 3.1:1 respectively. The NMS also display comparatively large regional disparities (see Table 10.5, column 3), although the range of values slightly narrowed.

STIMSON PAGINATION (M2469).indd 191

20/12/2010 15:12

192

STIMSON PAGINATION (M2469).indd 192

20/12/2010 15:12

– – – – 2.426 – 2.184 2.734 – 3.054

12 300 19 700 9 800 11 000 13 800 16 400 11 000 7 200 18 300 12 200

1.4 0.7 2.3 3.4 10.1 0.4 38.2 21.6 2.0 5.2

7.7 10.2

Col 4

– – – – 0.173 – 0.109 0.159 – 0.246

0.119 0.155

Col 5

Sources: Compiled by the authors using data available from Eurostat (2007).

1.917 2.628

7 200 16 400

Bulgaria Czech Republic Estonia Cyprus Latvia Lithuania Hungary Malta Poland Romania Slovenia Slovakia

Col 3

Col 2

GDP Per Highest/Lowest Population in Regional Gini Inequality 2004, Capita in 2004, Regional Index for Million Euros PPS GDP Per GDP Inhabitants Capita Per Capita in 2004 in 2004

1/1 – 1/1 1/1 6/7 1/1 16/16 8/8 1/1 3/4

6/6 7/8

Col 6

615 108 1 031 1 379 2 837 81 11 202 – 423 1 544

– 2 404

Col 7

3 404 213 4 531 6 775 22 890 840 66 553 19 213 4 101 10 912

6 674 25 883

Col 8

Indicative Structural and Number of Convergence Cohesion Funds, Allocations for 2004–06, Regions Convergence Million Euros Against Total Objective, (Prices 2004) Number of 2007/13, National Million Euros Regions (Current Prices)

Economic disparities and allocations for Convergence Objective in the new member states

Col 1

Country

Table 10.5

Implications of European Union structural assistance

193

The dynamics of economic development between the regions in one country can diverge almost as widely as it does between regions in different countries. The greatest discrepancy is displayed by Romania, where the GDP per capita in the most dynamic region (namely Bucharest-Ilfov) increased six times more than in the least developed one (namely North-East). The high regional inequalities of growth dynamics within the NMS are largely determined by the strong economic dominance of their capital regions. In all the NMS – and in some of the EU-15 countries – a substantial share of economic activity is concentrated in the capital regions, which usually have the highest GDP per capita. The larger GDP per capita of those capital city regions is mainly the effect of a substantially higher productivity, and it is also due to in-commuting, which provides a larger labour force relative to the inhabitants of the capital region. Implications The new financial perspective for the seven years between 2007 and 2013 is thus for an EU comprising 27 member states that are displaying increased economic inequalities. Disparities in the levels of development in the enlarged EU imply the need for assistance to the least developed regions and member states by means of an appropriate allocation of Structural and Cohesion Funds. As previously mentioned, the new round of cohesion policy will be focused on investment in a limited number of priorities organized around the three main objectives; namely convergence, regional competitiveness and employment and territorial cooperation. The Convergence Objective is designed to diminish the amplitude of the inter-regional disparities, focusing mainly on the least developed regions. Eligible regions for funding under this objective are the current NUTS 2 regions whose GDP per capita (measured in purchasing power parities) is below 75 per cent of the average GDP in EU-25 for the period 2000–02. In EU-27 there are 84 regions in this category belonging to 17 member states. From these, 51 regions (out of a total of 55 regions in NMS) belong to 11 NMS. The overall level of allocations available under the Convergence Objective amounts to 282.8 billion euros, representing 81.5 per cent of the total budget for the EU’s cohesion policy. Within these allocations, 199.3 billion euros are directed to the Convergence Objective aiming to speed up the convergence of the least developed regions, preponderantly belonging to the NMS (see Table 10.5, column 8). The specific level of allocations to each member state is calculated on the basis of relative regional and national prosperity and the unemployment rate of the eligible regions. The resulting indicative allocations of the

STIMSON PAGINATION (M2469).indd 193

20/12/2010 15:12

194

Endogenous regional development

Convergence Objective for the current round of cohesion policy – 2007–13 – reflect the dimensions and development levels of the 51 eligible convergence regions in 11 new member states (see Table 10.2 and Table 10.4). Following the last two enlargements of the EU, the EU-25 average GDP/capita decreased by 8 per cent, making 16 regions ineligible to receive convergence funds and which previously received ‘Objective 1’ funding. Based on the fact that the economic development level in these regions had not really improved, these so-called ‘phasing-out’ regions will still receive transitional funds that amount to 12.5 billion euros. Another 13 regions (with a total of 19 million inhabitants) – the so-called ‘phasing-in’ regions – will receive special financial allocations (10.4 billion euros) due to their former status as ‘Objective 1’ regions.

CAPACITY OF NMS TO ABSORB EU FUNDS As demonstrated earlier in this chapter, the NMS will be by far the most important net beneficiaries of the EU structural assistance funds. The financial transfers are designed to increase the economic and social cohesion among the member states, mostly via enhancing a faster catching-up process of the less developed states and regions in terms of income per capita. This issue is of a particular importance to the NMS, since Structural Funds are more important when the economy is weak, the marginal benefit of an efficient use of Structural Funds being higher in less developed economies (Daianu, 2003). However, there are experts who question the possibility of effective, productive absorption of the substantial financial transfers by the former centralized economies given all their structural, institutional and administrative problems (Kalman, 2002). Moreover, especially in the academic debate, some authors doubt about the ability of fiscal transfers to bring about economic convergence for the current net recipient member states, or, in general, about whether convergence can be achieved and, even if so, whether fiscal transfers are best tools for enhancing convergence (Boldrin and Canova, 2001). In that debate, various convergence concepts (absolute, conditional) have been discussed,5 but given the current options of the EU cohesion policy, our objective here is to evaluate the capacity of the NMS to absorb the large amount of allocated funds. Methodology Used by the EU to Assess NMS Absorption Capacity The question of absorption capacity started to concern the EC at the beginning of the previous financial exercise (2000–06) when the current

STIMSON PAGINATION (M2469).indd 194

20/12/2010 15:12

Implications of European Union structural assistance

195

NMS were preparing their accession to the EU. This concern is about institutional effectiveness. In 2002 the NEI Rotterdam developed a study commissioned by the DG-Regio/DG-Enlargement, which proposed a methodology for evaluating the ‘capacity of the candidate countries to effectively manage the Structural Funds’ (NEI, 2002, p. 1). The methodology analysed the administrative absorption capacity only for the design (of Structural Funds) phase, considering that it was premature to address the other two phases, namely: (1) performance, or the extent to which the Structural Funds have been managed efficiently and effectively; (2) functioning of Structural Funds. The design assessment focused on a series of indicators regarding: ● ● ● ● ●

management; programming; implementation; evaluation and monitoring; financial management and control.

These indicators were calculated for three main components, namely, structure, human resources and systems and tools: 1. 2.

3.

Structure, which refers to the clear assignment of responsibilities and tasks to deal with the Structural Funds; Human resources, which relate to detailed tasks and responsibilities at the levels of preparing job description, the number and qualifications of staff and fulfilling recruitment needs; Systems and tools, which refer to the availability of instruments, methods, guidelines, manuals, procedures, forms, and so forth.

The information provided by the candidate countries followed detailed questionnaires sent out by the Commission in the spring of 2003. As a result of this assessment the main message was that ‘acceding countries need to further strengthen their administrative capacity’ (Press Releases Rapid, 2003). Following that first exercise, experts and researchers employed the same methodology for their own studies and evaluations – Papadopoulos (2003) for ten candidate countries; Horvat (2004) for five countries (Hungary, Czech Republic, Slovakia, Estonia and Slovenia); and Oprescu et al. (2005) for Romania. For example, the study carried out by Oprescu et al. showed a weak capacity in the case of Romania, but at levels comparable with the other five former candidate countries at approximately the same

STIMSON PAGINATION (M2469).indd 195

20/12/2010 15:12

196

Endogenous regional development

Table 10.6

Administrative absorption capacity – design phase: results of the evaluation of the main indicators by country Romania Hungary

Horizontal evaluation Management C (72%) Programming C (52%) Implementation C (53%) Vertical evaluation Structure B (76%) Human resources C (51%) Systems and D (45%) instruments

Czech Rep.

Slovakia Estonia Slovenia

B (87%) B (80%) C (72%)

B (75%) C (63%) B (87%) C (71%) B (80%) D (40%) B (87%) B (80%) C (56%) C (52%) C (68%) C (52%)

B (84%) C (74%) C (60%)

B (79%) B (79%) A (95%) B (74%) C (71%) D (41%) B (82%) C (59%) C (50%) D (40%) C (60%) C (50%)

Notes: A = Strong capacity: system ready for the Structural Funds (at least 90%). B = Sufficient capacity, but weak points should be addressed (75–90% from the maximum score). C = Capacity not sufficient yet, serious weaknesses must be addressed (50–75%). D = Insufficient capacity, there is no base for administrating the Structural Funds. Source: Compiled by the authors using the evaluation by Oprescu et al. (2005) and Horvat (2004).

time before accession, suggesting that the delays could be recovered and its accession at the beginning of 2007 was still possible (see Table 10.6). Subsequent Evaluations of Absorption Capacity After the 2004 accession wave, further new studies have been carried out, concentrating on the challenges that NMS had to face in implementing the structural assistance allocated for 2004–06. What they reveal is discussed in what follows. A study undertaken by McMaster and Bachtler (2005) provides a comparative analysis concerning how the NMS have accomplished three essential functions: ● ● ●

programming and structural assistance; institutional training; the implementation of the funds.

In the programming process the difficulties were generated by the decision-making with regard to the political choices of the strategic

STIMSON PAGINATION (M2469).indd 196

20/12/2010 15:12

Implications of European Union structural assistance

197

development areas and the setting of clear targets and long-term objectives. In most cases short- and medium-term objectives were preferred over long-term ones and simple and direct interventions were preferred to complex ones, able to combine several objectives simultaneously. For example, direct support to enterprises was preferred to setting up services for businesses, or modernizing the existing transport infrastructure to developing combined or alternative transport modes. Regarding the choices between the national and regional dimensions of development, that is, between interventions at national level meant to support general development and economic growth and those aiming at stimulating the endogenous potential of regional and local development, Baleanu (2007) shows that they were largely in favour of the former. While in the EU-15 the regionalization of the Structural Funds management has been carried on for more than one decade, in almost all NMS that joined the EU in 2004 the governments have chosen to use centralized management systems. Regionalization requires the transfer of many programming and implementation responsibilities regarding Structural Funds to the regional authorities, which is not to the advantage of the countries with still weak regional and local administration. For this reason Poland, Czech Republic and Slovakia even gave up the design-specific regional programmes in the programming period 2004–06. Instead they incorporated them into sectoral programmes or into a single national programme (for example, Joint Operational Programme in the Czech Republic). Centralized management can have a negative effect on smooth implementation of the programmes and, as a consequence, it can diminish the absorption speed of the funds, as shown by the low absorption rates recorded. In the autumn of 2006, two years after the programmes had been launched, an average of only one-third of the funds allocated to NMS-8 was used (see Table 10.7). According to the n+2 rule, the EU funds could be spent by the end of 2008, based on a system of annual reallocation, so that the final results regarding the absorption rate will depend to a great extent on the ability of administration to strengthen its institutional capacity. So far the current situation shows that after accession the NMS administrations did not maintain the pace of reform, which resulted in a low ability to manage public funds. The implementation of advanced human resource management systems was in general limited and inconsistent. A survey conducted by the World Bank (2006) has shown that innovation is still isolated in public management: the administrative function of general coordination is at much lower standards compared with advanced countries, being unable to keep up with the requirements, politicization has reappeared in

STIMSON PAGINATION (M2469).indd 197

20/12/2010 15:12

198

Table 10.7

Endogenous regional development

Absorption rate in NMS-8 between May 2004 and September 2006

Country Czech Republic Estonia Hungary Latvia Lithuania Poland Slovakia Slovenia

Absorption rate (%) 26 29 32.5 25 25.5 24.5 27.5 34

Source: Compiled by the authors using data available from the European Commission, as quoted by Baleanu (2007).

administration, new payment and incentive systems for civil servants are not put into practice. At the same time, the experience regarding partnership has been difficult to assess. This is a major principle in the management of Structural Funds in order to be able to increase the effectiveness of the programmes and to achieve the commitment of actors involved in various stages of the programming cycle, as well as to create good practices in administration. In many cases it has demonstrated that the participation is not authentic, but mimic and formal, with negative consequences on the partners’ commitment and on assuming the ownership of projects’ results (see Baleanu, 2007). In order to reinforce the administrative reforms in the NMS and in the old MS whose administrations still do not function at the required level, the EC has introduced in the programming of Structural Funds for 2007–13 a priority concentrating on the modernization of public service. This is financed by the European Social Fund. The priority aims to stimulate good governance practices and to strengthen the capacity of administrations to meet the requirements for planning and implementing development plans and for increasing the administrative effectiveness of public service at national, regional and local level. Supposing that all these measures will lead to a better administrative capacity, and hence a high rate of absorption of the EU funds, a further question relates to the ‘demand side’ and impact of Structural Funds on the economic and social welfare in the recipient state. In other words, what are the effects of Structural Funds on economic growth and in achieving real convergence? Considering the original idea behind regional policy

STIMSON PAGINATION (M2469).indd 198

20/12/2010 15:12

Implications of European Union structural assistance

199

in which economic convergence was an ultimate goal, the input-oriented approach based on administrative capability does not seem to be sufficient. Impact of Structural Funds Regarding the impact of Structural Funds, a series of so-called ‘absorption problems’ have been identified and require careful consideration on behalf of policy-makers. They are pertaining to large-scale fiscal transfers that can emerge for various reasons and can be significant in preventing the economy from achieving its optimal growth path (Hervé and Holzmann, 1998; Kalman, 2002). According to Kalman (2002, pp. 5–9) they can be summarized as follows: 1. Administrative absorption problems. They result in a difference between transfers and the increase in the productive capital. For any given administrative capability there is a ceiling of absorption capacity and therefore the authors suggest that transfers should only be phased in gradually, starting from a low level and adjusted upwards. 2. Rent-seeking. This phenomenon refers to the people who interfere for the use of funds in purposes that afterwards lead to gaining personal advantages by sharing the benefits resulting from the newly created economic activities (Myrdal, 1972; Krueger, 1974). Rent-seeking becomes manifest through external forms of corruption like bribery, money laundering, traffic of influence and goods, black market, and so on. In the EU context rent-seeking might appear at three levels: between national governments and Brussels, between central and local governments, between any government and private sector agents who benefit from regional policy (e.g., consultants, supported SMEs, construction companies, etc.). The forms of rent-seeking range from legal lobbying activities to illegal forms such as bribery. 3. Use of funds for consumption instead of investment. As a result of external funds injection, domestic investment financing may well be reduced, which means that, unless effective constraints on the use of external funds are imposed, the impact of transfers on capital accumulation and growth will be lower than expected. Several empirical studies (e.g., Hervé and Holzmann, 1998) in development economics confirm that a substantial part of foreign aid is in fact directed to increasing consumption. 4. Timing-related problems. The long-term focused public investment and infrastructure development-related decisions have significant opportunity costs in the short run, such as delays in private investment

STIMSON PAGINATION (M2469).indd 199

20/12/2010 15:12

200

5.

6.

7.

Endogenous regional development

decisions owing to increased uncertainty or modified expectations, or private investment even being crowded out by the public sector. Information disadvantage of the transfer-generating authority/ ‘principal agent’ problems. Agents (private or public, eligible for structural funding) who propose projects have better information on local conditions, thus on the expected private and social rates of returns of the project than the allocating principals. Hence, they may try to reap various advantages whereas the principal is not able to correct the information disadvantage or may do it only at very high costs. Multiple priorities leading to sub-optimal choice. For example, when economic growth is not the sole priority of the recipient country or regional government, other considerations being followed as well, such as equity or fiscal expansion for re-election purposes, they might lead to non-optimal outcomes, such as selection of sub-optimal investment projects either deliberately or not. Problems resulting from relative price changes induced by transfers. Two well-known examples are the ‘Dutch disease’6 and immiserizing growth phenomena. The Dutch disease refers to the case when transfers through excess demand effects in the non-tradable sector (e.g., construction) lead to upward pressure on the overall wage and price level (inflation) and eventually determine a decline of the tradable sectors. Immiserizing growth occurs when economic distortions influence the industrial structure in tradable sectors and transfers provide benefits mostly to protected sub-sectors. An industrial restructuring towards protected sectors might occur, which may be harmful for the overall growth path of the economy and some backward regions may become further disadvantaged.

CONCLUSION In order to support the proper functioning of the single market and also to ensure solidarity among its members, the EU cohesion policy has an overall objective to stimulate the process of reducing the disparities between states and between regions via the so-called convergence process. Those disparities have significantly increased since the 2004 and 2007 accession waves. The structural financial assistance associated with the cohesion policy plays a central role in this process, which has a special significance to the NMS as the main net beneficiaries of the financial transfers.

STIMSON PAGINATION (M2469).indd 200

20/12/2010 15:12

Implications of European Union structural assistance

201

Without neglecting the importance of the amount of funds allocated, it is even more important how these funds are used by the recipient state. The absorption rate does matter, but at the same time the qualitative aspects of the impact of structural assistance matter as well. Overall the following conclusions may be drawn: (1) On the supply side, the success of Structural Funds-based programmes is conditioned to a great extent by the quality of public administration. The higher the quality, the higher the impact of Structural Funds on the economic and social welfare in the recipient state. (2) On the demand side, the scope of public benefits is conditioned by the way the funds are employed: if they are invested in viable projects, with big value-added and significant multiplier effects, the impact of the funds will be also important. The NMS should learn from the experience of the countries that have successfully used the structural assistance that they should have open economies, solid internal public policies and administrations able to implement it.

NOTES 1. The ‘financial chapters’ refer to Common Agriculture Policy, Structural and Cohesion Funds and financial obligations of a particular country to the EU budget. 2. These are regions where GDP per capita would be below 75 per cent of the EU-15. 3. These regions are distinguished by their low population density and considerable distance from mainland Europe. There are seven ‘outermost regions’: Guadeloupe, French Guiana, Martinique and Réunion (the four French overseas departments), the Canaries (Spain) and the Azores and Madeira (Portugal). 4. Even if according to the Berlin Summit (Council Regulation (EC) No. 1260/1999, Art. 7, 8) the upper limit for Structural and Cohesion Funds was set up at 4 per cent, for the 2007–13 period, the European Council (Dec. 2005) decided that the maximum level of transfers towards individual member states had to be reduced and the upper limit has been established between 3.71 and 3.2 per cent (and below) depending on the GNI per head. 5. In this case the debate is around the possibility of market forces – let alone – to lead to the convergence of income in the long run. 6. This refers to the effects of huge positive income shocks for the Netherlands in the 1970s.

REFERENCES Baleanu, A. (2007), ‘The Impact of Structural Funds – Qualitative Aspects’, Working Papers Series, No. 20, September, European Institute of Romania, Bucharest. Berlin European Council (1999), Council Regulations, No. 1260/1999. Boldrin, M. and Canova, F. (2001), ‘Inequality and Convergence: Reconsidering EU Regional Policies’, Economic Policy, 16(32), 207–53.

STIMSON PAGINATION (M2469).indd 201

20/12/2010 15:12

202

Endogenous regional development

Copenhagen European Council (2002), Presidency Conclusions, December 2002. Daianu, D. (2003), ‘Absorption Capacity of EU Funds Plays a Critical Role for Membership Seeking Countries’, Southeast European Times, Bucharest, 26 May. Euractiv (2007), The New EU Cohesion Policy, available at: http://www.euractiv. com/en/regional-policy/new-eu-cohesion-policy-2007–2013-archived/article131988; accessed 26 May, 2010. European Commision (2006), Fourth Progress Report on Cohesion: The Growth and Jobs Strategy and the Reform of European Cohesion Policy – Summary, available at: http://europa.eu/legislation_summaries/regional_policy/review_and_future/ g24238_en.htm; accessed 15 June 2010. Eurostat (2006, 2007, 2008), Regional Yearbook, 2006, 2007 and 2008. Hervé, Yves and Holzmann, Robert (1998), Fiscal Transfers and Economic Convergence in the EU: An Analysis of Absorption Problems and Evaluation of Literature, Nomos, Baden-Baden. Horvat, A. (2004), Absorption Problems in the EU Structural Funds. Some Aspects Regarding Administrative Absorption Capacity in the Czech Republic, Estonia, Hungary, Slovakia and Slovenia, National Agency for Regional Development of Slovenia. Hubner, D. (2007), ‘Future of Cohesion Policy: Towards Exploring Options’, Informal Ministerial Meeting on Territorial Cohesion and Regional Policy, Ponta Delgada, Azores, Portugal, 24 November 2007, available at: http:// europa.eu/rapid/pressReleasesAction.do?reference=SPEECH/07/742&format= HTML&aged=1&language=EN&guiLanguage=en; accessed 15 June 2010 Kalman, J. (2002), Possible Structural Funds Absorption Problems – The Political Economy View with Application to the Hungarian Regional Development Institutions and Financial System, available at: http://lgi.osi.hu/publications/2002/105/Marcou-Hungary.pdf; accessed 26 May 2010. Kasman, A., Kasman, S. and Turgutlu, E. (2005), ‘Nominal and Real Convergence Between the CEE Countries and the EU: A Fractional Cointegration Analysis’, Applied Economics, 37(21), 2487–500. Krueger, A.O. (1974), ‘The Political Economy of the Rent-seeking Society’, The American Economic Review, 46(3), 291–303. Leonardi, R. (2006), ‘Cohesion in the European Union’, Regional Studies, 40(2), 155–66. McMaster, I. and Bachtler, J. (2005), ‘Implementing Structural Funds in the New Member States: Ten Policy Challenges’, available at: http://www.eprc. strath.ac.uk/eprc/Documents/pdf_files/12A07_McMaster-Bachtler_paper.pdf, European Policies Research Center; accessed 26 May 2010. Myrdal, Gunar (1972), Asian Drama: An Inquiry Into the Poverty of Nations, Random House, New York. NEI (2002), ‘Key Indicators for Candidate Countries to Effectively Manage the Structural Funds: Final Report’, available at: http://www.evaluace.cz/doku menty/hodnot_zpr_eu/souhrnna_studie.pdf, prepared by the NEI Regional and Urban Development for the EC-DG REGIO/DG-ENLARGEMENT, Rotterdam; accessed 26 May 2010. Oprescu, G., Constantin, D.L., Pislaru, D. and Ilie, F. (2005), Analysis of Absorption Capacity of the EU Funds in Romania, The European Institute in Romania, Bucharest. Papadopoulos, A. (2003), Administrative Capacity Study – Phare Region, Phase Two. Country Reports, European Commission, March.

STIMSON PAGINATION (M2469).indd 202

20/12/2010 15:12

Implications of European Union structural assistance

203

Press Releases Rapid (2003), ‘Structural and Cohesion Funds: Acceding Countries Need to Further Strengthen their Administrative Capacity’, available at: http:// europa.eu/rapid/pressReleasesAction.do?reference=IP/03/103i&format=HTM L&aged=0&language=EN&guiLanguage=en; accessed 26 May 2010. Brussels, July. World Bank (2006), EU-8. Administrative Capacity in the New Member States: The Limits of Innovation, September.

STIMSON PAGINATION (M2469).indd 203

20/12/2010 15:12

11.

Macroeconomic and territorial policies for regional competitiveness: theory and empirical evidence from the EU Roberto Camagni and Roberta Capello

INTRODUCTION In this chapter we examine the new required policy targets and styles for assuring regional performance and competitiveness for the European Union (EU) and provide an empirical analysis of territorial policies for regional competitiveness, with a focus on investigating territorial capital and regional growth. We first address the theoretical basis of the importance of regional policies. The strongest argument in favour of regional policies lies in the long-term persistence and even widening of inter-regional disparities. The starting point in that theoretical reasoning is that regional policies are fundamental for the competitiveness of regional economies. Unlike countries, regions compete on the basis of an ‘absolute’ advantage principle, and, whenever non-competitive, regions cannot rely on any automatic mechanism in order to maintain some export specialization. Thus, in the case of the EU, the fate of regions is mass unemployment and, in the case of insufficient public income transfers, emigration and possibly desertification. In this chapter we stress the idea that the factors determining regional performance do not lie just in each region’s internal development capability. In fact, among the causes of regional success and failure, one can find, on the one hand, some pervasive characteristics of the national economy and, on the other hand, its general performance. This aspect, linked to the fact that national policies are not space-invariant within each country, brings us to the conclusion that national policies also can explain regional performance. That includes interest rate policies, monetary and fiscal policies driving movements of the exchange rate, and also such policies such as transportation and TENs (Trans-European Network) policies, policies

204

STIMSON PAGINATION (M2469).indd 204

20/12/2010 15:12

Macroeconomic and territorial policies

205

for R&D and agricultural policies. All such policies bring selective effects to different typologies of regions due to a wide array of transmission channels or preferential regional targets, and, therefore will strongly contribute to the economic performance of regions. Of course beyond the national component, the crucial component of regional performance is each region’s internal development capability. These policies are discussed based on a new concept, that of ‘territorial capital’, and that is analysed through a historical development perspective of policy targets and styles. Thanks to the existence of a regional growth forecasting model called MASST built by the authors (Capello, 2007; Capello et al., 2008), we also present simulations on the following: ●

● ●

The impact of macroeconomic policies at regional level in the EU. As the simulation will demonstrate, they are not space-invariant, as some regions gain more than others when, for example, a devaluation policy is pursued. The impact of European regional policies, like structural funds. The role of specific elements of territorial capital on regional growth patterns, on which some policies could be built.

PERSISTENCE OF REGIONAL DISPARITIES AS THE STRONGEST RATIONALE FOR REGIONAL POLICIES The strongest argument in favour of regional policies lies in the longterm persistence and even a widening of inter-regional disparities. In fact, the history of the entire European integration process is characterized apparently by slowly decreasing overall inter-regional disparities; but this process, in reality, results generally from strong processes of catching-up among nations in the presence of increasing disparities at the intranational level. These last trends of increasing disequilibria inside single countries may show a different intensity over time. The Williamson Law (Williamson, 1965) expects them to be naturally stronger in the early stage of the development or integration process, while in a later stage they slow down or may even reverse. This result was shown as empirically correct if interpreted inside the great macro-phases of European integration (Camagni and Gibelli, 1996): (1) increasing disparities in the first development and integration phase during the 1950s up to the mid-1960s, followed by a reduction until 1980, mainly thanks to the catching-up process of a new

STIMSON PAGINATION (M2469).indd 205

20/12/2010 15:12

206

Endogenous regional development

category of intermediate-income, ‘third’ regions like the ‘Third Italy’, Flanders, southern Germany, southern France and East Anglia; (2) an upswing of disparities during the industrial restructuring processes of the 1980s, the new integration phase of the Single Market and the path towards the single currency, up to 1995–2000, and still no clear evidence afterwards. Explanations for the Difficulties Encountered by Peripheral Regions The theoretical explanations of the causes of the difficulties encountered by less advanced or peripheral regions in their path towards integration and development may be classified into four main classes, which are discussed below: 1.

2.

Unlike countries, regions are not subject to the principle of ‘comparative advantage’ governing international specialization and trade, attributing each partner country some specialization sectors and a condition of full employment. The reason for this resides in the fact that the two equilibrating forces that in principle allow passing from an ‘absolute advantage’ to a ‘comparative advantage’ regime – namely price flexibility and currency devaluation, fully active in the case of countries – either do not work properly or do not exist at the interregional level.1 Therefore, regions are not granted some productive specialization and some role within the spatial division of labour whenever they prove less competitive in all production sectors with respect to the external territories (Camagni, 1992, 2002). Regions in fact compete on the basis of an ‘absolute’ advantage principle, and, whenever non-competitive, they cannot rely on any automatic mechanism in order to maintain some export specialization; their fate is, in this case, mass unemployment and, in the case of insufficient public income transfers, emigration and possibly desertification (GREMI, 1995). For these territories, the possible strategy for development (or survival) is threefold: complete autarchy (almost impossible); lobbying for public income transfers (to be rejected flatly), or improving competitiveness of some export sector; and attracting investments from other regions and from abroad.2 As a consequence, in the case of regions, cities and localities, caring about competitiveness and attractiveness is perfectly legitimate, and the criticisms that Paul Krugman recently addressed to such concerns (Krugman, 1996, 1998), labelling them as ‘wrong’ and ‘misleading’, do not apply in their case.3 The second argument refers to the fact that, even in the case that lagging regions could supply lower wages than advanced regions, their

STIMSON PAGINATION (M2469).indd 206

20/12/2010 15:12

Macroeconomic and territorial policies

3.

4.

207

advantage would be limited by the lower levels of labour productivity. In fact, what matters to firms in choosing their location are unit labour costs or ‘efficiency wages’ (wages/productivity): in advanced regions, a high productivity generated by a strong industrial culture, efficient services and good infrastructure may well out-weigh the disadvantage of higher salaries and generate a continuing external competitiveness, as in the well-known Dixon-Thirwall model (1975). All this leads to the further argument that, in absence of some basic ‘preconditions’, namely in terms of infrastructure, accessibility, general education and basic public services, growth could never start in a region, given the strong locational advantage of the other competitor sites (Rosenstein-Rodan, 1943; Armstrong and Taylor, 1993). The fourth argument, dynamic in nature, regards the cumulative nature of economic growth, due to: –

inter-sectoral relationships and forward/backward linkages in the general economy and along the main production filières (Hirschman, 1957; Krugman and Venables, 1996); – increasing returns to scale, when the expanding size of the local production fabric generates increasing productivity (once called the ‘Verdoorn Law’, utilized by Kaldor, 1970, and more recently rediscovered by Krugman, 1991); – demand–supply interaction on the goods and labour markets, which generates a cumulative development process in core regions (investment ⇒ development ⇒ in-migration ⇒ new local demand ⇒ new investment ⇒ new development) (Myrdal, 1957); – technological change, both ‘embodied’ in new machinery or ‘endogenous’ – in the sense of the cumulative expansion of local know-how and knowledge through learning-by-doing processes à la Arrow. These effects taking place in already developed regions are a measure to overcome the trend towards decreasing capital productivity in these regions (a consequence of the expanding capital stock), and to counterbalance the dispersion of economic activities expected by the simplified neoclassical model of regional growth of the early 1960s (Borts and Stein, 1964).4 The above arguments look strong enough to disprove the optimistic view about inter-regional convergence processes, mainly expressed by the neoclassical school, both in its traditional and modern modelling approaches.

STIMSON PAGINATION (M2469).indd 207

20/12/2010 15:12

208

Endogenous regional development

A Regional Convergence Process Might be Expected but Why Does it Not Occur? A convergence process is expected by the neoclassical school as a consequence of many factors. These include the following: ●

●

●

An out-migration process of unemployed people from lagging regions: this process takes place in fact, and statistically reduces percapita income disparities – as it reduces the denominator of the percapita income indicator in these regions. One may question though whether this can be considered a true convergence process, as it adds human resources in already developed regions reducing the inflationary effects of tight local labour markets, subtracting them to lagging regions, and whether inter-regional and international mass migrations could be still politically acceptable, at least in European countries. An outflow of capital from advanced regions, as a consequence of the diminishing returns showing up along with the capital accumulation process; this event may be easily counterbalanced by learning processes and cumulative technological change. An outflow of technology from advanced regions towards lagging ones. This argument suffers from the consideration of single technologies or technological paradigms, which for sure, once developed and introduced in some core localities, diffuse afterwards on the territory. But what about the overall pace of technological creation in the form of subsequent improvements to the single technology, development of integrated technological systems and new inventions leading to new technologies? And what about the capability of developing new applications of existing technologies? The pace of this technological creation process in core regions is likely to be faster than the pace of technology diffusion and catching-up by lagging regions, due to all sorts of learning processes, both internal to the firm or taking place in a ‘collective’ way on the territory (Camagni and Capello, 2000; Keeble and Wilkinson, 2000; Camagni, 2001).

Even modern neoclassical conditional convergence theory (Barro and Sala-i-Martin, 1991, 1992), which still expects diffusion processes and re-equilibrating processes (a sort of ‘entropic trend’ towards spatial homogeneity), acknowledges that these processes are ‘conditioned’ by other, opposite, endogenous or exogenous forces (industrial structure, propensity to save, institutional innovations like the creation of

STIMSON PAGINATION (M2469).indd 208

20/12/2010 15:12

Macroeconomic and territorial policies

209

custom unions, regionally asymmetric shocks, etc.), which diversify the catching-up capability of the single regions and are often a measure to postpone or overturn the process of convergence (Armstrong, 2002). But a second aspect, regarding the social costs of non-intervention in a context of increasing disparities and inter-regional competition (on the basis of an ‘absolute advantage’ principle), provides even clearer support to spatial development policies. A strategy of non-intervention in fact presents the following drawbacks: ● ●

●

●

●

huge social and political costs allowing the explosion of regional crises and the cultural and environmental costs of regional desertification; the risk of a super-concentration of population in the big urban areas of lagging regions (a phenomenon that is typical of developing countries), as a consequence of the crisis of the surrounding areas and not of the attractiveness of these urban areas, of a push and not of a pull factor; the high opportunity cost of adding successful activities in already successful areas: in a context of full employment, new workers for new activities are found at the expense of existing activities, while in weak areas they are drawn from the unemployment reservoir, and their opportunity cost is close to zero; the channelling of a wide share of national savings towards the building and construction industry and real estate speculation in advanced regions and cities, as a consequence of the migration processes, subtracting it from more productive uses; a lower exploitation of the creativity potential of all regional communities, constrained by the presence of some basic locational disadvantage (accessibility, services, infrastructure, and so on).

All the previous arguments may be considered, by and large, as variations on the theme of market failures. But other kinds of failures are invoked by the neo-liberal opponents of any policy intervention, namely: (1) government failures, or the unintentional drawbacks of public policies; (2) the reduction of the competitive climate and the likely formation of new social classes, specialized in incentive intermediation and lobbying, can reduce the reaction capability of the local entrepreneurship and the local economy, which is the well-known argument put forward by the public choice school concerning the inability of bureaucracy to substitute for the market signals and its orientation towards self-referential goals instead of the public good. All these arguments for which there are numerous confirmations in the case of long-term lagging regions, supposedly call for a substantial

STIMSON PAGINATION (M2469).indd 209

20/12/2010 15:12

210

Endogenous regional development

reduction in the role of governments – regional, national, European – even in the presence of the social costs of continuing territorial disparities (Buchanan and Tullock, 1982; Pennington, 1999). A possible answer in this case concerns the need – and the actual possibility – of providing the necessary countervailing powers and political control tools, coming from transparent, democratic and participatory practices and procedures. Many recent positive examples show that policy failures are not the necessary outcome, while the failures of the market mechanism are in-built into its constituency and operating rules. Spatial Development Policies Derived from the Globalization Debate An element that recently contributed to lowering the general attention to re-equilibrium issues and spatial development policies derived from the globalization debate. If, in the new integrated context, competitiveness is the main issue, and if champion firms and territories act as driving forces for the entire territorial system, a proper spatial policy – it is argued – should care more about strong than weak territories, about winners rather than losers, and should therefore concentrate investment and innovation (e.g., public expenditure in advanced infrastructure and human capital) on core regions and big global cities. Two answers may be given to this argument, referring to two different territorial levels, namely, the inter-regional and the intra-regional. As far as the inter-regional level is concerned, a distinction has to be made between the provision of advanced assets and infrastructure – that has to be secured to all regions and to advanced regions in particular – and the public support to the financing of this provision, which should be inversely proportional to the capability of each region to provide the same assets through a private financial procedure (e.g., through project financing). Regarding the second level, the suggestion coming from the previous argument can be acceptable: when intervening with public development policies on limited territories like regions or sub-regions, an important policy rule should be to select places with a maximum development potential (big  cities, specialized medium-sized cities), in order to maximize probability of success and save public money. At the intra-regional level, spread effects expected in the long run from the intervention on ‘champion territories’ would be probably faster and stronger than at the inter-regional level, and the social cost of letting some areas lag behind in the initial stage more acceptable. This issue presents itself very clearly in the case of the 12 new member countries in the EU. There is in fact no doubt about the polarized and regionally concentrated character of the new industrialization and

STIMSON PAGINATION (M2469).indd 210

20/12/2010 15:12

Macroeconomic and territorial policies

211

development phase, triggered by the accession to the EU, something that will probably continue in the next decade, due to cumulative territorial effects. Capital cities and territories located along the border belt with the 15 old member countries are the preferred locations for new activities, both industrial and tertiary, both in the production and in the commercial spheres. Migratory processes from peripheral regions towards core territories; investments linked directly to these migratory trends, in all kinds of ‘residential’ sectors; infrastructure provision that naturally follows the priorities indicated by the demand for mobility and accessibility; all these trends will generate an increase in internal disparities, very similar indeed to what happened in Western European countries during the early phases of integration. Furthermore, political attention is more and more directed to core areas, as they are rightly seen as the true assets in international territorial competition. This situation is nonetheless full of risks. An excessive concentration of economic growth in a few areas is likely to very quickly determine tensions on the local labour markets and land markets, pushing wages and land rents upwards in an unsustainable way, namely well beyond the increase in productivity.5 All this, coupled with the usual indirect effects of congestion, social costs and environmental decay, will soon generate a fall in competitiveness and attractiveness of these areas, jeopardizing the economic future of these countries. As usually happens, the potential trend is not only costly in terms of territorial cohesion and equity, but particularly on the grounds of economic efficiency itself. A wise strategy in these cases will be, on the contrary, to widen the potentially eligible areas for foreign and internal investments, strengthening second-rank cities and city-regions in the national urban hierarchy, and linking them with the capital cities and the big foreign agglomerations and markets through efficient transportation and communication networks. A polycentric urban development, both at the inter-regional and intra-metropolitan level, similar to what is suggested by the European Spatial Development Perspective, could be a proper way for the construction of a consistent spatial strategy, sustainable at the same time in economic, social and environmental terms.

THE NATIONAL COMPONENT OF REGIONAL GROWTH AND THE DIFFERENTIAL SPATIAL EFFECT OF MACROECONOMIC POLICIES The factors determining regional performance do not lie just in each region’s internal development capability. In fact, among the causes of

STIMSON PAGINATION (M2469).indd 211

20/12/2010 15:12

212

Endogenous regional development

regional success and failure, one can find, on the one hand, some pervasive characteristics of the national economy and, on the other hand, its general performance. The former elements refer to: (1) institutional factors like the performance of the high functions of the nation state – legislative, justice and government functions – to organizational factors like the efficiency of services of general interest like education, transportation, communication, health and security services; (2) economic factors like general fiscal pressure, effectiveness of public expenditure, pervasiveness of environmental regulations and efficiency of contract enforcement procedures, general price-competitiveness in case of less advanced countries. The second element linking regional economies to the general performance of the national economy is represented by the high inter-regional, within-countries integration, relative to international integration, in terms of exchange of goods, services and production factors, due to proximity effects and absence of any kind of institutional or linguistic barrier. All this is reflected in the empirical evidence in the EU. By and large, the variance of regional growth rates within countries is lower than the variance of international growth rates, all regions benefiting from a good short-term and long-term performance of their national economy.6 This second element looks even more relevant if interpreted in the light of the goal of reducing inter-regional disequilibria. As illustrated in Figure 11.1, the results of the MASST simulations and foresights on European regions (see later in the chapter) show that, between now and 2015, disparities in the EU-27 will decrease only thanks to the national growth component, namely to fast processes on international catching-up (mainly by new member states), with a slight increase in intra-national disparities (see Figure 11.1a).7 In conclusion, if aggregate, national development represents an important part of regional development, then it follows that first class territorial development policies reside in sound and consistent policies internal to each country, addressed towards a pervasive effectiveness of the public administration and provision of public goods and externalities enhancing the development capability of all local economies. A further element, linked to the previous one, that is worth an indepth inspection is the impact of macroeconomic and structural policies managed at the national level in determining regional performance. In this case, going beyond the open question of the general effectiveness of these policies, it looks sound to expect highly differentiated effects on the different regions. In fact, interest rate policies, monetary and fiscal policies driving movements of the exchange rate, but also such policies as transportation and TENs policies, excellence policies in R&D, agricultural

STIMSON PAGINATION (M2469).indd 212

20/12/2010 15:12

Macroeconomic and territorial policies

213

(a) EU-27 countries

0.25

0.2

0.15

0.1

0.05

0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Total (EU-27)

Between countries (EU-27)

Within countries (EU-27)

(b) 15 old and 12 new countries

0.25 0.2 0.15 0.1 0.05

0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Total (old 15) Between countries (old 15) Within countries (old 15)

Total (new 12) Between countries (new 12) Within countries (new 12)

Source: The authors using simulations through the MASST model.

Figure 11.1

The evolution of per-capita income disparities in the EU, 2002–15

policies, bring selective effects to different typologies of regions, due to a wide array of transmission channels or preferential regional targets. The case of agricultural policies is easy to understand, bringing direct support to rural regions that, at present, are often among the richest, at least in many of the 15 old member countries (Arkleton Institute, 2005). But also other policies have selective effects: TENs and excellence policies,

STIMSON PAGINATION (M2469).indd 213

20/12/2010 15:12

214

Endogenous regional development

for example, are naturally biased in favour of stronger regions, where most of the favourable preconditions for successful competitiveness policies are present and where the demand for new transportation infrastructure is higher and guarantees the highest economic return to investment. Another typology of national policies, concerning the macroeconomic sphere, is usually considered as neutral in terms of inter-regional disparities. On the contrary, selective effects are visible also in this case, due to inter-sectoral reasons. In fact, a rise in interest rates is likely to hit most regions specialized in manufacturing and building and construction rather than tertiary regions; a similar effect may be expected as a consequence of a real revaluation of the currency, generating wider tensions in industries and regions characterized by a wider degree of openness to international trade. Furthermore, a revaluation process is likely to hit more, ceteris paribus, those regions specialized in labour-intensive industries, as it raises labour costs expressed in international currency (while the cost of capital will remain unchanged, at the level determined internationally, especially if the revaluation process is interpreted as the effect of a strong and potentially fast-growing economy).

THE ENDOGENOUS COMPONENT OF REGIONAL GROWTH: THE CONCEPT OF TERRITORIAL CAPITAL Beyond the national component, of course the crucial component of regional performance is each region’s internal development capability. In the interpretation of this endogenous capability, spatial economic theory has shown a triple paradigm shift in the last few decades, namely: ● ●

●

from development (or even location) factors to innovation factors (Nijkamp, 1986); from hard to soft factors, residing either in intangible, atmospheretype, local synergy and governance factors (Becattini, 1990; Camagni, 1991), or in human capital and knowledge assets (Foray, 2000); from a functional approach to a cognitive approach.

The last and more recent trend deserves some inspection. A cognitive approach is increasingly superseding the traditional functional approach to show that cause–effect, deterministic relationships should give way to other kinds of complex, inter-subjective relationships that impinge on the way economic agents perceive economic reality, are receptive to external

STIMSON PAGINATION (M2469).indd 214

20/12/2010 15:12

Macroeconomic and territorial policies

215

Functional approach (substantive rationality)

Infrastructure Factor endowment Industrial structure Financial incentives

Human capital Competences Education and training Quality of life

Development factors

Innovation factors Connectivity Interconnection Openness ‘Go-global’

Knowledge, identity Receptivity Milieu effect Collective learning

Cognitive approach (procedural rationality) Source:

The authors.

Figure 11.2

Paradigm shifts in regional development factors (and policies)

stimuli, can react creatively and are able to cooperate and work synergetically. Local competitiveness is interpreted as residing in cooperation, trust and sense of belonging rather than in pure availability of capital; in creativity rather than in the pure presence of skilled labour; in receptivity to new business ideas and organizational styles more than in the presence of SMEs per se; in connectivity and relationality more than in pure accessibility; in local identity as well as local efficiency and quality of life (Figure 11.2). The theoretical elements that support the new methodological approach may be found in the following: ●

●

The theory of bounded rationality and decision-making under conditions of uncertainty, from the seminal contributions of Simon (1972) to their application to industrial innovation (Dosi, 1982; Nelson and Winter, 1982). The institutional approach to economic theory based on a ‘theory of contracts’, which emphasizes the importance of rules and behavioural codes, and of institutions that ‘embed transactions in more protective governance structures’ (Williamson, 2002, p. 439),

STIMSON PAGINATION (M2469).indd 215

20/12/2010 15:12

216

●

Endogenous regional development

reducing conflicts and allowing mutual advantages to be gained from exchange. The cognitive approach to district economies and synergies, which comprises the Italian school (Becattini, 1990), the French ‘proximity’ approach (Gilly and Torre, 2000), the GREMI approach to local innovative environments (Camagni and Maillat, 2006), and Michael Storper’s concept of ‘untraded interdependencies’ (Storper, 1995). The GREMI group conceives proximity space or the local ‘milieu’ as an uncertainty-reducing operator that works through socialized transcoding of information, cooperation enhancing and the supply of the cognitive substrate – represented mainly by the local labour market – in which processes of collective learning are embedded (Camagni, 1991; Capello, 2001).

All the above elements – which add to, and do not substitute for, more traditional, material and functional approaches – may be encompassed and summarized by a concept that, strangely enough, has only recently made its appearance, and has done so outside a strictly scientific context: the concept of ‘territorial capital’. This was first proposed in a regional policy context by the OECD in its Territorial Outlook (OECD, 2001), and it has been recently reiterated by DG Regio of the Commission of the European Union: Each Region has a specific ‘territorial capital’ that is distinct from that of other areas and generates a higher return for specific kinds of investments than for others, since these are better suited to the area and use its assets and potential more effectively. Territorial development policies (policies with a territorial approach to development) should first and foremost help areas to develop their territorial capital. (CEC, 2005a, p. 1)

In our view territorial capital may be seen as the set of localized assets – natural, human, artificial, organizational, relational and cognitive – that constitute the competitive potential of a given territory. In this very large sense it encompasses (Camagni, 2008; Camagni and Capello, forthcoming): ● ● ● ● ● ●

natural resources and social overhead capital; impure public goods or mixed public/private (p/p) goods (landscape, cultural heritage); agglomeration and district externalities; club goods such as proprietary networks; private fixed capital stock and relational private services; social and relational capital;

STIMSON PAGINATION (M2469).indd 216

20/12/2010 15:12

Macroeconomic and territorial policies ● ● ●

217

human capital, entrepreneurship, creativity and leadership; cooperation networks and strategic p/p partnerships in knowledge creation; governance structures.

It is evident that the concept of territorial capital encompasses not just immaterial and relational assets, but also material ones, in the form of natural and cultural resources, public goods and general urban structure. In this second respect, it helps us understand the role of a well-shaped geographical structure and form of settlements for both the efficiency-competitiveness of territories and the general welfare conditions of populations. In this sense, ‘integrated spatial/urban development policies’ were recently indicated by OECD (2001) and the EU (CEC, 2005b; Ministers for Spatial Planning, 2007) as the consistent new policy approach, integrating economy and territory and targeting the best utilization of the economic potential and the territorial capital of each local economy. The concept comes close to the French tradition of aménagement du territoire, with its integrated and wide-area approach to spatial development.

RENEWED REGIONAL POLICIES FOR REGIONAL COMPETITIVENESS As stated before, the new scenario in which territorial systems compete and act nowadays is a scenario of increasing complexity and uncertainty. Decision-making centres have to manage increasingly interdependent and integrated systems, globally stable but locally fragile; logical causal chains present wide and embarrassing holes; the behavioural logics of the different actors show increasing instability, limiting forecasting possibilities; transformations in production and consumption habits appear faster and faster; available information grows exponentially, but our capability of selection, interpretation, transcoding and evaluation does not grow in a parallel way. As a consequence, our cognitive models evolve in new directions, requiring new rationalities in the decisionmaking process. The logics founded upon a ‘substantive’ rationality, widely utilized in economic thinking, appear inadequate, as do the pretentions of the socalled ‘rational-comprehensive’ planning models of the recent past. But all this does not question the very existence and necessity of spatial policies, as some modern approaches seem to imply; it only requires a new type of rationality, acknowledging its limitations and fully assuming uncertainty

STIMSON PAGINATION (M2469).indd 217

20/12/2010 15:12

218

Endogenous regional development

as a normal condition in cognitive processes. Decision-making procedures shall be characterized by flexibility, partnership, iterative processes of decision-monitoring-assessment-decision, precautionary and ‘no regret’ principles as far as environmental matter is concerned and search for consensus by the concerned communities. Even if a ‘strong’ rationality is useless, we still need some form of rationality in spatial policies: we still need to find good solutions, to evaluate the effects of past decisions, to provide consistency and compatibility among different decisions (about projects and schemes that are frequently managed separately), to assess the likely impact of single projects not just in terms of their direct effect on employment, income, mobility, environmental quality, but on the general long-term competitiveness and liveability of territories.8 During the last half-century, both theoretical reflections and the evolution itself of spatial systems have determined a substantial enrichment of policy strategies and policy tools in spatial development practices, shifting the emphasis successively towards new, more suitable goals and keywords. We can summarize these evolutions as follows: ● ● ● ● ● ●

1950s: infrastructure, as precondition for growth; 1960s: attraction of external activities, development poles, export industry; 1970s: endogenous development: SMEs, local competencies; 1980s : innovation, technological diffusion, innovative milieux; 1990s: knowledge base, intangible factors, local culture; 2000s: ‘relational capital’, collective learning, interconnection, territorial capital.

The general approaches followed in the earliest period, in Europe in particular, fully recognized the ‘balanced’ nature of the economic development process; all preconditions had to be present, or provided, at the same time. Subsequent reflections, typical of the 1960s, were mainly oriented by the sense of urgency and the need to find short-cut solutions to development needs of territories: a ‘big push’ and an ‘unbalanced growth’ strategy was felt to be economically and politically superior. Development poles, ‘cathedrals’ or ‘white elephants’ forced into ‘the desert’ were supposed to bring faster, even if less certain, effects (Parr, 1999). During the 1970s, taking advantage of the results achieved mainly through the diffusion of public infrastructure and services throughout the countries (but also trying to counterbalance the negative effects that in many cases derived from the location of big external plants, separated from the local economy and society), development policies addressed themselves towards the strengthening of the local fabric of SMEs, in a first

STIMSON PAGINATION (M2469).indd 218

20/12/2010 15:12

Macroeconomic and territorial policies

219

stage through generalized incentives and subsequently trying to enhance local synergies and local specializations (the so called ‘development from below’ model: Stöhr, 1990). Subsequently, during the 1980s, both the theorization on the importance of innovation processes for spatial development (Ewers and Wettmann, 1980; Camagni and Cappellin, 1985; Nijkamp, 1986) and that on ‘industrial districts’ (Bagnasco, 1977; Brusco, 1982; Becattini, 1990) and milieux innovateurs developed by the GREMI (Camagni, 1991) helped the understanding of the necessity of valorization of local entrepreneurship, even if weak and of a mainly handicraft nature; of collaboration between the public administrations and the organizations of entrepreneurs, devising common projects and schemes addressed towards the strengthening of the local productive ‘vocation’; the necessity of concentrating in single sites different policy tools, from economic and fiscal incentives to education and training services and infrastructure provision; the necessity of stimulating innovation more than simply efficiency (Camagni, 1995). But the reflection on local milieux also helped in drawing attention once again to the characteristics of the territory: not only to its physical configuration (infrastructure, accessibility, environmental amenities) but especially to its social and relational aspects. Incentives to innovation through interfirm cooperation; provision of local schemes developed in private–public partnership; orientation of public funding towards specialized territorial ‘districts’; all these measures represented policy innovations during the 1980s and 1990s, recently revisited through the new tool represented by the territorial ‘pacts’. Since the mid-1980s, the knowledge base as a source of continuous innovation was highlighted as one of the major factors for long-term growth (Knight, 1984; Castells, 1985; Tatsuno, 1986). Also in this case, different stages were experimented with in policy strategies: interventions addressed directly to the R&D functions of firms; interventions establishing facilities and places devoted to knowledge construction outside firms (technology poles, science parks, etc.), and more recently interventions addressed towards creating a local atmosphere conducive to innovation, through wide training programmes dealing with general education and extensive use of new technologies, inside and outside firms, or building advanced schemes in public/private partnership (Ewers and Allesch, 1990; Cooke and Morgan, 1998). The core issue is increasingly placed on enhancing local creativity (Andersson, 1985), recreating the particular atmosphere felt (in particular periods) in such innovative milieux as Silicon Valley, Orange County, the Swiss Jura and the Italian industrial districts. Along the same logical trajectory, trust (Knack and Keefer, 1997), relational capital (Camagni, 1999; Camagni and Capello, 2000)

STIMSON PAGINATION (M2469).indd 219

20/12/2010 15:12

220

Endogenous regional development

and collective learning (Camagni, 1991; Lundvall and Johnson, 1994; Capello, 1999a and 1999b; Keeble and Wilkinson, 1999) are more recently theorized as the main factors or preconditions for a creative environment that calls for completely new policy styles, still to be fully understood and developed. We have spoken about strategy ‘enrichment’, in the sense that previous goals and keywords were not really abandoned subsequently: rather, the policy vision was enriched and widened, the main limits of previous policies corrected, emphasis shifted to new issues, on a path towards complexity and integration. Spatial development is increasingly understood as a complex, multi-dimensional phenomenon, and the illusion about the existence of simple, short-cut strategies progressively abandoned. As stated before, a general long-term trend is apparent, in the direction of some new target elements for a renewed and modern territorial development policy: ●

●

●

intangible factors, like human capital and knowledge, and the ‘operators’ that could translate their potential in actual growth projects; relational factors, creating synergies, promoting cooperation and partnership, exploiting the richness of local relationships that define a productive ‘vocation’, a local know-how and a local culture: social and relational capital; advanced communication networks and communication services, in order to get a global reach on markets, information, business opportunities: public goods (but also ‘toll goods’ provided by private bodies) addressed towards the efficiency of territory.

But also a change in policy styles is needed, residing in the goal of: ●

● ●

preparing territories for innovation, enhancing their adaptability to a changing external context, promoting their openness and receptivity to new business ideas and organizational styles, rather than forcing the locational decisions of single firms; negotiating the terms for a fruitful cooperation between territories and firms, rather than just supplying favourable location factors; reinterpreting a bottom-up, generative approach to development rather than a top-down, ‘competitive’ one where regions and cities fight among each other for the attraction of a given (and increasingly scarce) amount of public resources and private investments, in a zero-sum game.

STIMSON PAGINATION (M2469).indd 220

20/12/2010 15:12

Macroeconomic and territorial policies

221

EMPIRICAL EVIDENCE FROM THE MASST MODEL: MACROECONOMIC AND REGIONAL POLICIES IN THE EU A Brief Introduction to the MASST Model The econometric model estimating the system of cause–effect relationships – termed the MASST (MAcroeconomic, Sectoral, Social and Territorial) model – is a new one, conceptually defined for the purpose of investigating regional growth, its determinants and its territorial evolution. It draws on the most advanced theories of regional growth, without denying the importance of the achievements accomplished by the traditional theories. MASST explains relative regional growth through territorial and spatial factors such as agglomeration economies, territorial capital and spatial spillovers (i.e., the influence of each region on the growth trajectories of neighbouring regions). These factors determine the cumulative nature of regional growth patterns, as widely emphasized by the new endogenous growth theories and the ‘new economic geography’ rooted in Myrdal’s (1957) and Kaldor’s (1970) cumulative causation theory.9 Social elements (demographic change due to natural population change or migration flows) are included in MASST and have a role in explaining regional growth patterns together with the widely recognized factors of local competitiveness, namely accessibility, presence of human capital and local resource endowment. The MASST model is structured as follows. It comprises two blocks of equations, one explaining national growth, and the other explaining regional differential growth. The sum of the two provides, by definition, total or absolute regional growth. This structure differs substantially from the existing econometric regional growth models, which in general move towards a direct interpretation of absolute regional growth either by replicating national macroeconomic models, or by constructing complex systems of equations for each region linking the region to both the national aggregate economy and to the other regional economies through input–output technical coefficients. The advantage of the MASST model’s structure is that a strong interconnection between regional and national growth is established: national macroeconomic trends and policies generate an effect on both national and regional growth, but at the same time regional structures and policies affect both regional and national performance in an interactive national– regional manner. This structure allows account to be taken of complex vertical feedbacks between the regional and national economy without imposing a complex system of interlinked equations.

STIMSON PAGINATION (M2469).indd 221

20/12/2010 15:12

222

Endogenous regional development

The most innovative aspect of MASST is that, thanks to its simulation algorithm, it can be considered a ‘generative’ model of regional growth in the sense defined by Richardson (1969), even if it also encompasses macroeconomic and institutional aspects, which are typically national and top-down. In MASST, regional growth has a role in determining national performance. The model thus supersedes the limiting and erroneous role given in general to the regional side of growth models: that of simply distributing national growth among regions in a typical top-down approach. The scenario-building methodology on which simulations lie is based on identification of the institutional, socio-demographic and economic driving forces of change, and their possible alternative trajectories, which give rise to different opportunities for growth and patterns of territorial distribution. Once the driving forces have been identified, they are translated into quantitative assumptions concerning the independent causal variables of the forecasting model. Consequently, growth rates of GDP and population as well as their levels for each year up to 2015 are simulated. The aim is not to achieve precise quantitative values of economic elements, nor, on the other hand, is it merely to provide a qualitative image of what the economic system will look like; the aim is to show the main trends and relative behavioural paths that will be at work under specific assumptions on how the main driving forces of change will evolve. Both the values assigned to the target variables and the regional values emerging from the final results indicate an order of magnitude and some relative behavioural classes (high-medium-low increase or decrease), rather than precise quantitative values. The results of the simulations presented here are based on a baseline scenario built on the assumption that the present trends affecting growth and the associated policies put in place will continue in the future. The Regional Impact of Macroeconomic Policies Macroeconomic policies are usually considered to be neutral in terms of inter-regional disparities. Yet, for inter-sectoral reasons, selective regional effects may be expected: capital-intensive vs. labour-intensive, exposed vs. non-exposed sectors to international trade, are differently hit by policies having effects on real interest rates or exchange rate movements. These differentiated effects have been demonstrated by means of targeted simulations with the MASST model.10 In fact, by focusing on the spatial effects of specific national policies and changes in macroeconomic contexts, it was possible to disentangle the effects of single policy trends and, in particular, of a revaluation of the EU currencies generated by monetary policy or macroeconomic conditions. In general terms, a revaluation would

STIMSON PAGINATION (M2469).indd 222

20/12/2010 15:12

Macroeconomic and territorial policies

223

negatively affect growth in all regions of Europe, but less intensively so in the east, where cost competitiveness is higher for historical reasons (Figure 11.3). The Impact of EU Policies: the Case of Structural Funds Figure 11.4 concerns the impact of structural funds on regional growth. The map shows the regional average growth rate between 2002 and 2015 if structural funds are set to 0. The impact is easy to envisage, for structural funds provide direct support to peripheral regions. The figure shows that the Objective 1 regions suffer the most from the lack of structural funds. All Objective 1 regions lose growing capacity: all Greece, most of Spain, Ireland, north of Great Britain, the north of Scandinavian countries, and the south of Italy retain a negative growth rate. Eastern countries are also penalized by the lack for structural funds that in the baseline simulation are also devoted to them. What is less immediate to envisage is that the lack of structural funds has an indirect negative impact on most European regions, since in the regional integration perspective like the one modelled by MASST, the negative growth rates of peripheral regions pervade the national economic systems (via regional spillovers and national growth rates); negative growth rates are registered in most central regions, like most of the French regions, German and Dutch regions, in the UK, in Scandinavian countries. The very few exceptions regard the regions with capital cities or with a dynamic city located within them; it is the case of London and some of its surrounding regions, of Paris, Helsinki, Milan, Rome, Bologna, Amsterdam and Rotterdam, just to quote the most relevant ones. Territorial Capital and Regional Growth Figure 11.5 contains the results of the contribution of territorial capital to regional growth. Territorial capital is presented here in four categories, in particular social overhead capital, receptivity,11 entrepreneurship, creativity – all elements on which appropriate and renewed territorial policies may impinge. The first important result is that each component produces a rather different picture. The map in Figure 11.5a shows the contribution of social overhead capital to regional growth. Peripheral areas in Northern Europe (the Scandinavian countries and Scotland) together with most areas in Eastern countries receive the highest contribution to regional growth from transport infrastructure, thanks to efforts to overcome remoteness. However, also, central regions in the Pentagon (London-Paris-Milan-Munich-Hamburg), some advanced regions in

STIMSON PAGINATION (M2469).indd 223

20/12/2010 15:12

224

Endogenous regional development

Politecnico di Milano - MASST model - December 06 Effects of a revaluation of the exchange rate –0.199 – –0.184 –0.184 – –0.174 –0.174 – –0.159 –0.159 – –0.134 –0.134 – –0.108 –0.108 – –0.086 –0.086 – –0.057 –0.057 – –0.015 EU mean = –0.133

Source: The authors.

Figure 11.3

Spatial effects of a revaluation of the exchange rates across EU regions

STIMSON PAGINATION (M2469).indd 224

20/12/2010 15:12

Macroeconomic and territorial policies

225

Politecnico di Milano - MASST model - June 2007 Difference in average annual % GDP growth rate 2002-2015 in Baseline Scenario if Structural Funds are set to 0 < –0.2418 –0.2418 – –0.1493 –0.1493 – –0.0905 –0.0905 – –0.0505 –0.0505 – –0.0318 –0.0318 – –0.0178 –0.0178 – 0 >0

Source: The authors.

Figure 11.4

The impact of structural funds on regional growth rates across EU regions

northern Italy, some agglomerated areas like Barcelona and Madrid in Spain, and Porto and Lisbon in Portugal perform relatively well thanks to the infrastructure increase, probably thanks to the efforts to overcome congestion.

STIMSON PAGINATION (M2469).indd 225

20/12/2010 15:12

226

Endogenous regional development Social overhead capital –0.523 – –0.334 –0.333 – –0.202 –0.201 – –0.148 –0.147 – –0.106 –0.105 – –0.066 –0.065 – –0.019 –0.018 – –0.031 –0.32 – –0.092

(a)

Source: The authors, derived from MASST model.

Figure 11.5

Contribution of territorial capital to regional growth across EU regions: (a) social overhead capital

A different picture is presented by the map in Figure 11.5b, where the role of receptivity in regional growth (i.e., that part of regional growth that is dependent on the performance of neighbouring regions, like a sort of growth spillover) is highlighted. The map shows that receptivity has a positive effect on growth in central, ‘Pentagon’ Europe, spreading around major capital and ‘mega’ regions (London, Paris, Milan, Munich,

STIMSON PAGINATION (M2469).indd 226

20/12/2010 15:12

Macroeconomic and territorial policies

227

Receptivity –0.418 – –0.127 –0.126 – 0.001 0.002 – 0.267 0.268 – 0.523 0.524 – 0.713 0.714 – 0.904 0.905 – 1.049 1.050 – 1.203

(b)

Figure 11.5

(continued) (b) receptivity

Brussels, etc.), but, interestingly enough, also scattered towards more peripheral territories. The map in Figure 11.5c is devoted to the contribution provided by entrepreneurship to regional growth. It is evidently shown that the role of entrepreneurship in eastern regions’ growth is rather limited, and it is instead more important in peripheral countries of the 15 old member states, like Italy, Spain and Greece. A relatively important role is also played by entrepreneurship in some regions of the Pentagon area. As expected, entrepreneurship plays a limited role in the dynamics of capital

STIMSON PAGINATION (M2469).indd 227

20/12/2010 15:12

228

Endogenous regional development

Entrepreneurship 0.140 – 0.244 0.245 – 0.330 0.331 – 0.429 0.430 – 0.535 0.536 – 0.676 0.677 – 0.866 0.867 – 1.157 1.158 – 2.097

(c)

Figure 11.5

(continued) (c) entrepreneurship

regions, the latter being probably more influenced by the presence of value-added functions, not represented by a variable of self-employment. Interestingly enough, the map in Figure 11.5d presents a rather unexpected result for what concerns the contribution of creativity to regional growth: a decisive contribution of creativity to regional growth is shown in eastern regions, while in the 15 old member states the highest values are registered in some regions of the Pentagon. If all effects are counted together, the outcome is the one presented in the map in Figure 11.5e. Territorial capital provides an important contribution to regional growth in most Pentagon regions, in some peripheral

STIMSON PAGINATION (M2469).indd 228

20/12/2010 15:12

Macroeconomic and territorial policies

229

Creativity –0.276 – –0.238 –0.237 – –0.196 –0.195 – 0.008 0.009 – 0.024 0.025 – 0.056 0.057 – 0.137 0.138 – 0.233 0.234 – 0.424

(d)

Figure 11.5

(continued) (d) creativity

areas, like Greece and part of Spain, and France. It is evidently lacking in eastern countries, in peripheral countries in the north (Scandinavian countries, UK and Italy). Interestingly enough, capital regions, like Madrid, Lisbon, Paris, Athens, London, Copenhagen, Oslo, Helsinki, and important agglomerated regions like the regions in Northern Italy, Barcelona, Côte d’Azur, receive a small contribution from territorial capital to regional growth. We are inclined to say that, as with all productive factors, territorial capital also shows decreasing marginal productivity.

STIMSON PAGINATION (M2469).indd 229

20/12/2010 15:12

230

Endogenous regional development

Total effect –0.261 – 0.128 0.129 – 0.413 0.414 – 0.655 0.656 – 0.869 0.870 – 1.044 1.045 – 1.236 1.237 – 1.482 1.483 – 2.514

(e)

Figure 11.5

(continued) (e) total effect

CONCLUSIONS The objective of this chapter was twofold: on the one hand to present the rationale for regional competitiveness policies, highlighting the theoretically based reasons for their existence; and on the other to present a historical development perspective of regional policy targets and styles. The main messages stemming from the analysis can be summarized as follows. The persistence of regional disparities justifies the relevance of

STIMSON PAGINATION (M2469).indd 230

20/12/2010 15:12

Macroeconomic and territorial policies

231

regional policies; regions compete on the basis of absolute advantages, and therefore need solid policy tools to enhance their competitiveness. The policies for regional competitiveness are both national and regional policies: the former are generally felt to be space-invariant policies; what we claim in this chapter is that, on the contrary, they have a different impact according to the structural characteristics of regions; and the latter require a new approach, widening their scope in the direction of soft and knowledge components of territorial capital. Based on simulation exercises through the MASST model, the chapter presents empirical evidence on the impact of: ● ● ●

macroeconomic policies, and in particular a devaluation policy; the impact of a European policy, and in particular the impact of structural funds; the role of territorial capital elements on regional growth patterns.

The results support and highlight important elements. Macroeconomic policies are not neutral in terms of inter-regional disparities; for intersectoral reasons, selective effects are visible. In fact, a real revaluation of the currency generates wider tensions in industries and regions characterized by a greater degree of openness to international trade. Furthermore, revaluation is likely to hit more, ceteris paribus, those regions specialized in labour-intensive industries, because it raises labour costs expressed in international currency (while the cost of capital will remain unchanged at the level determined internationally, especially if the revaluation is indicative of a strong and potentially fast-growing economy). Moreover structural funds turn out to provide direct support to peripheral regions, and to have an indirect positive impact on most European regions, due to the inter-regional integration that characterizes the economy. The simulation exercise clearly shows that in those regions where territorial capital assets play an important role on regional growth, the overall performance of the regions is higher. Moreover, it clearly demonstrates that territorial capital, as with all production factors, is subject to strong decreasing returns to scale: in fact, in those regions (agglomerated and mega-regions) in which the level of territorial capital is higher, its effects on regional growth are more contained. A last interesting aspect emerges: the different factors of territorial capital analysed (receptivity, social overhead capital, entrepreneurship and creativity) play a different role on local growth according to the settlement structure and relative location of regions. All this allows us to say that territorial capital appears as a new, fruitful concept on which modern territorial policies have to be built.

STIMSON PAGINATION (M2469).indd 231

20/12/2010 15:12

232

Endogenous regional development

NOTES 1.

2.

3. 4.

5.

6.

7.

8.

9.

The ‘classical’ equilibrating process relies on downward flexibility of prices and wages, which is hampered by the existence of national wage contracts in both private and public structures and by the homogeneity of import prices (remembering that regions are very open economies). The second, ‘modern’ process relies on the devaluation of the currency, and it is automatically excluded in an inter-regional context. The importance of this last reflection is magnified by the consideration that, with the creation of exchange rate agreements and large monetary unions, states will increasingly lose control over the external value of their currency, and will increasingly behave inside the unions like regions inside a nation. Krugman speaks of a general ‘competitiveness obsession’, mainly referring to the case of countries. From a modelling point of view, already 15 years ago it was shown that, if the linearity assumptions of the traditional neoclassical model were abandoned and some nonlinearities and increasing returns introduced, the same neoclassical model could well accommodate varied possible outcomes – both spatial diffusion and concentration (Miyao, 1987); and also ‘endogenous growth’ models, which include cumulative processes into a neoclassical production function (Romer, 1986), mainly end up with an inter-regional divergence process. The Italian experience of the first economic boom, 1957–64, totally concentrated in only three areas – Milan, Turin and Genoa, the ‘north-western triangle’, is telling in this respect: wage and cost inflation, massive migrations from the south, social and political tensions, deterioration of environmental quality due to excessive speed in economic transformation, generated the anticipated end of the expansion phase, an early vanishing of previous competitiveness and 15 years of slow growth in these areas up to the end of the following decade. History might repeat itself. In the simulation experiment of regional growth in the EU up to 2015, realized through the MASST model (see later in the chapter) this fact is evident, almost all countries showing a standard deviation in inter-regional growth rates lower than the international standard deviation of growth rates of EU-27 countries. This is particularly true for the 15 old member countries (with the exception of Ireland and the partial exception of Spain and Portugal, showing an internal standard deviation similar to the international one). On the other hand, all 12 new member countries show a higher variability of internal regional growth rates, bringing support to expectations à la Williamson about the increasing inter-regional disparities in the first phase of a development or integration process. This general trend is generated by two differentiated evolutions inside the two blocks of EU countries. In fact, in the 12 new member countries, inter-regional disparities will increase, due to a strongly widening intra-national dualism between core and periphery regions, insufficiently counterbalanced by slowly decreasing international disparities. On the other hand, a partly different trend is likely to characterize the 15 old member countries: here too intra-national disparities will increase, though at a much slower pace than in the 12 new member countries, while international disparities will also increase, for the first time, though only slightly. Some countries in fact, like Ireland, have already more than completed their catching-up process, and their persisting miracle will widen international disparities; other countries, relatively wealthy, like Holland and Italy, are losing momentum, while some others, less advanced, like Greece or partially Portugal, are not expected to show convincing performances (see Figure 11.1a and 11.1b). An operational methodology for Territorial Impact Assessment of EU policies, along the lines suggested by the Commission, enphasizing the three dimensions of ‘territorial cohesion’, namely territorial efficiency, quality and identity, was recently developed with the TEQUILA model – Territorial Efficiency, Quality, Identity Layered Assessment; see Camagni (2006). For a wide explanation of MASST, see Capello (2007); Capello et al. (2008).

STIMSON PAGINATION (M2469).indd 232

20/12/2010 15:12

Macroeconomic and territorial policies 10. 11.

233

These simulations were performed as part of the ESPON 3.4.2 Project exploring the ‘Territorial Impact of EU Economic Policies’. For the complete report, see the ESPON website, www.espon.eu; accessed 27 May 2010. Receptivity is here defined as that part of regional growth that is dependent on the performance of neighbouring regions, like a sort of growth spillover.

REFERENCES Andersson, A. (1985), ‘Creativity and Regional Development’, Papers of the Regional Science Association, 56(1), 5–20. Arkleton Institute, University of Aberdeen (2006), The Territorial Impact of CAP and Rural Development Policy, ESPON Project 2.1.3., available at: www.espon. eu accessed 27 May 2000, Luxembourg. Armstrong, H.W. (2002), ‘Regional Policy: Reconciling the Convergence and Evaluation Evidence’, in J.R. Cuadrado and M. Parellada (eds), European Union Regional Policy: Reconciling the Convergence and Evaluation Evidence’, Springer Verlag, Berlin, pp. 231–72. Armstrong, H.W. and Taylor, J. (1993), Regional Economics and Policy, Harvester Wheatsheaf London. Bagnasco, A. (1977), Le Tre Italie, Il Mulino, Bologna. Barro, R.J. and Sala-i-Martin, X. (1991), ‘Convergence Across States and Regions’, Brookings Papers in Economic Activity, 22(1), 107–82. Barro, R.J. and Sala-i-Martin, X. (1992), ‘Convergence’, Journal of Political Economy, 100(2), 223–51. Becattini, G. (1990), ‘The Marshallian Industrial District as a Socio-economic Notion’, in F. Pyke, G. Becattini and W. Sengenberger (eds), Industrial Districts and Inter-firm Cooperation in Italy, ILO, Geneva. Borts, G.H. and Stein, J.L. (1964), Economic Growth in a Free Market, Columbia University Press, New York. Brusco, S. (1982), ‘The Emilian Model: Productive Decentralisation and Social Integration’, The Cambridge Journal of Economics, 6(2), 167–84. Buchanan, J.M. and Tullock, G. (eds) (1982), Towards a Theory of the Rent Seeking Society, A&M Press, Austin. Camagni, R. (1991), ‘Technological Change, Uncertainty and Innovation Networks: Towards a Dynamic Theory of Economic Space’, in R. Camagni (1991), Innovation Networks: Spatial Perspectives, Belhaven-Pinter, London. Camagni, R. (1992), ‘Development Scenarios and Policy Guidelines for the Lagging Regions in the 1990’s’, Regional Studies, 26(4), 361–74. Camagni, R. (1995), ‘The Concept of Innovative Milieu and its Relevance for Public Policies in European Lagging Regions’, Papers in Regional Science, 74(4), 317–40. Camagni, R. (1999), ‘The City as a Milieu: Applying GREMI’s Approach to Urban Evolution’, Revue d’Economie Régionale et Urbaine, 3, 591–606. Camagni, R. (2001), ‘Policies for Spatial Development’, in OECD (2001), pp. 147–69. Camagni, R. (2002), ‘On the Concept of Territorial Competitiveness: Sound or Misleading?’, Urban Studies, 39(13), 2395–412. Camagni, R. (2006), ‘Territorial Impact Assessment – TIA: A Methodological

STIMSON PAGINATION (M2469).indd 233

20/12/2010 15:12

234

Endogenous regional development

Proposal’, in Scienze Regionali – Italian Journal of Regional Science, 5(2), 135–46. Camagni, R. (2008), ‘Territorial Capital and Regional Development’, in R. Capello and P. Nijkamp (eds), Handbook of Regional Dynamics and Growth: Advances in Regional Economics, Edward Elgar, Cheltenhan, UK and Northampton, MA, USA, pp. 118–32. Camagni, R. and Capello, R. (2000), ‘The Role of Inter-SME Networking and Links in Innovative High-tech Milieux’, in D. Keeble and F. Wilkinson (eds), High-technology Clusters, Networking and Collective Learning in Europe, Ashgate, Avebury, pp. 118–55. Camagni, R. and Capello, R. (forthcoming), ‘Territorial Capital, Entrepreneurship and Regional Development in the EU, Regional Studies. Camagni, R. and Cappellin, R. (1985), Sectoral Productivity and Regional Policy, CEC, Brussels. Camagni, R. and Gibelli, M.C. (1996), ‘Cities in Europe: Globalisation, Sustainability and Cohesion’, in Presidenza del Consiglio dei Ministri – Dipartimento Politiche Comunitarie, European Spatial Planning, Rome, Poligrafico dello Stato. Camagni, R. and Maillat, D. (eds) (2006), Milieux Innovateurs: Théorie et Politiques, Economica, Paris. Capello R. (1999a), ‘A Measurement of Collective Learning Effects in Italian High-tech Milieux’, Revue d’Economie Régionale et Urbaine, 3, 449–68. Capello, R. (1999b), ‘Spatial Transfer of Knowledge in High-technology Milieux: Learning vs. Collective Learning Processes’, Regional Studies, 33(4), 353–65. Capello, R. (2001), ‘Urban Innovation and Collective Learning: Theory and Evidence from Five Metropolitan Cities in Europe’, in M.M. Fischer and Froehlich J. (eds), Knowledge, Complexity and Innovation Systems, Springer, Berlin, Heidelberg, New York, pp. 181–208. Capello, R. (2007), ‘A Forecasting Territorial Model of Regional Growth: The MASST Model’, Annals of Regional Science, 41(4), 753–87. Capello, R., Camagni, R., Chizzolini, B. and Fratesi, U. (2008), Modelling Regional Scenarios for the Enlarged Europe: European Competitiveness and Global Strategies, Springer Verlag, Berlin. Castells, M. (1985), High Technology, Space and Society, Sage, Beverly Hills. CEC – Commission of the European Communities (2005a), Territorial State and Perspectives of the European Union, Scoping Document and Summary of Political Messages, Brussels, May. CEC – Commission of the European Communities (2005b), Cohesion Policy in Support of Growth and Jobs: Community Strategic Guidelines, 2007–2013 (Communication from the Commission), COM(2005). Cooke, P. and Morgan, K. (1998), The Associational Economy. Firms, Regions and Innovation, Oxford University Press, Oxford. Dixon, R.J. and Thirwall, A.P. (1975), ‘A Model of Regional Growth Rate Differences on Kaldorian Lines’, Oxford Economic Papers, 27(2), 201–14. Dosi, E. (1982), ‘Technological Paradigms and Technological Trajectories: A Suggested Interpretation of the Determinants and Directions of Technical Change’, Research Policy, 11(3), 147–62. Ewers, H.J. and Allesch, J. (eds) (1990), Innovation and Regional Development, De Gruyter, Berlin. Ewers, H. and Wettman, R. (1980), ‘Innovation Oriented Regional Policy’, Regional Studies, 14, 161–79.

STIMSON PAGINATION (M2469).indd 234

20/12/2010 15:12

Macroeconomic and territorial policies

235

Foray, D. (2000), L’Economie de la Connaissance, La Découverte, Paris. Gilly, J.P. and Torre, A. (eds) (2000), Dynamiques de Proximité, L’Harmattan, Paris. GREMI (1995), ‘Cohesion and the Development Challenge Facing the Lagging Regions’, Regional Development Studies, No. 24, directed by R. Camagni and M. Quevit, European Commission, Bruxelles. Hirschman, A.O. (1957), ‘Investment Policies and “Dualism” in Underdeveloped Countries’, The American Economic Review, 47(5), 550–70. Kaldor, N. (1970), ‘The Case for Regional Policies’, Scottish Journal of Political Economy, 17(3), 337–48. Keeble, D. and Wilkinson, F. (1999), ‘Collective Learning and Knowledge Development in the Evolution of Regional Clusters of High-technology SMEs in Europe’, Regional Studies, 33(4), 295–304. Keeble, D. and Wilkinson, F. (eds) (2000), High-technology Clusters, Networking and Collective Learning in Europe, Ashgate, Aldershot. Knack, S. and Keefer, P. (1997), ‘Does Social Capital have an Economic Payoff? A Cross-country Investigation’, Quarterly Journal of Economics, 112(4), 1251–88. Knight, R. (1984), ‘The Advanced Industrial Metropolis: A New Type of World City’, paper presented at the seminar The Future of the Metropolis, Berlin, October. Krugman, P. (1991), ‘Increasing Returns and Economic Geography’, Journal of Political Economy, 99(3), 483–99. Krugman, P. (1996), ‘Making Sense of the Competitiveness Debate’, Oxford Review of Economic Policy, 12(3), 17–25. Krugman, P. (1998), Pop Internationalism, MIT Press, Cambridge, MA. Krugman P. and Venables (1996), ‘Integration, Specialisation and Adjustment’, European Economic Review, 40, 857–80. Lundvall, B.A. and Johnson, B. (1994), ‘The Learning Economy’, Journal of Industrial Studies, 1(2), 23–42. Ministers for Spatial Planning (2007), Leipzig Charter on Sustainable European Cities, Informal Meeting of Ministers, Leipzig, May. Miyao, T. (1987), ‘Dynamic Urban Models’, in E. Mills (ed.), Urban Economics; Handbook of Regional and Urban Economics, North-Holland, Amsterdam, vol. 2, pp. 877–925. Myrdal, G. (1957), Economic Theory and Underdeveloped Regions, Gerald Duckworth & Co., London. Nelson, R. and Winter, S. (1982), An Evolutionary Theory of Economic Changes, Harvard University Press, Cambridge, MA. Nijkamp, P. (ed.) (1986), Technological Change, Employment and Spatial Dynamics, Springer Verlag, Berlin. OECD (2001), Territorial Outlook, Paris. Pennington, M. (1999), ‘Free Market Environmentalism and the Limits of Land Use Planning’, Journal of Environmental Policy & Planning, 1(1), 43–59. Parr, J. (1999), ‘Growth Pole Strategies in Regional Economic Planning: a Retrospective View. Part 2. Implementation and Outcome’, Urban Studies, 36(8), 1247–68. Richardson, H.W. (1969), Regional Economics, World University, Redwood Press Limited, Trowbridge, Wiltshire. Romer, P. (1986), ‘Increasing Returns and Long-run Growth’, Journal of Political Economy, 94(5), 1002–37.

STIMSON PAGINATION (M2469).indd 235

20/12/2010 15:12

236

Endogenous regional development

Rosenstein-Rodan, P.N. (1943), ‘Problems of Industrialisation of Eastern and South-Eastern Europe’, The Economic Journal, 53(210/211), 202–11. Simon, H. (1972), ‘From Sustantive to Procedural Rationality’, in C.B. McGuire and R. Radner (eds), Decision and Organization, North Holland, Amsterdam, pp. 116–76. Stohr, W. (ed.) (1990), Global Challenge and Local Responses, United Nations University Press/Mansell, London. Storper, M. (1995), ‘The Resurgence of Regional Economies Ten Years Later: The Region of Untraded Interdependencies’, European Urban and Regional Studies, 2, 191–221. Tatsuno, S. (1986),The Technopolis Strategy, Prentice Hall, New York. Williamson, J.G. (1965), ‘Regional Inequality and the Process of National Development: A Description of the Patterns’, Economic Development and Cultural Change, 13(4), 3–45. Williamson, O. (2002), ‘The Lens of Contract: Private Ordering’, American Economic Review, Papers and Proceedings, 92(2), 438–53.

STIMSON PAGINATION (M2469).indd 236

20/12/2010 15:12

12.

Endogenous employment growth and decline in Australian capital city statistical divisions* Alistair Robson

INTRODUCTION The Australian economy has experienced significant changes in its industrial structure in recent decades (Productivity Commission, 1998, p. xiv). Broadly, there has been a large shift in employment from the manufacturing industry to the services industries. Such change is similar in many other OECD countries. Research on regional structural economic change is copious, and the Productivity Commission report of 1998 provides a good overview – albeit 12 years old – on the topic. There are, however, considerable regional differences in the nature and magnitude of changes in employment structures across industry sectors. In this chapter the reasons for these changes are considered, with the focus being explicitly on the nation’s state capital city metropolitan areas with populations exceeding 1 million (hereafter referred to as the five-capital cities). Those five-capital cities – in descending order of population size – are: Sydney (the capital city of the state of New South Wales); Melbourne (Victoria); Brisbane (Queensland); Perth (Western Australia); and Adelaide (South Australia). The chapter begins with an overview of the economic structure of the state capital cities. That is followed by a shift-share analysis of change in employment in industry sectors over the decade 1996 to 2006 to explain the divergent employment growth between the cities. The emphasis is on analysing the regional shift component as a surrogate measure of endogenous performance of the industry sectors within each of the capital cities. The chapter concludes with a summary of the differences between the cities in their endogenous employment performance for the decade 1996 to 2006.

237

STIMSON PAGINATION (M2469).indd 237

20/12/2010 15:12

238

Endogenous regional development

OVERVIEW OF THE ECONOMIC STRUCTURE OF THE CAPITAL CITIES Capital City Dominance In Australia the capital city regions1 are dominant in terms of population, employment and production. They are effectively primate cities in their respective states according to the Jefferson (1939) criteria. The latest available data from the 2006 Census of Population and Housing shows these regions accounted for 63 per cent of the nation’s population and 65 per cent of total employment.2 However, historically the capital city regions have not always been so dominant. For example, the capital city regions accounted for only 36 per cent of Australia’s population in 1901 (Australian Bureau of Statistics, 1997, p. 18). But since that time they have reached their current level of dominance (see Figure 12.1). The dominant pattern of the five-capital cities is evident in all states, but it is less so Queensland, which is a comparatively special case where the regions to the south, west and north3 of Brisbane have strong economic linkages to each other that is not as evident in other states’ capital city regions (such as in Sydney with Newcastle and Wollongong). Referred to as the South East Queensland region, it accounts for about two-thirds of both the population and the employment in the state, a level broadly on a par with the other capital city regions. The capital cities have a high employment to working age population. And they also tend to have relatively higher income per capita compared with other non-metropolitan regions. However, in an analysis of regional economic change in Australia over the period 1991 to 1996, Harding et al. (2000) found a divergence in economic performance – as measured by incomes – between the capital cities. Changing National Shares Ideally, production data for the five-capital cities would be used in an analysis of their economic performance. However, this data does not exist for the capital city regions. But state production data does exist and it provides a context for capital city employment changes. Over the period 1996 to 2006, falls in the share of national real Gross Domestic Product were recorded in New South Wales (down 1.4 percentage points to 34 per cent), South Australia (down 0.5 of a percentage point to 7 per cent), Victoria (down 0.3 of a percentage point to 24 per cent) and Tasmania (down 0.2 of a percentage point to 2 per cent). Increases in the share were recorded in Queensland (up 1.9 percentage points to 18

STIMSON PAGINATION (M2469).indd 238

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

239

2.0

Change in share (p.p:s)

1.5 1.0 0.5 0.0

1

2

3

4

6 5

10 11 7

8

9

19 20 12 13 14 15 16 17 18

–0.5 –1.0 –1 –2.0

Industry

1 Agriculture/forestry & fishing 2 Mining 3 Manufacturing 4 Electricity/gas/water & waste services 5 Construction 6 Wholesale trade 7 Retail trade 8 Accommodation & food services 9 Transport/postal & warehousing 10 Information media & telecommunications

11 Financial & insurance services 12 Rental/hiring & real estate services 13 Professional/scientific & technical services 14 Administrative & support services 15 Public adminstration & safety 16 Education & training 17 Healthcare & social assistance 18 Arts & recreation services 19 Other services 20 Inadequately described/Not stated

Source: Compiled by the author using data available from the Australian Bureau of Statistics (2007).

Figure 12.1

Percentage point change in capital Cities’ share of national employment by industry, 1996–2006

per cent) and Western Australia (up 0.6 of a percentage point to 12 per cent). The shifts in production over this period are largely replicated in the shifts in employment. The states losing national share of employment were New South Wales (1.6 percentage points), South Australia and Tasmania (both 0.2 of a percentage point). The states recording an increase in their share of national employment were Queensland (1.6 percentage points), Western Australia (0.3 of a percentage point) and Victoria (0.2 of a percentage point). Capital City Shares of State Population and Employment Over the decade from 1996 to 2006 there was little change in the relative dominance of the capital city regions compared with their states as a

STIMSON PAGINATION (M2469).indd 239

20/12/2010 15:12

240

Table 12.1

Endogenous regional development

Capital city shares of state employment and population, 2006 Employment

Working Age Population (15–64)

Employment to Working Age Population (15–64) Ratio

Level in Change Level in Change Level in Change 2006 1996–2006 2006 1996–2006 2006 1996–2006 (% of (pps of (% of (pps of (% of (pps of state) state) state) state) state) state) Sydney Melbourne Brisbane Adelaide Perth Five-capital city weighted average

65 74 47 74 74 65

−0.2 0.3 0.4 — 1.4 —

65 74 45 74 73 65

1.2 1.2 0.6 −0.1 0.6 0.6

101 100 103 100 101 101

−2.2 −1.2 −0.4 0.1 1.1 −1.0

Source: Compiled by the author using data available from the Australian Bureau of Statistics (2007).

whole in terms of population and employment (see Table 12.1). In 2006, the five-capital cities’ aggregate share of state employment was 65 per cent (unchanged from 1996), the same as their share of state population (up 0.6 of a percentage point from 1996). Overall, the ratio of employment to the working age population (15–64) was 1.0 per cent higher in the five-capital cities than their states, a fall of 1.0 percentage points from 1996, hence the patterns detected in the state data can be inferred as a good guide to the performance of the capital city regions. Indeed, over this time no capital city, or rest of state area, recorded a fall in either population or working age population (the age groups 15 to 64 years). Very prominent was the relatively large increase of 30 per cent in employment and 20 per cent in the working age population in Queensland over the decade to 2006, which was considerably above the national figures of a 19 per cent and 13 per cent increase respectively. The stronger growth in Queensland’s employment compared with working age population was reflected in the 6.9 per cent rise in its employment to working age population ratio. That compared with 5.3 per cent for Australia as a whole and 4.2 per cent for the five-capital cities aggregate average. That stronger growth in the employment to working age population ratio in Queensland occurred even though it was equal to the national average in 1996.

STIMSON PAGINATION (M2469).indd 240

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

241

Divergent Rates of Growth and Change in the Capital Cities One explanation for the divergent growth rates of employment is the different industrial structures of the capital city economies. Some of the five-capital cities may have relatively more employment in industries that are growing. Comparatively, other capital cities may have relatively more employment in industries that are declining. As shown in Table 12.2, the industry sector with the most variation in employment share in 1996 was the manufacturing sector. The lowest share of employment in the manufacturing industry was in Perth (10 per cent of total employment), followed by Brisbane (11 per cent of total employment). Comparatively, Melbourne had the highest proportion of employment in the manufacturing industry (17 per cent), followed by Adelaide (15 per cent) and Sydney (12 per cent). Employment in the manufacturing industry in the rest of Australia represented 10 per cent of total employment. This represents one of the major structural differences between capital city regions and the rest of Australia. By 2006, the manufacturing industry sector recorded some relatively large falls in share of employment (see Table 12.3). In Melbourne, its share fell by 3.7 percentage points over the decade to 2006 – the largest percentage share fall of any industry by capital city region. The share of employment in the manufacturing industries sector also fell in Sydney (by 2.2 percentage points), in Adelaide (by 2.1 percentage points), Brisbane (by 0.5 percentage points) and Perth (by 0.4 percentage points). Over this same period, the share of employment in the five-capital cities increased the most in the construction industry (up 1.4 percentage points). The largest share increases were in Perth (up 2.0 percentage points) and Adelaide (up 1.9 percentage points). Nationally, overall the most prominent changes in the share of national employment over the decade 1996 to 2006 were in the manufacturing industries (down 1.6 percentage points), and in both the agriculture, forestry and fishing industries and the wholesale trade industries (down 1.2 percentage points). Comparatively, there were increases in the share of national employment over the same period in the construction industry (up 1.6  percentage points), the retail trade industry (up 1.2 percentage points), and the health care and social assistance industries (up 1.1 percentage points).

STIMSON PAGINATION (M2469).indd 241

20/12/2010 15:12

242

STIMSON PAGINATION (M2469).indd 242

20/12/2010 15:12

1 0 17 1 6 6 10 5 5 3 5 1 8 3 5 7 9

6 7 10 6 6 4 6 2 8 3 5 7 9

Melbourne (%)

1 0 12 1

Sydney (%)

3 7 8 10

7

4 2

6 6 10 6 5 2

1 0 11 1

Brisbane (%)

Capital city industry structure share of employment, 1996

Agriculture\ forestry & fishing Mining Manufacturing Electricity\ gas\ water & waste services Construction Wholesale trade Retail trade Accommodation & food services Transport\ postal & warehousing Information media & telecommunications Financial & insurance services Rental\ hiring & real estate services Professional\ scientific & technical services Administrative & support services Public administration & safety Education & training Health care & social assistance

Table 12.2

3 6 8 12

6

4 2

5 6 10 6 4 2

1 0 15 1

Adelaide (%)

3 6 8 10

7

4 2

7 6 11 6 4 2

1 2 10 1

Perth (%)

3 5 7 9

7

5 2

6 6 10 6 5 3

1 0 13 1

2 7 8 9

4

2 1

6 4 10 7 4 2

10 2 10 1

Five-capital Rest of cities Australia (%) (%)

243

STIMSON PAGINATION (M2469).indd 243

20/12/2010 15:12

Source:

Note:

100

1 4 3 100

2 5 3 100

1 5 3 100

2 5 4

Compiled by the author using data available from the Australian Bureau of Statistics (2007).

Figures rounded to nearest whole percentage point.

Arts & recreation services Other services Inadequately described/Not stated Total employment (%) 100

2 5 4

100

2 5 3

100

1 4 3

244

STIMSON PAGINATION (M2469).indd 244

20/12/2010 15:12

−0.3 −0.1 −3.7 0.1 1.6 −0.7 1.3 0.4 0.1 −0.6 0.0 0.2 0.6 0.6 0.2 0.6 1.1

0.9 −1.1 0.9 0.2 −0.2 −0.5 0.4 0.2 0.7 0.1 0.3 0.6 1.0

Melbourne (pps)

−0.2 0.0 −2.2 0.0

Sydney (pps)

0.4 0.3 0.2 1.0

0.1

−0.2 0.2

1.5 −1.7 0.9 −0.1 0.1 −0.8

−0.4 0.1 −0.5 0.3

Brisbane (pps)

0.4 1.2 0.0 1.2

0.4

0.0 0.0

1.9 −1.6 1.6 0.0 0.0 −0.5

−0.1 0.2 −2.1 0.2

Adelaide (pps)

0.1 1.0 0.1 1.0

0.5

−0.5 0.0

2.0 −1.6 0.8 −0.2 −0.2 −0.7

−0.5 0.9 −0.4 0.1

Perth (pps)

0.3 0.5 0.4 1.0

0.5

0.0 0.1

1.4 −1.2 1.1 0.1 0.0 −0.6

−0.3 0.1 −2.2 0.1

0.5 0.8 0.2 1.3

0.4

−0.2 0.1

1.9 −1.2 1.3 0.0 0.0 −0.5

−2.6 −0.1 −0.7 0.0

Five-capital Rest of cities (pps) Australia (pps)

Capital city change in industry structure share of employment over the decade 1996 to 2006

Agriculture\ forestry & fishing Mining Manufacturing Electricity\ gas\ water & waste services Construction Wholesale trade Retail trade Accommodation & food services Transport\ postal & warehousing Information media & telecommunications Financial & insurance services Rental\ hiring & real estate services Professional\ scientific & technical services Administrative & support services Public administration & safety Education & training Health care & social assistance

Table 12.3

245

STIMSON PAGINATION (M2469).indd 245

20/12/2010 15:12

Source:

Note:

0.0

0.0 −0.7 −0.3 0.0

0.1 −0.9 −0.5 0.0

−0.1 −0.8 −0.5 0.0

−0.2 −0.9 −1.6

Compiled by the author using data available from the Australian Bureau of Statistics (2007).

Figures rounded to nearest whole percentage point (0 refers to less than 0.5).

Arts & recreation services Other services Inadequately described/Not stated Total employment 0.0

−0.2 −0.9 −1.2

0.0

0.0 −0.8 −0.6

0.0

0.0 −0.4 −0.9

246

Endogenous regional development

A SHIFT-SHARE ANALYSIS OF EMPLOYMENT CHANGE OVER THE DECADE 1996 TO 2006 As we will see later in the chapter, the changes in industry employment structure in the five-capital cities discussed above explain little of the actual employment growth performance of the capital city regions. One method that allows us to explain the employment growth performance of these regions is shift-share analysis. This technique allows for a decomposition of change in employment that is attributable to: ● ● ●

the share of national growth (national shift effect, NS); the mix of industries (industry mix effect, IM); the shift change in activities toward (or away from) a capital city region (differential or regional shift effect, RS).

The regional shift effect provides us with the net impact of factors affecting employment from in the region; or alternatively the endogenous factors. Hence the technique can provide the new insight that is central to the concern of this chapter. These factors can be useful to planners and researchers as they can be influenced more so than factors from outside the region (national shift effect and industry mix effect that are exogenous to the region). Background and Method Shift-share analysis has a long history of use in regional economic analysis. It was developed to project future employment in a particular region and is an off-shoot of the Constant Share Approach (for a discussion see Wadley and Smith, 2003). Like most methods there has been much debate on the efficacy and accuracy of the approach. There is also considerable debate about how much of the regional shift effect, which is a residual, truly reflects endogenous factors. There have been numerous refinements to the shift-share technique, particularly a version by Stilwell (1969), which provided a traditional shift-share technique with a working tableau and relevant equations (see Wadley and Smith, 2003, p. 1). In response to this, Chalmers (1971), Edwards et al. (1978), Markusen et al. (1991) and Noponen et al. (1997) have provided refinements to his method. While Wadley and Smith (2003, p. 1) argue that the method still has a way to go, the benefits of the method are its simplicity and low level of required data. Nonetheless, shift-share analysis is a powerful technique and is widely used in regional economics. Recent uses in Australia include a study by Mitchell (2008) who used

STIMSON PAGINATION (M2469).indd 246

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

247

labour force survey employment data to determine regional patterns of employment change adjusting for part-time and full-time status. Labour force survey data is not used in the research reported in this chapter as data from the Census of Population and Housing is considered by the author to be more robust and reliable (with the trade-off being fewer observations). The Australian Bureau of Statistics makes very clear that while it has attempted to improve the quality of employment data at geographic levels below the state and territory level, care should be taken in the interpretation of regional estimates due to relatively high standard errors (Australian Bureau of Statistics, 2009). Nonetheless, broad findings from that study are similar to those shown here. In the shift-share analysis in this chapter, the national shift, industry mix and regional shift effects are calculated by using a modified shiftshare technique developed by Haynes and Dinc (1997). There are more advanced methods – such as using efficiency measures using production data – but the data needed to use these methods are unavailable at the capital city level. In the Haynes and Dinc methodology the total change in employment over a time period is referred to as the ‘total shift’ (TS). As with all shift-share techniques, this methodology decomposes the total shift in employment into its exogenous and endogenous shift effects. The exogenous shift effects in this methodology are those effects attributable to both the increase in national employment (known as the ‘national shift’ effect, NS) and the impact of growth in certain industries nationally (known as the ‘industry mix’ effect IM). Comparatively, the endogenous shift effect (known as ‘regional’ or ‘differential’ shift effect, RS) is due to the impact of factors within the local area. This can be shown mathematically as: TS = NS + IM + RS The NS effect is the change in employment in a certain industry that would have taken place in employment had it grown at the same rate of that for all industries nationally. It is calculated by multiplying the level of employment in the local area by the growth rate of employment nationally: thus: NS = Ei × gn where: NS = the national shift effect; Ei = the employment in industry i at the initial year; gn = the growth rate of employment nationally.

STIMSON PAGINATION (M2469).indd 247

20/12/2010 15:12

248

Endogenous regional development

The IM effect is the change in employment in a certain industry that would have take place if it had grown at the same rate as that industry nationally: thus: IM = Ei × gi, n where: IM = the industry mix effect; Ei = the employment in industry i at the initial year; gi,n = the growth in employment in industry i nationally. The RS effect (also known as the differential shift effect) is the residual between the total shift and the combination of the national shift effect and the industry mix effect. It can be calculated as either the difference between those three, or more properly as: RS = Ei × (gi, r – gi, n) where: RS = the regional shift effect; Ei = the employment in industry i at the initial year; gi, r = the regional growth rate of employment in that industry; gi, n = the national growth rate of employment in that industry. The RS effect can be derived in aggregate for a region, or its individual industry components. The RS effects for each industry in a region are different from the IM effects. Attempts to separate exogenous and endogenous shift effects on employment for Australian regions have been few (Mitchell, 2008, p. 3). Recent work, apart from that by Mitchell, has been performed by Stimson et al. (2005, 2009). The technique used here is the same as those publications, albeit looking at capital city regions rather than non-metropolitan regions. The analysis reported here uses data based on the place of enumeration count methodology. This was because the Australian Bureau of Statistics releases time series data on CD, only using this count method for all Statistical Divisions.4 For consistency purposes this is the count method used throughout this chapter. The implications of using this count method is that employment data refer to people who are employed based on where they were counted. As such, a higher level of employment does not necessarily mean an area has more jobs in that area. This is not an

STIMSON PAGINATION (M2469).indd 248

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

249

issue for large Statistical Divisions, such as the capital city metropolitan areas, as they are largely self-contained. All data is sourced from the 2006 Census of Population and Housing Time Series Profile (Australian Bureau of Statistics, 2007). There are particular issues with this data set. These include the facts that over time, geographies and industry classifications change. The Australian Bureau of Statistics adjusts (called concordance) for these changes by particular methods. Another issue is that people filling out the industry of employment on the census form may not accurately identify the industry of their employer. An example the Australian Bureau of Statistics (2006, p. 198) provides is a person who works for a coal mining company as a driver of the company’s coal trucks. The individual’s occupation is truck driver. However, the industry of the individual’s employer is coal mining and not transport. To provide some measure of standardization, the regional shift effect throughout this chapter is compared with the change in working age population. There are issues with this measure, particularly where the level of employment self-containment varies between capital cities. Nonetheless, standardizing the regional shift effect provides a relatively good comparative tool. This is the first time such a standardization method has been applied to the results of the Haynes and Dinc method to the knowledge of the author. Results Using the shift-share method discussed above, the NS, IM and RS effects were obtained for the five-capital city regions for the period 1996 to 2006. The results are given in Table 12.4. Two distinct patterns are evident: In aggregate the five-capital cities recorded negative RS effects in employment of −15 668 or −1 job per 100 change in working age population. But most of this was attributable to Sydney, which recorded a negative RS effect of −113 034 (or −37 jobs per 100 increase in working age population). Adelaide recorded a much smaller negative RS effect of −17 193 jobs (or −37 per 100 increase in working age population). In contrast, positive RS effects were recorded in Brisbane (+66 832 or 29 jobs per 100 increase in working age population), Melbourne (+22 948 or +7 jobs per 100 increase in working age population) and Perth (+24 714 or +17 jobs per 100 increase in working age population). When the decade 1996 to 2006 is segmented into the two inter-census periods 1996–2001 and 2001–06 we get the results shown in Table 12.5. The following conclusions may be drawn: ●

Most of Sydney’s negative RS effects over the decade 1996 to 2006 occurred in the five-year period 2001 to 2006. There was a noticeable

STIMSON PAGINATION (M2469).indd 249

20/12/2010 15:12

250

STIMSON PAGINATION (M2469).indd 250

20/12/2010 15:12

72 87 89 146 96 94 87 107

Per 100 increase in working age population (no.) 322 080 267 520 126 212 83 925 106 380 1 467 941 906 064 561 798

106 82 56 180 73 94 86 109

10 126 −4 412 9 366 1 638 8,598 — 25 323 −25 323

3 −1 1 4 6 — 2 −5

−113 034 22 948 66 832 −17 193 24 714 — −15 668 15 668

−37 7 29 −37 17 — −1 3

Per 100 increase in working age population (no.)

(No.)

Per 100 increase in working age population (no.)

(No.)

(No.)

Per 100 increase in working age population (no.)

Regional Shift Effect

Industry Mix Effect

National Shift Effect

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

219 197 286 066 202 382 68 374 139 700 1 467 862 915 719 552 143

(No.)

Total Shift

Shift-share results for capital city Statistical Divisions, 1996–2006

Sydney Melbourne Brisbane Adelaide Perth Australia Five-capital cities Rest of Australia

Table 12.4

251

STIMSON PAGINATION (M2469).indd 251

20/12/2010 15:12

Data is not additive.

−37 7 29 −37 17 — −1 3

Per 100 increase in working age population (no.) −13 336 27 475 14 133 −8 604 3 665 — 23 333 −23 333

(No.)

−7 17 14 −39 5 — 4 −10

Per 100 increase in working age population (no.)

1996–2001

−98 021 −8 532 51 436 −7 945 19 419 — −43 643 43 643

(No.)

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

Note:

−113 034 22 948 66 832 −17 193 24 714 — −15 668 +15 668

(No.)

1996–2006

Capital city regional shift effects: 1996–2001, 2001–06 and 1996–2006

Sydney Melbourne Brisbane Adelaide Perth Australia Five-capital cities Rest of Australia

Table 12.5

−85 −5 40 −32 26 — −9 16

Per 100 increase in working age population (no.)

2001–06

252

●

●

●

Endogenous regional development

acceleration in the negative RS effect in Sydney from −7 jobs per 100 increase in working age population to −85 jobs. Melbourne recorded a positive RS effect in the 1996–2001 period of +17 jobs per 100 increase in working age population, which was followed by a negative RS effect of −5 jobs per 100 increase in working age population. Adelaide’s RS effect over these two periods was negative, but marginally less negative in the later period at −32 jobs per 100 increase in working age population compared with −39 jobs per 100 increase in population in the period 1996–2001. Both Brisbane and Perth recorded accelerating positive RS effects over the two periods (Brisbane with +14 jobs per 100 increase in working age population in the 1996–2001 period to +40 jobs in the 2001–06 period; and Perth with +5 jobs per 100 increase in working age population in the 1996–2001 period to +26 jobs in the 2001–06 period).

INDUSTRIAL EXPLANATIONS FOR REGIONAL SHIFT EFFECTS IN CAPITAL CITIES While in aggregate there have been some significant movements in RS effects for Australia’s five-capital cities, by looking at the RS effects by industry it is possible to see which particular industries have experienced strong endogenous employment growth. This is, of course, different from the IM effect described previously. The industry sectors mainly affecting aggregate RS effects – as opposed to the IM effect described earlier – were: manufacturing (particularly for Sydney and Melbourne) and construction (particularly for Sydney).The large negative RS effect of manufacturing employment in Sydney and Melbourne, and to a lesser degree Adelaide, was countered by positive RS effects in the rest of Australia, Brisbane and Perth. Proportionate to their 1996 employment levels, the mining industry recorded large RS effects in the five-capital cities in aggregate. But Sydney and Melbourne recorded strong negative proportionate RS effects, while Perth, Adelaide and Brisbane recorded strong positive proportionate RS effects. Other proportionately large RS effects were recorded in the electricity gas and water supply industries for Brisbane (positive), the construction industry for Sydney (negative) and the arts and recreation services industries for Adelaide (negative). Overall, most of the negative RS effects for all the five-capital cities in aggregate over the period 1996 to 2006 were concentrated in these industry sectors:

STIMSON PAGINATION (M2469).indd 252

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities ● ● ●

253

the manufacturing industries (−20 838 jobs or more than the total negative RS effect); the construction industry (−7468 jobs or 48 per cent of the total negative RS effect); the administrative and support services industries (−6078 jobs or 39 per cent of total negative RS effects jobs).

The highest positive RS effects were recorded in: ● ● ●

the wholesale trade industry (+8629 jobs); the mining industry (+6372 jobs); the financial and insurance services industries (+6080 jobs).

Proportionately, compared with 1996 employment levels over this period, the following patterns are evident: (1) The highest positive RS effect was in the mining industry at 32 per cent, followed by the electricity, gas and water supply industries at 6 per cent, 3 per cent for wholesale trade industry and for information, media and telecommunications industries, and 2 per cent for education and training. (2) The highest negative proportionate RS effects over this period were recorded in the agriculture, forestry and fishing industry at −8 per cent, followed by the administrative and support services industries at −4 per cent, and then at −3 per cent manufacturing, construction and other services industry sectors. The focus now turns to an examination of some individual industries that have had relatively large contributions to RS effects in the five-capital cities. These are the manufacturing industries (large negative RS effect), the wholesale trade industry (also a large negative RS effect) and the construction industry (a relatively large positive RS effect). Manufacturing Industry Employment Large RS effects were recorded in the manufacturing industries sector in the five-capital cities. Overall, there was a negative regional shift of −20 838 jobs in the manufacturing employment over the decade 1996 to 2006 (see Table 12.6). This represented 3 per cent of the 1996 level of employment in the manufacturing industry sector. But there were marked differences in the RS effect between the cities: ●

The negative RS effects took place in Sydney (−22 609 jobs representing 11 per cent of the 1996 employment level), Melbourne (−21 824 jobs representing 9 per cent of the 1996 employment level) and

STIMSON PAGINATION (M2469).indd 253

20/12/2010 15:12

254

Endogenous regional development

Table 12.6

Regional shifts in the manufacturing industry, 1996–2006 Total Shift (no.)

Sydney Melbourne Brisbane Adelaide Perth Five-capital cities Rest of Australia

Regional Shift Effect (no.)

Contribution to Total Regional Shift (%)

Regional Shift c.f. 1996 Employment Level (%)

−16 270 −14 537 18 766 −362 11 440 −963

−22 609 −21 824 16 387 −2 434 9 642 −20 838

20 −95 25 14 39 133

−11 −9 22 −4 17 −3

30 080

20 838

39

8

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

●

●

Adelaide (2434 jobs representing 4 per cent of the 1996 employment level). In contrast, there was a positive RS effect in Brisbane (+16 387 jobs representing 22 per cent of the 1996 employment level) and Perth (+9642 jobs representing 17 per cent of the 1996 employment level). The rest of Australia recorded a positive RS in the manufacturing industries of 20 838 jobs or 8 per cent of the 1996 employment level. Most of this took place in the Gold Coast Statistical Division (+7266 jobs) and the Sunshine Coast Statistical Division (+2197 jobs).

In essence, employment in the manufacturing industry has shifted from the two largest capital cities in Australia (and to a minor degree Adelaide) to the rest of the country. This has occurred due to endogenous factors within these places. Wholesale Trade Industry Employment One of the offsetting factors for the negative RS effect in the five-capital cities in aggregate was the wholesale trade industry (see Table 12.7). The wholesale trade industry recorded a positive RS effect in employment of +8629 jobs for the period from 1996 to 2006. Again the performance varied between the cities:

STIMSON PAGINATION (M2469).indd 254

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

Table 12.7

255

Regional shifts in wholesale trade industry, 1996–2006

Sydney Melbourne Brisbane Adelaide Perth Five-capital cities Rest of Australia

Regional Shift c.f. 1996 Employment Level (%)

Total Shift (no.)

Regional Shift Effect (no.)

Contribution to Total Regional Shift (%)

−6 421 5 514 −1 827 −4 381 −2 725 −9 840

503 10 960 7 07 −2 867 −674 8 629

– 48 1 17 −3 −55

0 13 2 −12 −2 3

−16 473

−8 629

55

−7

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

●

●

●

Most of this positive RS performance was attributable to the strong positive RS effect in Melbourne (+10 960 jobs or 13 per cent compared with its 1996 employment level). There was a smaller positive RS effect recorded in Brisbane (+707 jobs or 2 per cent compared with its 1996 employment level) and Sydney (+503 jobs or less than 1 per cent compared with its 1996 employment level). Falls were recorded in Adelaide (−2867 jobs or 12 per cent compared with its 1996 employment level) and Perth (−674 jobs or 2 per cent compared with its 1996 employment level). In the rest of Australia there was a negative RS effect on employment, with much of it in New South Wales (with 27 per cent of the rest of Australian negative RS effect located in the Hunter and Illawarra Statistical Divisions).

Construction Industry Employment The next highest RS effect for employment impact in the five-capital cities was in the construction industry (see Table 12.8). On average, the fivecapital cities recorded a negative RS effect of 7468 jobs or 3 per cent of the 1996 level of employment in the industry. This was entirely due to Sydney: ●

For Sydney, over the period 1996 to 2006, it was equal to +22 395 jobs, which contributed 20 per cent of its total negative RS. It

STIMSON PAGINATION (M2469).indd 255

20/12/2010 15:12

256

Endogenous regional development

Table 12.8

Regional shifts in the construction industry, 1996–2006 Total Shift (no.)

Sydney Melbourne Brisbane Adelaide Perth Five-capital cities Rest of Australia

Regional Shift Effect (no.)

Contribution to Total Regional Shift (%)

Regional Shift c.f. 1996 Employment Level (%)

30 059 43 507 26 134 12 856 23 359 3 039

−22 395 3 687 4 566 2 503 4 172 −7 468

20 16 7 −15 17 48

−22 5 11 12 11 −3

3

7 468

48

7

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

●

●

represented 22 per cent of the 1996 level of employment in the construction industry. All other capital cities recorded positive RS effects in the construction industry, with Brisbane recording the highest (+4566 jobs or 11 per cent of 1996 levels of employment), followed by Perth (+4172 jobs or 11 per cent of 1996 levels of employment), Melbourne (+3687 jobs or 5 per cent of 1996 levels of employment) and Adelaide (+2503 jobs or 12 per cent of 1996 levels of employment). There was a positive RS effect of +7468 jobs in the rest of Australia, which is equal to 7 per cent of this industry’s level of employment in 1996.

FOCUSING ON EACH CAPITAL CITY Each individual capital is now examined to identify the impact of each industry sector on the total RS effect. Sydney Sydney is the capital city with the largest negative RS effect on employment of −122 993 over the decade 1996 to 2006, which compares with a total shift of +219 197 jobs (see Table 12.9). A little under two-thirds of the negative RS effect is attributable to four industries:

STIMSON PAGINATION (M2469).indd 256

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

Table 12.9

257

Regional shift effects in Sydney, 1996–2006 Total Shift (No.)

Agriculture\ forestry & fishing Mining Manufacturing Electricity\ gas\ water & waste services Construction Wholesale trade Retail trade Accommodation & food services Transport\ postal & warehousing Information media & telecommunications Financial & insurance services Rental\ hiring & real estate services Professional\ scientific & technical services Administrative & support services Public administration & safety Education & training Health care & social assistance Arts & recreation services Other services Inadequately described/Not stated Total employment

Regional Shift Effect (No.)

Contribution to Total Regional Shift (%)

Regional Shift c.f. 1996 Employment Level (%)

−2 623

−1 113

1

−16

−323 −16 270 2 194

−1 147 −22 609 −1 754

1 20 2

−33 −11 −13

30 059 −6 421 38 505 15 961

−22 395 503 −14 346 −4 434

20 0 13 4

−22 0 −9 −4

9 137

−8 636

8

−9

−2 135

2 757

−2

5

21 093

3 455

−3

3

6 796

−1 600

1

−6

31 770

−7 539

7

−6

9 819

−9 599

8

−17

17 769

−8 916

8

−10

25 391 37 938

−2 123 −12 465

2 11

−2 −8

2 686

−1 821

2

−7

−3 196 1 047

−4 034 4 823

4 −4

−5 9

219 197

−112 993

100

−7

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

STIMSON PAGINATION (M2469).indd 257

20/12/2010 15:12

258 ● ● ● ●

Endogenous regional development

manufacturing industry (−22 609 jobs or 20 per cent of total RS effect); construction industry (−22 395 jobs or 20 per cent of total RS effect); retail trade industry (−14 346 jobs or 13 per cent of total RS effect); health and community services and social assistance industries (−12 465 jobs or 11 per cent of total RS effect).

Despite this there were three industry sectors that recorded positive RS effects, namely: ● ● ●

financial and insurance services industries (+3455 jobs or 3 per cent of total RS effect); information, media and telecommunications industries (+2757 jobs or 2 per cent of total RS effect); wholesale trade industry (+503 jobs or less than 1 per cent of total RS effect).

Proportionately, and compared with 1996 employment levels, the highest negative RS effect in Sydney was in the mining industry (−33 per cent), followed by the construction industry (−22 per cent) and the administrative and support services industries (−17 per cent). The highest positive RS effect in Sydney was in the information, media and telecommunications industries (+5 per cent), the financial and insurance services industry (3 per cent) and the wholesale trade industry (less than 0 per cent). Melbourne Compared with Sydney, Melbourne recorded a modest positive RS effect of +22 976 jobs (compared with a total shift of +286 066 jobs) over the decade 1996 to 2006 (see Table 12.10). Most of the positive RS effect was attributable to these industry sectors: ● ● ●

wholesale trade industry (+10 960 jobs or 48 per cent of the total RS effect); accommodation and food services industries (+6481 jobs or 28 per cent of total RS effect); health care and social assistance industries (+5351 jobs or 23 per cent of total RS effect).

As with Sydney, in Melbourne there was a very large negative RS effect in manufacturing industries (−21 824 jobs or 95 per cent of the total RS effect). Other industries that recorded a negative regional RS were:

STIMSON PAGINATION (M2469).indd 258

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

Table 12.10

259

Regional shift effects in Melbourne, 1996–2006 Total Shift (No.)

Agriculture\ forestry & fishing Mining Manufacturing Electricity\ gas\ water & waste services Construction Wholesale trade Retail trade Accommodation & food services Transport\ postal & warehousing Information media & telecommunications Financial & insurance services Rental\ hiring & real estate services Professional\ scientific & technical services Administrative & support services Public administration & safety Education & training Health care & social assistance Arts & recreation services Other services Inadequately described/ Not stated Total employment

Regional Shift Effect (No.)

Contribution Regional Shift to Total c.f. 1996 Regional Employment Shift (%) Level (%)

−2 415

−823

−4

−11

−272 −14 537 3 378

−988 −21 824 880

−4 −95 4

−33 −9 10

43 507 5 514 50 417 21 558

3 687 10 960 3 997 6 481

16 48 17 28

5 13 3 9

14 931

2 808

12

4

−1 526

2 097

9

5

14 261

2 715

12

4

7 283

2 193

10

13

32 185

1 490

6

1

17 899

3 209

14

8

16 263

−3 946

−17

−6

29 609 43 280

5 351 1 464

23 6

5 1

6 188

2 153

9

10

−1 997 540

−2 708 3 781

−12 16

−4 8

286 066

22 976

100

2

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

STIMSON PAGINATION (M2469).indd 259

20/12/2010 15:12

260 ● ● ● ●

Endogenous regional development

public administration and safety industries (−3946 jobs or 17 per cent of total RS effect); other services industries (2708 jobs or 12 per cent of total RS effect); mining industry (988 jobs or −4 per cent of total RS effect); agriculture, forestry and fishing industries (823 jobs or 4 per cent total RS effect).

Proportionately, when compared with 1996 employment levels, the wholesale trade and the rental, hiring and real estate services recorded the highest positive RS effect at +13 per cent, followed by the electricity, gas and water supply and the arts and recreation services industries at +10 per cent. The highest negative proportionate RS effects were recorded in the mining industry at −33 per cent, followed by the agriculture, forestry and fishing industries at −11 per cent and the manufacturing industries at −9 per cent. Brisbane Brisbane recorded a strong positive RS effect of +66 805 jobs over the decade 1996 to 2006 compared with a total shift of +202 825 jobs (see Table 12.11). Most of this strong positive RS effect was attributable to these industry sectors: ● ● ●

manufacturing industries (+16 387 jobs or 25 per cent of total RS effect); health care and social assistance industries (+6634 jobs or 10 per cent of total RS effect); retail trade industry (+6222 jobs or 9 per cent of total RS effect).

There was a negative RS effect recorded in the: ● ●

information, media and telecommunications industries (−364 jobs or 1 per cent of total RS effect); agriculture, forestry and fishing industries (−110 jobs or less than 1 per cent of total RS effect).

Proportionately, when compared with 1996 employment levels, the electricity, gas and water supply industries had the highest positive RS effect at +48 per cent, and was followed by the mining industry at +38 per cent and the manufacturing industries at +22 per cent. There were negative proportionate RS effects in the agriculture, forestry and fishing industries at −3 per cent and the information, media and telecommunications industries at −2 per cent.

STIMSON PAGINATION (M2469).indd 260

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

Table 12.11

261

Regional shift effects in Brisbane, 1996–2006 Total Shift (No.)

Agriculture\ forestry & fishing Mining Manufacturing Electricity\ gas\ water & waste services Construction Wholesale trade Retail trade Accommodation & food services Transport\ postal & warehousing Information media & telecommunications Financial & insurance services Rental\ hiring & real estate services Professional\ scientific & technical services Administrative & support services Public administration & safety Education & training Health care & social assistance Arts & recreation services Other services Inadequately described/ Not stated Total employment

Regional Shift Effect (No.)

Regional Contribution Shift to Total c.f. 1996 Regional Employment Shift (%) Level (%)

−1 042

−110

0

−3

1842 18 766 3753

1129 16 387 2 345

2 25 4

38 22 48

26 134 −1827 28 475 11 267

4566 707 6222 3144

7 1 9 5

11 2 9 8

12 057

5337

8

15

−1729

−364

−1

−2

6377

1931

3

8

5191

1515

2

12

15 790

2111

3

4

9180

2459

4

13

16 461

2792

4

6

17 204 28 454

4541 6634

7 10

9 10

1857

172

0

2

2366 1806

2028 3259

3 5

7 16

202 382

66 805

100

10

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

STIMSON PAGINATION (M2469).indd 261

20/12/2010 15:12

262

Endogenous regional development

Perth Perth recorded a positive RS of +24 728 jobs over the decade 1996 to 2006 compared with a total shift of +139 700 jobs (see Table 12.12). This was mainly attributable to these industry sectors: ● ● ●

manufacturing industries (+9642 jobs or 39 per cent of total RS effect); mining industry (+6589 jobs or 27 per cent of total RS effect); spublic administration and safety industries (+5356 jobs or 22 per cent of total RS effect).

The highest negative RS effects were in: ● ● ●

financial and insurance services industries (−1657 jobs or 7 per cent of total RS effect; administrative and support services industries (−1252 jobs or 5 per cent of total RS effect); arts and recreation services industries (−774 jobs or 3 per cent of total RS effect).

Proportionately, when compared with 1996 employment levels, the mining industries recorded the highest positive RS effect at +74 per cent, followed by both the manufacturing industries and the public administration and safety industries at +17 per cent. The highest negative proportionate RS effect was recorded in the agriculture, forestry and fishing industries at −13 per cent, followed by the arts and recreation services industries at −9 per cent, and the financial and insurance services industries at −8 per cent. Adelaide In Adelaide over the decade 1996 to 2006 there was a negative RS effect of −17 184 jobs compared with a total shift of +68 374 jobs (see Table 12.13). Marginally over half of this negative RS effect is attributable to these industry sectors: ● ● ●

health care and social assistance industries (−3366 jobs or 20 per cent of total RS effect); education and training industries (−2889 jobs or 17 per cent of total RS effect); wholesale trade industry (−2867 jobs or 17 per cent of total RS effect).

STIMSON PAGINATION (M2469).indd 262

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

Table 12.12

263

Regional shift effects in Perth, 1996–2006 Total Shift (No.)

Agriculture\ forestry & fishing Mining Manufacturing Electricity\ gas\ water & waste services Construction Wholesale trade Retail trade Accommodation & food services Transport\ postal & warehousing Information media & telecommunications Financial & insurance services Rental\ hiring & real estate services Professional\ scientific & technical services Administrative & support services Public administration & safety Education & training Health care & social assistance Arts & recreation services Other services Inadequately described/ Not stated Total employment

Regional Shift Effect (No.)

Regional Contribution Shift to Total c.f. 1996 Regional Employment Shift (%) Level (%)

−1517

−563

−2

−13

8731 11 440 1738

6589 9642 257

27 39 1

74 17 5

23 359 −2725 20 309 6912

4172 −674 899 67

17 −3 4 0

11 −2 2 0

4757

6

0

0

−1723

−696

−3

−6

2134

−1657

−7

−8

2871

−504

−2

−4

13 175

2102

8

5

5152

−1252

−5

−7

14 745

5356

22

17

11 435 20 521

704 2266

3 9

2 4

859

−774

−3

−9

724 −3197

426 −1639

2 −7

2 −8

139 700

24 728

100

4

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

STIMSON PAGINATION (M2469).indd 263

20/12/2010 15:12

264

Table 12.13

Endogenous regional development

Regional shift effects in Adelaide, 1996–2006 Total Shift (No.)

Agriculture\ forestry & fishing Mining Manufacturing Electricity\ gas\ water & waste services Construction Wholesale trade Retail trade Accommodation & food services Transport\ postal & warehousing Information media & telecommunications Financial & insurance services Rental\ hiring & real estate services Professional\ scientific & technical services Administrative & support services Public administration & safety Education & training Health care & social assistance Arts & recreation services Other services Inadequately described/ Not stated Total employment

Regional Shift Effect (No.)

Regional Contribution Shift to Total c.f. 1996 Regional Employment Shift (%) Level (%)

33

648

−4

22

1098 −362 1438

788 −2434 327

−5 14 −2

61 −4 8

12 856 −4381 15 413 4017

2503 −2867 527 −1170

−15 17 −3 7

12 −12 1 −5

2968

−510

3

−3

−1 022

−179

1

−2

2512

−365

2

−2

858

−1121

7

−17

5966

−1297

8

−5

3897

−896

5

−7

10 163

2395

−14

9

5586 14 293

−2889 −3366

17 20

−8 −6

−13

−1225

7

−18

−1122 −5824

−1361 −4691

8 27

−6 −30

68 374

−17 184

100

−4

Source: Compiled by the author using data available from the ABS 2006 Census of Population and Housing data.

STIMSON PAGINATION (M2469).indd 264

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

265

The industry sectors with positive RS effects were: ● ● ●

construction industry (+2503 jobs or 15 per cent of total RS effect); public administration and safety industries (+2395 jobs or 14 per cent of total RS effect); mining industry (+788 jobs or 5 per cent of total RS effect).

Proportionately, when compared with 1996 employment levels, the arts and recreation services industries recorded the highest negative RS effect at −18 per cent, followed by the rental, hiring and real estate services industries at −17 per cent and the wholesale trade industry at −12 per cent. In contrast, the mining industry had the highest proportionate positive RS effect at +61 per cent, followed by the agriculture, forestry and fishing industries at +22 per cent, and the construction industry at +12 per cent.

CONCLUSION The research reported in this chapter found that negative regional shift effects in employment in Sydney and Adelaide have been the drivers of relatively poor employment growth in those cities over the decade 1996 to 2006. In Sydney, Australia’s largest city, this was particularly so in the final half of that period after 2001. For Brisbane and Perth, the regional shift effects were positive over this period. These findings fit within the broader context of contemporary regional economic trends during this period. Given its size, the impact of Sydney’s relatively poor regional employment shift effects are important to the employment performance of the nation as a whole. Most of the negative regional shift effect in Sydney over the period was attributable to the manufacturing, construction and retail trade industry sectors. Comparatively, most of the positive regional shift effect in Brisbane – the nation’s rapidly growing ‘sub-belt’ metropolis – has been attributable to the manufacturing, health care and social assistance industry and the retail trade industry sectors. The national economy has experienced significant changes in its industrial structure in recent decades. This is also true for employment in the five-capital cities (those with populations in excess of 1 million) of Australia. One of the key determinants of change in employment for capital cities has been regional shift effects (i.e., due to endogenous factors). The research reported here has examined the regional shift effects on employment in the capital city regions for the period from 1996 to 2006, which was the last decade or so of the long boom experienced by Australia from the early 1990s.

STIMSON PAGINATION (M2469).indd 265

20/12/2010 15:12

266

Endogenous regional development

Future research may examine these regional shift effects for different count methods of employment. The optimal method would be to use place of employment (i.e., where the jobs are). Additionally, techniques to measure production at the capital city level may allow future research into regional shift effects for production in the capital cities.

NOTES * 1. 2. 3. 4.

The research reported in this chapter is part of a wider project analysing endogenous regional growth performance across Australia funded in a grant to Robert J. Stimson under the Australian Research Council Discovery Scheme Grant # DP0879819. As defined by Statistical Divisions. By place of enumeration count method. The southern region refers to the Gold Coast Statistical Division, the western region refers to the West Moreton Statistical Division, and the northern region refers to the Sunshine Coast Statistical Division. Place of usual residence method is available for most statistical divisions. This data needs to currently be downloaded one Statistical Division at a time – consuming a significant amount of time to download all the required data. For most Statistical Divisions there is little difference between the two count methodologies. On balance, I determined that the place of enumeration count method was the optimal count method to use at this point in time.

REFERENCES Australian Bureau of Statistics (1997), Australian Demographic Trends, Catalogue No. 3102.0, Canberra. Australian Bureau of Statistics (2006), 2901.0 Census Dictionary, 2006 (Reissue), Canberra. Australian Bureau of Statistics (2007), 2006 Census of Population and Housing, Catalogue No. 2069.0.30.003 – Census of Population and Housing: Time Series Profile Datapack, 2006, Canberra. Australian Bureau of Statistics (2009), 6262.0 - Information Paper: Regional Labour Force Statistics, February 2009, Canberra. Chalmers, J.A. (1971), ‘Measuring Changes in Regional Industrial Structure: A Comment on Stilwell and Ashby’, Urban Studies, 8(3), 289–92. Edwards, J.A., Harniman, K.F. and Morgan, J.S. (1978), ‘Regional Growth and Structural Adaption: A Correction to the Stilwell Modification’, Urban Studies, 15(1), 97–100. Haynes, K.E. and Dinc, M. (1997), ‘Productivity Change in Manufacturing Regions: A Multifactor/Shift-Share Approach’, Growth and Change, 28(8), 201–21. Harding, A., Hellwig, O. and Lloyd, R. (2000), ‘Regional Divide? A Study of Incomes in Regional Australia’, 29th Conference of Economists, Gold Coast, 3–6 July 2000. Jefferson (1939), ‘The Law of the Primate City’, Geographical Review, 29, 226–32.

STIMSON PAGINATION (M2469).indd 266

20/12/2010 15:12

Endogenous growth and decline in Australian capital cities

267

Markusen, A.R., Noponen, H. and Driessen, K. (1991), ‘International Trade, Productivity, and US Job Growth: A Shift-Share Interpretation’, International Regional Science Review, 14(1), 15–39. Mitchell, B. (2008), ‘Exploring Regional Disparities in Employment Growth’, Working Paper, No. 08-12, Centre of Full Employment and Equity, University of Newcastle, Australia. Noponen, H., Markusen, A. and Driessen, K. (1997), ‘Trade and American Cities: Who Has the Comparative Advantage?’, Economic Development Quarterly, 1(1), 67–87. Productivity Commission (1998), Aspects of Structural Change in Australia, Canberra, Australia. Stilwell, F.J.B. (1969), ‘Regional Growth and Structural Adaptions’, Urban Studies, 6(2), 162–78. Stimson, R.R., Robson, A.P.L. and Shyy, T. (2005), ‘Modeling Determinants of Spatial Variation in Regional Endogenous Growth: Non-metropolitan Regions in the Mainland States of Australia’, in Proceedings of 29th Annual Conference of the Australian and New Zealand Regional Science Association International, Sydney, Australia. Stimson, R.R., Robson, A.P.L. and Shyy, T. (2009), ‘Modeling Regional Endogenous Growth: An Application to the Non-metropolitan Regions of Australia’, Annals of Regional Science, 43(2), 379–98. Wadley, D. and Smith, P. (2003), ‘Straightening up Shift-Share Analysis’, Annals of Regional Science, 37(2), 259–61.

STIMSON PAGINATION (M2469).indd 267

20/12/2010 15:12

13.

A case study approach to investigating local development initiatives in rural small towns in Victoria John Martin*

INTRODUCTION An interest in endogenous development comes out of the research undertaken on Australian small towns in recent years, in particular research on small towns in the state of Victoria.1 Studying small towns is problematic because they are, paradoxically, similar in many ways, but cannot be treated as such (Powers, 1991). Across Victoria there are some 300 small towns with populations ranging from 250 to 10 000. The future of these places has been the basis of much contentious discussion in recent years (see Sher and Sher, 1994; Forth, 2000; O’Connor et al., 2001; Stimson, 2002) and the continuing public policy response seems to be to leave them to the competing forces of globalization and their own capacity to prosper and survive. Of course, there are also many smaller communities, communities with fewer than 250 people, simply categorized by the demographers as ‘rural other’. A key policy question facing Australian state governments in recent decades has been how to assist small towns with the considerable change impacting them (Garlick, 2005). That includes change in the scale of agriculture, more extreme climatic regimes, innovations and improvements in transportation and communications technologies and the ageing of people living in these places as the flight of youth takes this cohort to education and employment opportunities and the excitement of larger cities. A clear policy preference by state governments for these places has come out of the social capital literature, especially that reflecting Robert Putnam, a regular speaker at state and local government forums in Australia, where he advocates intervention strategies based on participation and engagement (see Eversole and Martin, 2005 for recent Australian cases). Yet when one looks closely at these programmes one cannot find clear measures as

268

STIMSON PAGINATION (M2469).indd 268

20/12/2010 15:12

Initiatives in rural small towns in Victoria

269

to their impact in the places they are applied. Most evaluations are based on opinion surveys with central players with no real tangible evidence of improved or sustainable development relating to investment, production, employment and the overall standard of living of people in these communities. The initial idea for this chapter grew out of the idea of being able to identify how intra-regional endogenous growth is measured, thus influencing the focus on local policy development. In his keynote address at the annual conference of the Australian and New Zealand Regional Science Association International (ANZRSA) held in Manukau in September 2007, Roger Stough referred to the doctoral research of Mark De Santis (1993) on intra-regional endogenous growth. De Santis’ research was on the relationship between leadership, resource endowment and exogenous and endogenous drivers of regional economic development. The author became interested in how it might apply to better understanding of small town development. The focus was on inter-regional endogenous growth at a scale several times greater than our interest in small rural towns. Herein lay the complication I was to discover when researching small town development strategies in two rural shires in Victoria. The question that arose was: ‘Can the factors informing inter-regional success be applied to small rural towns on an intra-region comparison?’ Local government policy-makers, who are the place managers responsible for ongoing support for these small towns, would like to know whether these factors also apply. Stimson et al. (2005) have proposed their ‘virtuous circle for sustainable regional development’, which also reflects De Santis’ findings. Thus, the question is: ‘Do they apply to intra-regional assessment of small towns in large rural shires?’ After a brief discussion of methodology, there is a discussion of some of the literature relating to community economic development. That is followed by a review of two rural shires, and several towns in each shire, which are used as case studies. There is then a discussion of both the relevance of the social capital concept and measurement and the applicability of the ‘virtuous circle for sustainable regional development’ model as it applies to the four case study small towns. The focus in the chapter reflects a wider research interest in indicators for small town development, which represents a much smaller spatial scale of analysis than is usual in the conventional studies on regional development.

PURPOSE AND METHOD In the De Santis study referred to above, the sample was 35 metropolitan areas in the US, each with an urban centre of at least 50 000 people and

STIMSON PAGINATION (M2469).indd 269

20/12/2010 15:12

270

Endogenous regional development

a total population of at least 100 000. While the conceptual framework driving his research was interesting, the regional scale was much greater than that faced in dealing with small towns in Victoria. The regional shires in Victoria range in size from 15 to 25 000 people in total, with maybe one urban centre approaching 5 000 to 10 000 people, and all-up these shires might be lucky to have a total of 15 000 people. The overall population density is low. These are rural places with overall an ageing and mostly static population, with perhaps some growth on the back of city dwellers making lifestyle purchases with their rural block or holiday house. In focusing on small rural towns, the question arises as to why does one town do so much better than another of similar size and apparent resources? The Australian inter-regional comparisons have been ably done by others (see, for instance, O’Connor et al., 2001). Comparing small towns within Australian regions (local government authorities) is not something that has been systematically researched in the way that the De Santis study in the US did by analysing change over a decade. For the study reported here the author was curious to see if it was possible to identify the criteria by which one could identify and evaluate change in order to compare inter-town performance and success within shires. As part of the research, secondary sources on small town development were obtained for two Victorian shires – identified as Shire A and Shire B for purposes of confidentiality – for reasons that will become apparent when I report on how challenging it is to get information from local governments, even when one of them has won a national award for its programme and the other has presented itself at regional conferences as being an innovator and leader in small town engagement and development. It is clear that there is a managerial and political overlay at the micro-level that comes into play, therefore it is necessary to protect the people in these small towns who could easily be identified. With the focus on local government processes, what is being sought is to identify successes and failures in the eyes of local government officials responsible for these programmes aimed at enhancing local development. It would have been possible to have gone to a local government shire that did not have a comprehensive small town engagement strategy across its jurisdiction, but that would have taken more time and resources to get information on different town performance. So the primary focus was to begin to identify those factors that are critical to the success of small rural towns (in the state of Victoria), and to measure those factors and determine how to use them in any assessment of small town performance within a local government jurisdiction.

STIMSON PAGINATION (M2469).indd 270

20/12/2010 15:12

Initiatives in rural small towns in Victoria

271

THEORETICAL VIEWS ON ENDOGENOUS DEVELOPMENT De Santis’ (1993) conclusion to his intra-regional comparison of US cities referred to above is summarized in Table 13.1. High economic performance was found in those places where there were both high levels of leadership and resource endowments. De Santis’ findings thus support the sometimes hypothesized causal relationship between social capital and regional economic performance. Places rich in social capital – in this case leadership – are more likely to have higher economic performance when they are also endowed with appropriate material resources, than places that have less amounts of these two factors. In the social capital literature it is typically presented as a metaphor for advantage (Burt, 2001). The more social capital a place has the more likely it will succeed on a number of grounds, including economic performance. More recent research by Eastwood (2007) has shown that the ‘combination of creative entrepreneurs networking in an economic free environment’ (p. 157) is the basis for sustainable economic development. The question is: ‘Does this align with De Santis’ conclusion that effective leadership in resource-endowed cities leads to high economic performance?’ I would suggest that these two empirical studies are not at odds with each other. A more general assessment is that in places endowed with economic resources, however widely defined, and where individual creativity is valued and fostered, economic growth and development is more likely to occur than in places that do not have these characteristics. The research focus in the study of small towns in Victoria discussed Table 13.1

Relationship between leadership, resource endowments, and economic performance Leadership

Resource Endowments

High

Low

Source:

High

Low

High economic performance Moderate economic performance

Moderate economic performance Low economic performance

Compiled by the author and derived from De Santis (1993, p. 91).

STIMSON PAGINATION (M2469).indd 271

20/12/2010 15:12

272

Endogenous regional development

here was to identify the factors that contribute to small town success – including social and economic success – with the aim being to inform policy-makers as to how best they might work with such places for their continuing sustainability. Defining success is thus critical. The criteria will determine what to measure, and begin to value. Stimson et al. (2005) presented their ‘virtuous circle for sustainable regional development’ with ten factors that constitute the basis for identifying the success of local community approaches to sustainable development. They condense them into leadership, institutions and resource endowments and market fit, and see how they might interact within what is termed a ‘regional competitiveness performance cube’. Those overarching factors are not, I believe, inconsistent with the conclusions drawn by De Santis (1993) and Eastwood (2007). My view is that the ten factors that constitute the conceptual model provide the basis for finding out the local development potential of a small town. For example the Stimson et al. (2005) premise is that strong proactive leadership that helps build a vision for future development is backed up by strategy, plans and processes and facilitates institutional change to enhance regional capacity and capability, then the local development process is enhanced. That requires effective institutions and regional infrastructure, and needs processes and mechanisms for using resource endowments so as to maximize the potential to tap market conditions. That may lead to sustainable development (after Stimson et al., 2005). It was clear in the research on small towns outlined below that many of these factors are evident in their small town development projects. The question is: ‘Why don’t the local shires use them as part of their ongoing evaluation?’ The literature on social capital is now extensive and central in efforts by researchers to understand social structures and processes (Putnam, 1993, 2000; Woolcock, 1998; Lin, Cook and Burt, 2001). Putnam’s argument that the degree of association or participation by individuals in society is related to the success of communities is central to the interests reflected in this chapter. There is a need to know what these processes are in small rural towns of several hundred people and how it impacts local economic development: ‘What are the indicators?’ ‘How do we measure them, and what does it tell us about what governments can do to facilitate local economic success?’ Burt (2001) also raises the idea of ‘brokerage’ across structural holes as the source of value-added, and ‘closure’ can realize value buried in holes. He contrasts processes of ‘external’ and ‘internal lack of constraint’ (see Figure 13.1). The sociometry of these relationships is presented diagrammatically in Figure 13.2. As will be seen from the discussion below, it is the

STIMSON PAGINATION (M2469).indd 272

20/12/2010 15:12

Initiatives in rural small towns in Victoria

273

D

High

A Disintegrated group of diverse perspectives, skills, resources

External lack of constraint

Minimum performance

Low Low

Maximum performance Cohesive group containing only one perspective, skill, resource

Internal lack of constraint

B

High

Source: Compiled by the author and derived from Burt (2001, p. 48).

Figure 13.1

Social capital matters

balance between external and internal attention that seems to be a critical factor in a community’s success in garnering development support. In his extensive theoretical discussion of the relationship between social capital and economic development, Woolcock (1998, p. 153) has commented on both the epistemology underpinning research and the scale at which it is focused. He asserts that: The subsequent discrediting of modernization theory in both its sociological and economic guise, however, and its replacement, respectively, by world systems and neo-classical growth theories have led to a situation where the units of analysis in contemporary studies of development are either nationstates and transnational corporations or ‘rational’ individuals and firms. The contribution of civil society and other institutional arrangements mediating the space between states and markets has been lost, incorporating themselves neither into a coherent body of knowledge nor sensitive and sensible policy prescriptions.

Woolcock (1998, p. 187) also provides a cautionary conclusion to the relationship between social structures and processes and development success. He asserts that: the mediating variable is the extent to which a mutually beneficial interaction coordinates specific levels, dimensions, and combinations of social relationships. The structure of the state, the nature and extent of its involvement in civic and corporate life, and the organization of society together constitute the key factors determining whether a country succeeds or fails in development.

Sociological studies that attempt to identify those associational factors that influence development in small communities also reflect Woolcock’s

STIMSON PAGINATION (M2469).indd 273

20/12/2010 15:12

274

Endogenous regional development

D

B

A Source: Compiled by the author and derived from Burt (2001, p. 48).

Figure 13.2

The sociometry of the relationships ‘social capital matters’

caution. In their excellent research on the structure of access to embedded resources in Canadian First Nations communities, Enns et al. (2008) apply Lin, Fu and Hsung (2001) position generator to these small coastal communities. Position generators:

STIMSON PAGINATION (M2469).indd 274

20/12/2010 15:12

Initiatives in rural small towns in Victoria

275

use a sample of ordered structural positions salient in a society and ask respondents to indicate contacts, if any, in each of these positions [thus] it becomes possible to construct measures of (1) range of accessibility to different hierarchical positions; (2) extensity or heterogeneity of accessibility to different positions; and (3) upper reachability of accessed social capital. (Lin, Fu and Hsung, 2001, p. 63; italics original)

Enns et al.’s (2008) findings support Putnam’s (1993, 2000) view that: ‘civic participation is related to the development of critical networks that are crucial in gaining access to embedded resources and developing social capital’ (quoted in Enns et al., 2008, p. 37). They also report a limitation to this social research technique – of interest to the research question being asked in this chapter – that: ‘the position generator is not able to explicitly capture the social processes whereby people utilize their network ties for social benefit’ (ibid., p. 36) Nevertheless, Enns et al.’s (2008) research is worthy of greater consideration in investigating the relationship between intra-region, or shire council, jurisdictions, as it tells us something about the relationship between, for example, gender, education and the strength of ties people have both within and outside their communities. Enns et al. support both Leonard and Onyx (2003) and Woolcock (1998) who also advocate for balanced networks with no particular emphasis on one type of tie in a given location . . . involvement in groups that span communities leads to both weak and strong ties located both inside and outside the community, and thus allows residents to form diverse, stable networks to resource-rich persons. (Enns et al., 2008, p. 38)

For Woolcock (1998) it is the intra-community (bonding) extracommunity  (bridging) ties working together that differentiates success between communities. Importantly Woolcock asserts the important link between social capital and economic success: you can’t have the latter without the former. Thus, the question is: ‘Can we assume that economic development within small towns will follow if there is strong social capital?’

THE CASE STUDY SHIRE COUNCILS2 Two rural/regional local government areas (called shires) in the state of Victoria, each containing numerous small towns, were chosen as the case studies for the research discussed here. They have these common characteristics:

STIMSON PAGINATION (M2469).indd 275

20/12/2010 15:12

276 ● ● ● ●

Endogenous regional development

there is only one large town of approximately 8000 population; numerous other small towns with a population above 250 (as well as many locales with fewer than 250 people); significant agricultural activity; an ageing population.

Both have active community development strategies.3 Victoria is the most closely settled state of mainland Australia, and the state government has had a deliberate small towns development strategy working through its dependent local government councils. Of the two councils one (Shire B) is a national award winner for its community development programme. In 2006 it was committing more than A$200 000 per year (for three years) to support Local Community Planning Groups representing ten communities and 100 localities. The other (Shire A) is also regarded across the state as a leader in this field. Statistical data and documents were reviewed, and the research team members spoke by phone to key players in the councils. Shire A Shire A has labelled its small towns programme as ‘The Community Connect Model’. The shire identified several characteristics as underpinning this programme. That includes the following: ● ● ● ●

open and honest communication between council and communities (trust?); plans reflect community vision (a shared view?); community control of the process (involvement); collaborative opportunities for council, communities and other agencies to work together (engagement).

The focus of the strategy was to engage citizens in a discussion about facilities and local government service improvement in each town. A deliberate planning process was set in place, which resulted in specific plans for each town. The processes used in Shire A were to: ● ● ● ● ●

conduct workshops to explore vision; learn how to plan and work together; support and learn from each other; identify community priorities and projects; apply for funding from state government.

STIMSON PAGINATION (M2469).indd 276

20/12/2010 15:12

Initiatives in rural small towns in Victoria

277

It is clear from Shire A’s plans and publicity information that it places a heavy emphasis on community engagement. The council’s first goal in its 2006–10 plan is: ‘Vibrant Communities: To create healthy, informed and proud communities’. Indicators of success identified are: ● ● ● ● ● ● ● ● ● ●

civic centre redevelopment; attendance at a community event in past six months; volunteer rate; community satisfaction with health and human services; immunization rates; reportable incidences of communicable diseases; percentage of population who feel they have opportunities to participate in affordable local community events; membership of an organized group; community satisfaction rating with recreation facilities; community satisfaction rating for local roads and footpaths.

These are outcome, output and process measures, taken in large part from state government indicators linked to funded programmes. They give no clear view of the overarching conceptual framework in which they are located. It is suggested that this is why the attempt to define and use a wide range of social indicators in local governments in Victoria largely falls on deaf ears at the local government level as there is no apparent utilization in the local decision-making. Shire A’s community plan development process is the primary source of information used for these two case study towns in the shire. Each town was engaged by the shire in a planning process that established a ‘community contact team’. Other key stakeholders for the planning process were identified by this group. The shire then developed a community profile for each town. Those two towns are now discussed. Town A1 In the case of Town A1 the profile lists population over time, age cohorts, employment, buildings, birth rate and a half-page profile, which reads like a tourism brochure. The community future recognizes that the town is on the edge of the growth area that is the southeast corridor of the Melbourne metropolitan region. There are and will continue to be a significant proportion of residents who commute elsewhere for employment. Economic activity reflects the changing land use in small towns around two hours’ drive from metropolitan Melbourne. While dairying and beef production remain prominent agricultural industries the town now has many

STIMSON PAGINATION (M2469).indd 277

20/12/2010 15:12

278

Endogenous regional development

self-employed tradespeople as well as small home-based businesses with people who live and work in the area. The community plan development process follows a strict formula: revisit our vision statement (who do we want to be) followed by a workshop process that identifies priority areas, projects, ‘scoring’ or polling participants to determine priorities. At a follow-up workshop after the shire officers processed the initial discussions, the participants reviewed and consolidated projects and priorities. The third and final workshop developed action plans and agreed the top eight projects. This ‘completed the series of workshops for input into the development of the [Shire’s] Community Plan’ (Town A1 Community Plan). The eight Community Plan priority areas and projects are: ● ● ● ● ● ● ● ●

sport and recreation; public infrastructure; public facilities and services; town beautification; business development; walking and cycling tracks; roads; education.

The projects all concern the provision of infrastructure in one form or another; for example, skateboard park, community centre, street scaping and parks, walking and cycling tracks. Importantly this Town A1 has also organized itself – independently of the shire council and its development efforts – to undertake two major local economic development projects. One is the establishment of a Bendigo Bank Community Bank®, which is a regional banking initiative taken more than a decade ago in Bendigo, a regional town in Victoria, and which has spread widely across rural areas of Australia. The other is setting up a town historical society (recognized by the shire council in its community plan, reflecting its involvement, but just how much is not clear). Both of these cases demonstrate that in many small towns, and we suspect this will be the majority,4 that individuals do cooperate to develop local social and economic institutions and facilities in spite of what governments do or do not do. The challenge for local policy-makers is to work with people when such initiative occurs. Town A2 This town is similarly located to Town A1 with respect to its proximity to the metropolitan region, predominant agricultural production around

STIMSON PAGINATION (M2469).indd 278

20/12/2010 15:12

Initiatives in rural small towns in Victoria

279

dairy and beef production and a significant number of commuters enjoying a rural lifestyle on a few acres. The community plan development process, priority areas and projects are also remarkably similar to Town A1. This is not surprising as the shire council developed the approach and applied it across a number of small towns in their jurisdiction. The eight priority areas were: ● ● ● ● ● ● ● ●

sport and recreation; public facilities and services; walking and cycling track and trails; environment; signage; events; town beautification and improvements; strategic planning.

The focus is also on infrastructure such as the improvement of playgrounds, provision of a mower, car parking, walking tracks, improved signage, and retaining walls. The last priority area – strategic planning – has a one-line project ‘Controlled Development in [Town]’. A defining characteristic of the outcomes of these two case study towns in Shire A is the provision of local infrastructure, which, at first glance, appears minor. But in these small towns such local improvements to amenity impact the lives of the local citizens making their town a better place in which to live as well as providing greater amenity for visitors. Shire B The overarching strategy for this Shire B is found in their award-winning ‘Community Development Programme’.5 From the shire’s website they outline the programme as follows: ● ● ● ● ●

a programme that sets out to fully engage people in shaping the future of their community and local economy; a public community meeting develops the plan for each participating district; public appointment of a community group steers action on the district plan; the district plan takes into account the strengths of each community; council works in partnership with the various groups, working on projects arising from the district plan.

STIMSON PAGINATION (M2469).indd 279

20/12/2010 15:12

280

Endogenous regional development

The community plan is developed as follows: ● ● ●

● ● ● ●

participate in the annual public community meeting to set the plan for the district; join a community planning group to help steer action on your district’s plan; the members of community groups seek out interested community members, to join in the development of projects arising from their district’s plans; actively seek and listen to views; exchange ideas and opinions; exchange information; progress is reviewed at a public meeting each year.

There is a clear process set out by the shire council for community planning. This process provides indicators of success and can be measured. They are largely institutional and should not be overlooked as important factors contributing to endogenous development. Town B1 Located within an irrigated agricultural region, Town B1 has been very successful in leveraging capital works funds from the shire council to attract additional monies to upgrade infrastructure in the town including its business centre and street scaping. The town has also established its own local fuel outlet that operates out of the business centre being staffed by some 70 volunteers to ensure the locals do not have to drive many kilometres to the next largest town for fuel. Significantly many of these volunteers worked together to upgrade the fuel outlet to a modern service station. One of the local town leaders attributes the success to ‘people with ideas, motivations and leadership’. He asserted that ‘it’s all about getting excited about the little things’. Many of these volunteers also help out at the Bush Nursing Centre driving patients to other centres for their particular health needs. In 1992 the community formed a progress association, and one of their first projects was community housing for retirees. They now have 12 units completed, with land for a total of 20 units. Another project was to convert an irrigation drain through the middle of town into an aquatic centre playground for the community and visitors. They now have an ornamental lake, an island, rotunda boardwalk and fountain, which make their town attractive for both residents and visitors. A more recent project is their living heritage complex. This is an historical centre showcasing the development of agriculture in the region. Volunteers have together spent

STIMSON PAGINATION (M2469).indd 280

20/12/2010 15:12

Initiatives in rural small towns in Victoria

281

thousands of hours and also raised tens of thousands of dollars for the centre. This town is clearly one that has been more successful than most in the shire in obtaining resources from external sources as well as from amongst the residents. It is, in fact, held up as a model of small town success and the stories presented in the media typically show a range of citizens taking an active role in their community. As the shire organization applied its community development model to the town it responded readily and was able to capitalize on the processes put in place. Town B2 Town B2 is also located within the irrigation schemes off the Murray River but has been impacted by water restrictions over the last decade. It contains a major milk manufacturing facility, now with international owners who control this factory from the state’s capital city, Melbourne. Like Town B1, Town B2 has also engaged in a series of planning meetings to improve local infrastructure and services, ranging from improvements to the public swimming pool, construction of a multi-purpose stadium adjacent to the local school, road intersection realignments and signage. It has also developed several projects independently from the shire-initiated community planning process. One was to establish a banking service with the Bendigo Bank as part of a longer-term development to establish a community bank. Another recent success was the establishment of a Rural Transaction Centre with support from the Australian government. The leaders of these initiatives, who were marginal to the council-initiated process, also developed a major national campaign ‘Sponsor a Cow’. The district development committee, we are referring to here, created this programme to ‘give city folk and farmers the opportunity to “Build Bridges” between the city and country and for sponsors to learn first hand the extent of this drought and the impact and affect it is having on the whole population’ (from Sponsor a Cow website). This project has support from the Victorian Farmers Federation and Dairy Australia. The sponsorship has a real impact enabling dairy farmers to ‘save some of their breeding stock and maintain some dignity during this time, which is the worst drought in living memory’. While the programme ran for two years and raised more than A$100 000 for participating farmers, the stories of families coming up from Melbourne to visit their cow and meet the dairy farming family tell of a much more powerful outcome: of urban Australians feeling what it is to be a dairy farmer in trying environmental and economic times. Both Town B1 and Town B2 have benefited from the initiatives of the shire council with its award-winning community development planning programme, so much so that it has been put on hold because of funding

STIMSON PAGINATION (M2469).indd 281

20/12/2010 15:12

282

Endogenous regional development

difficulties across the whole shire programme. This suggests that for most small towns their expectations are for funding support from elsewhere for their local projects. Our two case study towns have also shown they are able to work together beyond the (exogenous) shire support to develop their own (endogenous) development strategies.

CONCLUSION The common characteristics of small town development processes in the two case study shires were the structuring of relationships to address the provision of basic infrastructure and local leadership through community planning processes. It is not known which priorities have been agreed by each council and subsequently developed. But what was learned through the studies enquiring into four small towns across two shires in rural Victoria was that local economic development projects have been initiated in some towns, but are not as evident in the others. It cannot be denied that such local initiative does not occur in those other places. It was only in two where a community-wide approach to such development was found. Coincidentally it was in the shire that won a national award for its community development programme a few years earlier. In the other shire, the community planning process is still relatively new. It will be interesting to learn if local economic development initiatives spin off from the community planning processes established by the shire and the communities themselves attend to local social and economic development. What intra-regional endogenous development factors can be effectively evaluated? From the discussion of the social capital literature and a brief review of intra-regional endogenous development in four small towns across two rural shires there appear to be several characteristics that are key to local development: 1.

2.

3.

The application of the position generator to these places will tell local government policy-makers who talks to whom about what and how often they do this. Quite simply is there a sufficient level of engagement in the community such that they can have a dialogue, a conversation about what needs to happen and how they can make it work together? The balance of internal community and external networks needs to be right, and local leaders can influence these networking opportunities once they appreciate the importance of striking the right balance, which will be different for different leaders, at different times working on different issues. Leadership pervades all of these processes. If appropriate individuals

STIMSON PAGINATION (M2469).indd 282

20/12/2010 15:12

Initiatives in rural small towns in Victoria

283

with the skills and motivation to lead are in sufficient number then De Santis’ findings for inter-regional success will also apply for intraregional success. But just how many leaders are required is a vexed question. Of course there is another view that leadership is what is required, something that many people in a community can display. The community planning processes developed by both the case study shires would have clearer conceptual direction if they followed, for example, the virtuous circle for sustainable regional development proposed by Stimson et al. (2005) outlined earlier in this chapter. It may be argued that a comprehensive, over-arching measurement strategy for local development should start with an assessment of social capital using Lin, Fu and Hsung’s (2001) position generator, as demonstrated by Enns et al. (2008). This should be a baseline for future surveys to learn how the external and internal links had changed as a result of an intervention strategy by either the shire council, or when local circumstances change through the actions of locals, regardless of the involvement of government. The Stimson et al. (2005) model for sustainable development is reflected in the small town development strategies implemented by the two shires and four towns reported in this chapter. While such a concept was not revealed by the informants in the scoping study undertaken in the case studies, it is believed that it would certainly resonate with them. The big question is: ‘Would these shires be prepared to enter into a pre- and postevaluation strategy as suggested here?’ Or alternatively: ‘Does local political control dictate such processes in that other agendas, including being seen to take action are as important as any genuine attempt at achieving an outcome for the sustainability of these places?’ It may be that the localized evaluation research may prove all too challenging, for political and managerial reasons, like: ‘Do we have the staff and resources to do this analysis?’ and ‘If we do, then what would we do with this information?’ The concepts and measures are available to monitor local social and economic development. Their use by local government councils, especially in small and under-resourced councils, will be dependent on their understanding of the concepts and tools involved in this type of research and the preparedness of local councillors and professional staff to face both good and bad news that will come from such an assessment.

NOTES *

The author thanks Ms Barbara Beattie who provided excellent research assistance in obtaining information from both shires and the four towns for the research on which this chapter is based.

STIMSON PAGINATION (M2469).indd 283

20/12/2010 15:12

284

Endogenous regional development

1. See The Study of Small Towns in Victoria Revisited, available at: http://www.dse.vic. gov.au/CA256F310024B628/0/E352E96216614FDDCA2572CF0021B941/$File/07_08_ Small_Towns_Revisited.pdf; accessed 28 May 2010. 2. These are referred to as Shire A and Shire B. 3. Although one shire has just signalled to its community leaders that their small towns strategy is currently on hold due to a lack of funding. This notification came about as we started to interview key players and the purpose of our research became clear. This is unfortunate as our intentions are to identify how best to measure intra-regional small town development success to assist local governments in their future planning for these places. 4. See The Study of small towns in Victoria Revisited, available at: http://www.dse.vic. gov.au/CA256F310024B628/0/E352E96216614FDDCA2572CF0021B941/$File/07_08_ Small_Towns_Revisited.pdf; accessed 28 May 2010. 5. From the shire’s website.

REFERENCES Burt, R. (2001), ‘Structural Holes versus Network Closure as Social Capital’, in N. Lin, K. Cook and R.S. Burt (eds), Social Capital: Theory and Research, Aldine De Gruyter, New York, pp. 31–56. De Santis, M. (1993), Leadership, Resource Endowments, and Regional Economic Development, DPA, George Mason University, Fairfax, Virginia. Eastwood, B.M. (2007), ‘Explaining U.S. Urban Economic Growth 2000–2004: The Role of the Creative Class, Social Capital, Economic Freedom, Distributive Politics and Defense Spending’, PhD thesis, West Virginia University, Morgantown. Enns, S., Malinick, T. and Matthews, R. (2008), ‘It’s Not Only Who You Know, It’s Also Where They Are: Using the Position Generator to Investigate the Structure of Access to Embedded Resources’, in N. Lin and B.H. Erickson (eds), Social Capital: An International Research Program, Oxford University Press, New York, pp. 255–81. Eversole, R. and Martin, J. (eds) (2005), Participation and Governance in Regional Development: Global Trends in an Australian Context, Aldershot, UK: Ashgate. Forth, G. (2000), ‘Following the Yellow Brick Road and the Future of Australia’s Declining Country Towns’, paper presented at the First National Conference on the Future of Australia’s Country Towns, La Trobe University, Bendigo, Australia. Garlick, S. (2005), ‘Regional Growth, Enterprising Human Capital and Community Engagement’, Monograph, Institute for Sustainability, Health and Regional Engagement, University of the Sunshine Coast, Queensland, Australia. Leonard, R. and Onyx, J. (2003), ‘Networking Through Loose and Strong Ties: An Australian Qualitative Study’, Voluntas: International Journal of Voluntary and Non-profit Organizations, 14(2), 189–203. Lin, N., Fu, X.-C. and Hsung, R.-M. (2001), ‘The Position Generator: Measurement Techniques for Investigations of Social Capital’, in N. Lin, K. Cook and R.S. Burt. (eds), Social Capital: Theory and Research, Aldine De Gruyter, New York, pp. 57–81.

STIMSON PAGINATION (M2469).indd 284

20/12/2010 15:12

Initiatives in rural small towns in Victoria

285

Lin, N., Cook, K. and Burt, R.S. (eds) (2001), Social Capital: Theory and Research, Aldine De Gruyter, New York. O’Connor, K., Stimson, R.J. and Daly, M. (2001), Australia’s Changing Economic Geography: A Society Dividing, Oxford University Press, Melbourne. Powers, R. (1991), Far From Home: Life and Loss in Two American Towns, Doubleday, New York. Putnam, R.D. (1993), Making Democracy Work: Civic Traditions in Modern Italy, Princeton University Press, Princeton, NJ. Putnam, R.D. (2000), Bowling Alone: The Collapse and Revival of American Community, Simon and Schuster, New York. Sher, J.P. and Sher, K.R. (1994), ‘Beyond the Conventional Wisdom: Rural Development as if Australia’s Rural People and Communities Really Mattered’, Journal of Research in Rural Education, 10(1), 2–43. Stimson, R. (2002), ‘Planning for Fewer People’, available at www.uq.edu.au/ news/?article=3069, accessed 28 May 2010. Stimson, R.J., Stough, R.R. and Salazar, M. (2005), ‘Leadership and Institutional Factors in Endogenous Regional Economic Development’, Investigaciones Regionales, 7, 23–52. Woolcock, M. (1998), ‘Social Capital and Economic Development: Toward a Theoretical Synthesis and Policy Framework’, Theory and Society, 27(2), 151–208.

STIMSON PAGINATION (M2469).indd 285

20/12/2010 15:12

14.

Economic development incentives and the measurement of local endogenous growth: is there a need for modeling adjustment? Terry Clower

INTRODUCTION Over the past quarter-century there has been much addition to and improvement in theoretical constructs of endogenous growth models. Those improvements include addressing human capital characteristics and investments (Romer, 1986), the role of firms in regional economies (Markusen, 1985), innovation diffusion (McCombie, 1982), entrepreneurship (Acs and Armington, 2004), local leadership (Stough et al., 2007) and many others. Of course, this sampling of contributions does not include notable works by many researchers, nor does it cover the extensive theoretical debates centered on regional convergence/divergence. While the theoretical debates and theorem proofs are rich, relatively little of this research has employed empirical techniques to test these theories. The first, and in many respects foremost, challenge in assessing and quantitatively modeling endogenous growth at the local level is determining the best measure of growth to serve as a dependent variable. Unlike national and state-level analyses, which can rely on well-tested measures of gross product output, there are few demonstrably valid measures of regional product available from public sector sources for sub-state regions. Nascent efforts by the US Bureau of Economic Analysis are just beginning to offer a measure of regional product for Metropolitan Statistical Areas, but these data are only available for a few areas/years. The author is not aware of any public sources for local/regional output estimates for Australia, the UK or Canada. While there are estimates of US local area output available from several private vendors, these data can be cost-prohibitive for research requiring hundreds of cases to support complex endogenous growth regression equations.

286

STIMSON PAGINATION (M2469).indd 286

20/12/2010 15:12

Economic development incentives and measurement of growth

287

In response to these data limitations, the majority of local endogenous growth modeling efforts have relied on employment change as the bestavailable measure of growth. The chapter starts with a consideration of the strengths and weaknesses of employment change as a measure of local endogenous growth, highlighting a particular concern regarding the confounding influence of large non-local government incentives targeted to specific firms on employment change measures. This concern is presented first on a conceptual level illustrated by some infamous examples of state governments ‘buying’ jobs and then offered as a simulation exercise focused on a particular site location case in north central Texas. Then follows a discussion on how the influence of non-local incentives could be addressed through alternative or new measures on either the left- or righthand side of endogenous growth model equations.

MEASURES OF LOCAL ENDOGENOUS GROWTH As noted above, the traditional measure for growth has been one or more calculations of gross output. This works well for most national-level studies and is fairly well supported by gross product measures at the state/ provincial level. However, these data are not readily available at the local level. Some researchers have used measures of personal income for measuring local-level growth (see, for example, Gkritza et al., 2007). From a data availability standpoint, employment is the most consistently available measure of local growth, though some caution must be taken based on these measures. For example, in the US, the Department of Commerce offers annual estimates of firm employment by detailed North American Industry Classification System (NAICS) code. These estimates are based on place of work. In contrast, the US Bureau of Labor Statistics’ employment data are based on place of residence. This potential difference is not great if one defines the subject area based on labor markets.1 However, the main concern with using employment data as a measure of endogenous growth is the effect of non-local economic influences. One of the most useful and intellectually satisfying contributions to the literature on the practice of endogenous growth research is the methodology proposed by Stimson et al. (2002) using the regional share component of a shift-share equation to represent endogenous growth – essentially the dependent variable in any number of regression models accounting for regional development. More recently, Stimson et al. (2006) have adapted an enhanced shift-share model suggested by Haynes and Dinc (1997). Starting with Rigby and Anderson’s efforts to include output and productivity change to account for ambiguity in the source of employment shifts

STIMSON PAGINATION (M2469).indd 287

20/12/2010 15:12

288

Endogenous regional development

in the regional component, Haynes and Dinc (1997) offer an approach for separating the influences of labor and capital in the productivity change portion of the regional component. The intention of using the regional shift component is to remove any portion of employment change that is not attributable to local characteristics. In most cases, this approach is likely to work very well. However, there is a possibility that the regional shift component may not sufficiently partial out the influence of development incentives that are provided by non-local government entities. Specifically, the problem is caused when local economic development is advanced through one-off (non-recurring) grants or other incentives funded through non-local sources such as state government. Since the incentives are non-recurring, does the incentive represent an endogenous or exogenous characteristic?

THE INFLUENCE OF INCENTIVES IN LOCAL ECONOMIC GROWTH For practitioners and applied researchers, economic development is based on three sources, often called the legs of the economic development stool: 1.

2.

3.

Attraction refers to bringing existing businesses and organizations to the local market. This includes relocated plant sites or corporate headquarters and possibly large-scale government enterprises such as a military base or prison. Retention strategies focus on keeping whatever firms and activities that are located in the local community from relocating elsewhere or, in the current economic climate, helping companies stay in business. Business development is the creation of new local firms or the expansion of existing local businesses.

Business retention and development are clearly based on local factors of production. An argument can be made that business attraction can be exogenous growth. The business was created and built based on factors that may not have had anything to do with local factors of production.2 Site location decisions are typically made based on local characteristics. The Area Development annual survey of corporate executives ranks site location factors based on level of perceived importance. As shown in Table 14.1, traditional endogenous factors such as convenient infrastructure access, labor costs and availability, tax rates and proximity to markets are all important in site location decisions (Gambale, 2008). Governmentfunded incentives and tax exemptions make the top ten in important local

STIMSON PAGINATION (M2469).indd 288

20/12/2010 15:12

Economic development incentives and measurement of growth

Table 14.1

289

Site selection factors according to corporate executives

Rank Factor

2007

2006

1 2 3 4 5 6 7 8 9 10 10t 11 12 13 14 15 16 17 18 19 20 21 22 23 24

96.9 92.3 89.0 88.7 88.2 85.4 83.8 83.4 83.2 82.8 82.8 82.2 80.6 79.3 72.1 71.8 71.5 65.2 63.0 62.5 56.6 54.4 38.1 32.7 15.2

90.9 95.0 82.4 85.1 85.5 73.3 90.8 88.6 68.9 86.7 76.9 n/a 78.4 n/a 67.1 49.3 n/a 65.3 64.1 64.1 56.0 61.4 20.8 30.0 17.0

Highway accessibility Labor costs Energy availability and costs Availability of skilled labor Occupancy or construction costs Available land Corporate tax rate State and local incentives Environmental regulations Tax exemptions Proximity to major markets Availability of advanced ICT services Low union profile Availability of buildings Right-to-work state Proximity to suppliers Expedited or fast-track permitting Availability of unskilled labor Availability of long-term financing Raw materials availability Training programs Accessibility to major airport Railroad service Proximity to technical university Waterway or ocean port accessibility

Source: Compiled by the author using information available from Area Development (www.areadevelopment.com).

characteristics for companies ahead of buildings, availability of financing, training programs and other factors. The findings are similar for Area Development’s survey of site location consultants with incentives and tax exemptions ranking number 6 and 7 in importance, respectively (ibid.). Compared with most other developed nations, even those that share a history in English common law, the US federal system pushes more governmental functions and taxing authority to state and local governments. With the advent of first wave (Blakely and Bradshaw, 2002) industrial recruitment strategies that included direct incentives, states and the locales began competing for firm relocations. Some have even termed this competition the ‘economic war among the states’ (Burnstein and Rolnick, 1995).

STIMSON PAGINATION (M2469).indd 289

20/12/2010 15:12

290

Endogenous regional development

One reason that states and local government have increased their use of incentives has been effective convergence in state and local tax burdens driven by interstate competition. According to the Tax Foundation, the average combined state and local tax burden in the US is 10.10 percent of per capita income (CNN/Money, 2008). Thirty-five of the 50 states have a combined average state and local tax burden within ± 1.0 percent (ibid.). Based on an ongoing survey conducted by the Council for Community and Economic Research, 25 of the 50 US states have non-targeted3 grants as a part of their incentive programs (C2ER, 2008). Bartik (2005) concludes that lowering taxes has a statistically significant but small effect on business activity levels. This effect is larger for suburban communities in intra-regional competition for employers. Peters and Fisher (2004) conducted a meta-review (their terminology) of the literature, reporting that incentives were found in most studies to have ambiguous or very minor impacts. Gabe and Kraybill (2002) found a negative relationship between employment change and state incentives in Ohio. There are, of course, many other studies that could be cited, but the results are consistently unclear. There is no consistent evidence that incentives lead to growth, especially employment growth. Why then would there be an issue with the measurement of endogenous growth? Simply stated, the magnitude of incentives is tracking upwards and in many cases has become large enough that dispassionate practitioners strongly feel that they play a role in site location decisions. Admittedly, this observation is based largely on casual empiricism, but one can find compelling evidence to support the notion that incentives can trump other business location factors. Alabama The state of Alabama has taken a very aggressive approach to attracting auto assembly plants. The first to arrive was Mercedes Benz in 1993. Mercedes Benz artfully played North Carolina, South Carolina and Alabama’s state negotiating teams off against each other to drive up the value of the incentives being offered (Buchholz, 2009). In the end, Mercedes Benz, arguably one of the premier auto manufacturers in the world, chose Vance, Alabama, population 480, as the site for a new assembly plant. Vance, Alabama, is located between the cities of Birmingham (45km) and Tuscaloosa (20km). In exchange, the company received incentives, including direct payments for workers’ salaries in the early years of the agreement, that were valued at US$300 million, or somewhere between US$153 000 and US$220 000 per job created. Demonstrating further that these incentives were beyond the capacity of local government, the state of

STIMSON PAGINATION (M2469).indd 290

20/12/2010 15:12

Economic development incentives and measurement of growth

291

Alabama eventually had to borrow money from its own state employees’ pension fund to pay for the promised incentives (ibid.). Cautioned to a small extent by the near financial disaster of the Mercedes Benz deal, the state of Alabama and local governments offered a comparatively paltry US$158 million incentive package to Honda in 1999 or about $105 000 per job created (Anonymous, 1999). The state’s largess grew once more in the early part of this decade with a state and local incentive package for Hyundai totaling about US$253 million, or about US$117 000 per promised job (Beyerle, 2002). Texas In Texas, the state legislature set up funding for the Texas Enterprise Fund to be used by the governor’s Office of Economic Development as a ‘deal closer’. These funds are effectively grants to attract or retain businesses, or to promote expansion in a targeted industry. In Fiscal Years 2004–07 (the state of Texas operates on a biennium budget), the governor’s office granted over US$360 million to 39 companies, of which US$324.1 million has actually been distributed. Tables 14.3 and 14.4 provide information on the grants. All grants are tied to job creation and capital investment criteria. So far about US$300 000 has been recovered through clawbacks that are included in the grant contracts for failure to create the promised number of jobs. Included in the Texas Enterprise Fund recipients is Ruiz Foods, a maker of frozen Mexican food. Ruiz foods relocated from California to the city of Denison located in Grayson County, Texas. This will be the subject of the case study/simulation discussed later in the chapter. In exchange for locating in Denison and bringing 423 jobs to the local economy, Ruiz Foods received an Enterprise Fund (EF) grant of US$1.5 million in June of 2005. The EF grant is in addition to any local incentives such as tax abatements, grants, permit fee waivers and the like. There have been nine EF grants of at least US$10 million with two at the US$50 million level (Texas Instruments and Sematech).

THE PROBLEM FOR ENDOGENOUS GROWTH MEASURES As noted earlier, using gross regional employment data as the dependent variable in an endogenous growth model would potentially bias the resulting regression coefficients due to overall economic and industry-specific trends. The regional component of a shift-share analysis is intended to

STIMSON PAGINATION (M2469).indd 291

20/12/2010 15:12

292

Endogenous regional development

capture the local competitive effect of the sum of all local characteristics. These local effects are further refined by the Haynes and Dinc (1997) modeling approach. Still, even with these model improvements over traditional shift-share adjustments, employment change is still supposed to be related to local characteristics in the regional component. However, the shiftshare methodology has no recognized method for adjusting the regional component for exogenous incentives. The presence of a state-level incentive is, in and of itself, not a problem when assessing local economic development characteristics, as long as the incentive is available on a recurring basis. Otherwise, the assessment of any endogenous growth model would be unreliable because the assumed competitive advantage would not be present in another time period. Therefore, if the time period covered by the model coincides with a local firm receiving a large, one-off subsidy, the assessment of regional competitive strengths is potentially rendered invalid. This is especially true if the incentivized firm represents a substantial change in area employment, such as the aforementioned Mercedes Benz assembly plant, which could bias the coefficients of the growth model at unknown magnitudes. To illustrate this point I will use the traditional shift-share calculation for food manufacturing firms (NAICS 311) to assess the economy of the ShermanDenison Metropolitan Area (MA) of Texas. The Sherman-Denison Example Sherman-Denison MA is composed of Grayson County. In recent years two major employers in Sector 311, Pillsbury and Oscar Mayer, closed plant operations in Grayson County. This resulted in substantial employment declines in Grayson County while this industry was comparatively stable at the national level, potentially suggesting that this area’s endogenous characteristics were somehow not competitive (see Table 14.2). As noted earlier, Ruiz Foods moved their food processing plant to Denison, Texas in June 20054 occupying the building abandoned by Oscar Mayer (a meat processing plant) and employing 423 individuals. The availability of a building contributes to Grayson County’s positive endogenous characteristics. On the other hand, we have been informed by local economic development professionals that very few of the former employees of either Pillsbury or Oscar Mayer are currently employed by Ruiz negating the potential advantage of an available, specialized labor force. For illustrative purposes, the simulation focuses on the period subsequent to the last national recession. From 2003 through March of 2005, employment in Sector 311 continued to decline in Grayson County but rose after Ruiz Foods relocated to the area. As shown in Table 14.3, the

STIMSON PAGINATION (M2469).indd 292

20/12/2010 15:12

Economic development incentives and measurement of growth

Table 14.2

293

Employment and establishments in Grayson County and the United States Sector 311 food manufacturing and totals

Year

Grayson County 311 Employment

Total Grayson County Employment

US 311 Employment

US Total Employment

1998 1999 2000 2001 2002 2003 2004 2005 2006

1 445 914 749 728 387 139 125 111 404

38 179 38 147 39 519 39 252 36 693 37 337 36 644 37 036 37 459

1 464 419 1 464 354 1 468 254 1 470 146 1 443 766 1 495 998 1 482 261 1 469 730 1 458 738

108 117 731 110 705 661 114 064 976 115 061 184 115 061 184 113 398 043 115 074 924 116 317 003 119 917 165

Source: Compiled by the author using data available from the US Department of Commerce, Census Bureau and County Business Patterns.

national share component of the industry-specific shift share totaled eight jobs. The industrial share component, reflecting declining employment at the national level, is −11.5. That leaves an employment gain of 268.5 for the regional share, indicating that the entire gain in employment is attributable to Ruiz Foods. It appears that sector employment would have declined further from 2005 levels if Ruiz Foods had not located to the area, but local economic development practitioners suggest that the relocation would not have happened without the state incentive funds. It is reasonable to assume that other site location factors influenced Ruiz’s decision; however, the growth in the regional share for this sector in Grayson County is, at least in part, not attributable to endogenous growth factors. In broader assessments of endogenous growth, employment change is measured across all employment sectors of the local economy simultaneously, often at a two-digit data disaggregation (‘major industry’ sectors). Is the potential bias observed when examining a single industry a problem for assessments that look more broadly at local employment change? To examine this possibility, separate shift-share calculations are made for 19 major industry sectors present in Grayson County. As shown in Table 14.4, Grayson County shows substantial competitive weakness as indicated by the sum of the regional shares for the major industry categories at negative 1266 jobs on a base of 37 334 jobs in 2003. The strongest performing sectors were lodging and food services (349.0

STIMSON PAGINATION (M2469).indd 293

20/12/2010 15:12

294

Table 14.3

Endogenous regional development

Shift-share calculations for Sector 311 food manufacturing Grayson County, Texas 2003–06

Component National shift Industry shift Regional shift Total shift

% Change 119 917 165–113 398 043 1 458 738–1 495 998 404–139

0.0575 −0.0249 1.9065

Share 139* 0.0575

8.0

139* (−0.0249–0.0575) −11.5 139* (1.9065–0.0249)

268.5 265.0

Source: The author.

regional share), manufacturing (324.7), and retail trade (152.2). If you remove the effect of the Ruiz Foods relocation, traditional basic industries shrank substantially in Grayson County. The strongest performing sectors were in low-value-added services. Perhaps this indicates that Grayson County really requires exogenous intervention to promote basic industry development. It is noteworthy that without Ruiz Foods, the regional share for the manufacturing major industry category would have likely been negative. For Grayson County, the introduction of an incentivized firm distorts the regional component of the shift-share analysis at the detailed industry level (311) and the major industry group level of aggregation (31). It could also impact an assessment of the regional strength of basic industries when using a shift-share approach. However, even though the removal of the jobs associated with Ruiz foods would decrease the sum of the regional share components by an additional 30 percent, the presence of Ruiz Foods does not greatly alter the overall assessment that the Grayson County economy was not competitive during the 2003 to 2006 period. Nonetheless, for this case study the impact of adding a firm to the region through exogenous incentives could impact a statistical analysis of the relative impacts of endogenous growth characteristics on local economic performance. Addressing the Potential Problem The degree to which the potential problem of biased coefficients in an endogenous growth model needs to be specifically addressed is first a function of the sample used for the model. If the sample is large and the number of subject areas having a firm that was influenced in their location

STIMSON PAGINATION (M2469).indd 294

20/12/2010 15:12

Economic development incentives and measurement of growth

Table 14.4

295

Shift-share calculations Grayson County, Texas 2003–06

Industry 11: Agriculture 21: Mining 22: Utilities 23: Construction 31: Manufacturing 42: Wholesale Trade 44: Retail Trade 48: Transportation 51: Information 52: Finance 53: Real Estate 54: Professional Services 55: Management 56: Administrative Support 61: Education 62: Health Care 71: Arts & Entertainment 72: Lodging & Food Services 81: Other Services Totals

2003 2006 National Employment Employment Share

Industry Regional Share Share

15 80 198 2 322 6 855 995 5 861 693 486 2 465 418 1 107

10 85 225 2 347 6 937 1 085 6 368 764 486 2 322 412 877

0.9 4.6 11.4 133.5 394.1 57.2 336.9 39.8 27.9 141.7 24.0 63.6

−2.1 13.0 −29.4 214.9 −636.8 −28.9 17.9 0.8 −55.4 −71.8 11.1 44.0

−3.8 −12.6 45.0 −323.4 324.7 61.7 152.2 30.4 27.5 −212.9 −41.2 −337.7

54 2 073

60 1 720

3.1 119.2

−2.4 244.3

5.3 −716.5

435 7 935 377

455 7 892 308

25.0 456.2 21.7

6.8 46.0 7.3

−11.8 −545.2 −97.9

3 459

4 120

198.9

113.1

349.0

1 506 37 334

1 573 38 046

86.6 2 146.3

−60.9 −168.5

41.4 −1 265.8

Source: Compiled by the author using data available from US Department of Commerce, Census Bureau, and County Business Patterns.

decision by exogenous incentives is small, then the potential bias is likely inconsequential. For example, Stough et al. (2007) examined 245 metropolitan statistical areas in their endogenous growth model. In contrast, Stimson et al. (2008) studied local government areas by state in Australia with as few as 47 subjects in their model of the Victoria economy. It is beyond the scope of this analysis to perform sensitivity analyses for the described effect, but one could imagine that a few instances of exogenous incentives would have little measurable influence on the models, but the likelihood of such influence is greater with a small sample. Unfortunately, large samples do not necessarily shield studies of US sub-state economic areas from the effect described in this chapter. As

STIMSON PAGINATION (M2469).indd 295

20/12/2010 15:12

296

Endogenous regional development

noted earlier, the use of state-level, one-off incentives is hardly a rare occurrence. Therefore, one would need to at least suspect that many of the metropolitan areas studied by Stough et al. (2007) are home to firms influenced by exogenous incentives. Still, if there is reason to conclude that the number of industry share components impacted by the presence of firms receiving exogenous incentives is small, a note offering reader caution is probably sufficient.5 If regression modeling operated perfectly, the influence of any confounding factor would be partialed to the error term and left as unexplained variance. One would have to test for any correlations between the error term and the independent variables to make this assumption. Of course, cases of very large incentives targeted to a small economic area, such as Mercedes Benz in Alabama, are likely to appear as outliers in the model and would either be dropped from the model or have a statistical treatment to mute the influence of the case on model coefficients. Another argument for the ‘do nothing’ approach is to recognize that over the course of several years, the effect of the exogenous incentive would diminish. For example, recipients of funds from the Texas Enterprise Fund may guarantee that they will sustain their promised employment level for a certain number of years. But, having the firm choose to stay in the area, or in the best case where the incentivized firm grows, most likely reflects usual endogenous economic conditions in the local area. Adjustments to the dependent variable (employment) are problematic. While there are methods and techniques that could be applied to assessing the relative impact of a particular incentive on site location decisions, the incidence of strategic behavior on the part of business leaders makes the finding suspect. The author is not aware of any research that has gained acceptance among scholars and practitioners on the assessment of the contributions of incentives by type and magnitude on site location decisions that are scientifically rigorous. A more practical approach to dealing with the issue of partialing out the effects of exogenous incentives from endogenous growth models on the left-hand side of the equation is to choose a variable that measures growth that is not greatly sensitive to job counts. Per capita measures of income or perhaps local tax revenues could be tested, although this would address local/regional economic development, not just economic growth – an important difference. Still, it would be interesting to test existing endogenous growth models using per capita income or per capita government revenue to see how the explanatory variables relate to these alternative measures of economic performance. Theoretically, the issue raised here could be addressed in the right-hand side of the regression equation by including a variable(s) describing the

STIMSON PAGINATION (M2469).indd 296

20/12/2010 15:12

Economic development incentives and measurement of growth

297

competitive effect of the exogenous incentive. However, in practice this would be very difficult because the researcher would need to know what other offers the incentivized firm received before making their site location decision, which is impossible in practice. More importantly, the type of incentives discussed here are often highly political in nature. Their application may not follow any economically rational model. Another right-side-of-the-equation approach would be to consider variables that account for why a particular firm is chosen to receive nonrecurring exogenous incentives. As noted above, there are often political considerations in granting firms large state-level grants and other incentives. These political considerations could reside with the firm itself – call it political connectedness or influence. However, the political dimension to selecting incentive recipients could be based on the political influence of the host community. Stimson et al. (2005) introduced leadership and institutional factors to endogenous growth modeling. Though their variables do not specifically address political leadership, this provides a basis for the inclusion of variables that may explain additional variance in the endogenous growth model. Simply put, the ability of an economic area to attract directly, or through the firm, an exogenous incentive is effectively an endogenous local trait. A variable testing of this trait could be developed as an index such as a local area government equivalent of the Power Advantage Index (Mielcova and Cemerkova, 2008) or the Hirschmann-Herfindahl index of market share adapted for political representation.

CONCLUSIONS The influence of government incentives on site location decisions remains problematic for researchers and economic development practitioners, especially in the US. This chapter has sought to draw attention to the particular problem of potential bias caused by large, usually one-off exogenous incentives on local endogenous growth models when the dependent variable is job change from the regional share component of a shift-share calculation. The case study of a food manufacturing firm relocating to Grayson County, Texas, after receiving a large grant from the Texas Enterprise Fund illustrates the concern. It was shown that if one assumes that the incentive affects the firm’s location decision, then the regional share component is heavily biased if the analysis focuses on a single detailed industry. This case also shows the bias would exist to a lesser extent when the share component represents an aggregated major industry sector (manufacturing) and suggests bias if the model considered

STIMSON PAGINATION (M2469).indd 297

20/12/2010 15:12

298

Endogenous regional development

only traditional ‘basic’ sectors. However, the case study showed that the influence was greatly muted when major industry regional shares are summed for the subject area. Several alternatives are offered for addressing the potential problem including doing nothing, choosing an alternative dependent variable, or addressing the concern by including appropriate explanatory variables in the model to partial out the influence of exogenous incentives. It is suggested that testing alternative dependent variables and new explanatory variables may contribute to endogenous growth modeling exercises beyond addressing the particular concern highlighted in this chapter. The choice among the offered alternatives, or preferably the development of better alternatives, is left to the discretion of the research teams. However, even if a ‘do nothing’ alternative is chosen, researchers should be able to identify within the cases examined in their models instances of very large exogenous incentives occurring during their study period with appropriate cautions for the potential of biased coefficients. There is one additional perspective that could prove important. The argument presented in this chapter focuses on the potential for assigning a better assessment of economically competitive endogenous characteristics than is deserved by the subject area. However, another area is equally affected by the influence of exogenous incentives but in the opposite direction. The economic area that loses employment because a firm is lured away by large incentives may indeed have a competitive disadvantage but is it an endogenous weakness? In the US, where the frequency and magnitude of economic development incentives increasingly distort rational economic (site location) decision-making, one has to wonder if endogenous models at a local/regional level can really capture economic reality.

NOTES 1. Metropolitan Statistical Areas in the US are aggregations of counties based, in part, on worker commuting patterns. 2. This assumes that the local market is not a prime market area for the firm’s products and is not a source of scarce raw materials. 3. Non-targeted meaning that they are not specifically designated for a particular use, such as training, or to a particular area, such as enterprise zones. 4. Annual employment data presented in Table 14.2 is for the week of 12 March for the specified year; therefore Grayson County’s Sector 311 totals for 2005 do not reflect the presence of Ruiz Foods. 5. This would be similar to the caution offered by Stimson et al. (2008) on the potentially biasing impact of their data reflecting employee place of residence rather than place of work. It is a potential weakness in their analysis, but one for which there is little that can reasonably be done to correct.

STIMSON PAGINATION (M2469).indd 298

20/12/2010 15:12

Economic development incentives and measurement of growth

299

REFERENCES Acs, Z. and Armington, C. (2004), ‘Employment Growth and Entrepreneurial Activity in Cities’, Discussion Papers on Entrepreneurship, Growth, and Public Policy, No. 1304, Max Planck Institute for Research into Economic Systems, Jena, Germany. Anonymous (1999), ‘Incentive Deal of the Month’, The Site Selection Online Insider, available at: http://www.siteselection.com/ssinsider/incentive/ti9906. htm; accessed 28 May 2010. Bartik, T. (2005), ‘Solving the Problems of Economic Development Incentives’, Growth and Change, 36 (2), 139–66. Beyerle, D. (5 April, 2002), ‘State Unveils Hyundai Incentives’, Times Daily (Florence, Alabama), 1B, 2B, available at http://news.google.com/newspapers ?Nid=1842&dat=20020405&id=DWweAAAAIBAJ&sjid=VskEAAAAIBAJ& pg=5017,478953; accessed 10 June 2010. Blakely, E. and Bradshaw, T. (2002), Planning Local Economic Development: Theory and Practice, 3rd edition, Sage, California. Buchholz, D. (2009), ‘Business Incentives Reform. Case Study: Mercedes Benz: Deal of the Century’, CFED, available at http://www.cfed.org/assets/pdfs/ business-incentives-reform.pdf; accessed 10 June 2010. Burnstein, M. and Rolnick, A. (1995), ‘Congress Should Stop the Economic War Among the States’, The Region, 9, 3–20. C2ER (2008), State Development Incentive Database, see www.c2er.org for access. CNN/Money (2008), ‘How Tax-friendly is Your State?’, available at http://money. com/pf/ features/lists/taxesbystate2005/index.html; accessed 28 May 2010. Gabe, T. and Kraybill, D. (2002), ‘The Effect of State Economic Development Incentives on Employment Growth of Establishments’, Journal of Regional Science, 42(4), 703–30. Gambale, G. (ed.) (2008), The 22nd Annual Corporate Survey, available at: www. areadevelopment.com; accessed 28 May 2010. Gkritza, K., Sinha, K., Labi, S. and Mannering, F. (2007), ‘Influence of Highway Construction Projects on Economic Development. An Empirical Assessment’, Annals of Reginal Science, 42(3), 545–63. Haynes, K. and Dinc, M. (1997), ‘Productivity Change in Manufacturing Regions: A Multi-factor/Shift-Share Approach’, Growth and Change, 28(2), 201–21. Markusen, A. (1985), Profit Cycles, Oligopoly, and Regional Development, MIT Press, Cambridge, MA. McCombie, J. (1982), ‘How Important is the Spatial Diffusion of Innovations in Explaining Regional Growth Rate Disparities?’, Urban Studies, 19(4), 377–82. Mielcova, E. and Cemerkova, S. (2008), ‘Political Power of the Czech Representatives in the European Parliament’, Munich Personal RePEc Archive, paper no. 12258, available at: http://mpra.ub.uni-muenchen.de/12258/; accessed 28 May 2010. Peters, A. and Fisher, P. (2004), ‘The Failures of Economic Development Incentives’, Journal of the American Planning Association, 70(1), 27–38. Romer, P. (1986), ‘Increasing Returns and Long Run Growth’, Journal of Political Economy, 94(5), 1002–37. Stimson, R., Robson, A. and Shyy, T. (2006), ‘Modeling Regional Endogenous

STIMSON PAGINATION (M2469).indd 299

20/12/2010 15:12

300

Endogenous regional development

Growth: An Application to the Non-Metropolitan Regions of Australia’, Annals of Regional Science, online edition, 16 April, 1–20. Stimson, R., Robson, A. and Shyy, T. (2008), ‘Modeling Determinants of Spatial Variation in Regional Endogenous Growth: Non-metropolitan Regions in the Mainland States of Australia’, paper presented at the Annual Conference of the Australia-New Zealand Regional Science Association, Adelaide, December. Stimson, R., Stough, R. and Roberts, B. (2002), Regional Economic Development: Analysis and Planning Strategy, Springer, Berlin. Stimson, R., Stough, R. and Salazar, M. (2005), ‘Leadership and Institutional Factors in Endogenous Regional Economic Development’, Investigaciones Regionales, 7, 23–52. Stough, R., Song, C., Qian, H. and Wang, J. (2007), ‘Modeling Endogenous Growth in US Metropolitan Regions’, paper presented to the Western Regional Science Association Annual Meeting, Newport Beach, February.

STIMSON PAGINATION (M2469).indd 300

20/12/2010 15:12

15.

Regional growth and development theories revisited Roberta Capello and Peter Nijkamp

TRENDS IN REGIONAL ECONOMIC GROWTH THEORY Regional development covers a wide range of economic policy issues related to the need to exploit appropriate productive resources that may contribute, or form an impediment, to the welfare of a region, in either an absolute or a relative sense. Consequently, regional development is associated with both efficiency objectives, such as the optimal use of scarce factor inputs, and equity objectives (such as social cohesion and distribution of wealth), issues in modern jargon sometimes referred to as ‘territorial cohesion’. Clearly such elements are also prominent in conventional macroeconomic growth analysis (e.g., at a national scale), but a special reason for giving explicit attention to regional economic growth lies in the relatively small and open character of a region (or a system of regions). Since a region forms an intermediate or hybrid spatial unit between a nation and its citizens, regional economic growth theory uses elements from both macroeconomic growth policy and individual welfare theory. Against this background it comes as no surprise that in recent years endogenous growth theory has gained much popularity in regional economics; it is essentially a blend of microeconomics and macroeconomic growth theory, in which smart use of the indigenous resources of a region plays a critical role. It covers, inter alia, the linkages between income, employment, investments, infrastructure and suprastructure. In particular, much emphasis is placed on the study of spatial socio-economic disparities or convergence (including labour migration), with a particular view to the way spatial disparities can be influenced by the deliberate actions of stakeholders (e.g., industry, government). It is thus increasingly recognized that regional development is not only a spatial efficiency issue in economic policy. It is also an equity issue, because economic development normally exhibits a significant degree of spatial variability, and, furthermore, it is increasingly conceived of as a 301

STIMSON PAGINATION (M2469).indd 301

20/12/2010 15:12

302

Endogenous regional development

spatial sustainability issue with strong regional and urban dimensions. Over the past few decades, the continuous concern about unbalanced regional development has prompted various strands of important research literature, in particular: the measurement of inter-regional disparity; the causal explanation for the emergence or persistent presence of spatial variability in economic development; and the impact assessment of policy measures aimed at coping with undesirable spatial inequity conditions. The study of socio-economic processes and inequalities at the meso-scale and regional levels positions regions as the core places of policy action, and hence warrants intensive conceptual and applied research efforts. Economic analysis of regional growth and its distribution already has a long history and dates back to classical economists such as Adam Smith and Alfred Marshall. From an analytical perspective, the foundations of modern economic growth theory can be found in the early work of Solow (1956), in which he argues that, in a neoclassical economic world, the growth rate of a region (measured in per capita income) is inversely related to its initial per capita income, a thesis that offers an optimistic perspective for poor regions. Interesting regional growth models have been extensively developed in the 1960s, in particular in a neoclassical framework (Borts, 1960; Borts and Stein, 1964 and 1968). The spatial-economic convergence idea has attracted considerable attention over the years and has generated interesting applied research on evolving convergence versus persistent disparities (see, e.g., Barro and Sala-i-Martin, 1992). In addition to statistical analysis, this research and policy issue has also led to new insights into the contextual drivers of disparities, such as mobility, product diversification, monopolistic competition, institutional impediments, and so on. It is thus clear that, over the past few decades, a persistent unequal distribution of welfare among regions and/or cities has been a source of concern and inspiration for both policy-makers and researchers. Regional development is at the heart of this concern, as it is about the geography of welfare and its evolution. It has played a central role in such disciplines as economic geography, regional economics, regional science and economic growth theory. The concept is not static in nature, but refers to complex space–time dynamics of regions (or an interdependent set of regions). In addition, the actual measurement is also dependent on the geographical scale used. Changing regional welfare positions are often hard to measure, especially in a comparative multi-regional context. In practice, we often use Gross Domestic Product (GDP) per capita (or growth therein) as a statistical approximation (see Stimson et al., 2006). Sometimes alternative or complementary measures are also used, such as per capita consumption, poverty rates, unemployment rates, labour force participation rates, or access to public services. These indicators are more social in nature

STIMSON PAGINATION (M2469).indd 302

20/12/2010 15:12

Regional growth and development theories revisited

303

and are often used in United Nations welfare comparisons. An example of a rather popular index in this framework is the Human Development Index, which represents the welfare position of regions or nations on a 0–1 scale using quantifiable standardized social data (such as employment, life expectancy, or adult literacy) (see, e.g., Cameron, 2005). In all cases, however, spatial socio-economic disparity indicators show much variability. The history of economic research has witnessed an ongoing debate on income convergence among countries or regions, both theoretically and empirically, often with due emphasis on effective and efficient policy measures and strategies. This continues to be an important research topic, since regional disparities may have significant negative socio-economic cost consequences because of, for instance, social welfare transfers, inefficient production systems (e.g., due to the inefficient allocation of resources), and undesirable social conditions (see Gilles, 1998). In a neoclassical framework of analysis, these disparities (e.g., in terms of per capita income) are assumed to vanish in the long run, because of the spatial mobility of production factors, which ultimately results in an equalization of factor productivity in all regions. Clearly, long-range factors – such as education, R&D and technology – play a critical structural role in this context. In the short run, however, regional disparities may show rather persistent trends (see also Patuelli, 2007). Inter-regional disparities can be – as mentioned above – measured using various relevant categories, such as (un)employment, income, investment, growth, and so on. Clearly, such indicators are not entirely independent, as, for instance, is illustrated in Okun’s Law, which assumes a relationship between economic output and unemployment (see Okun, 1970; Paldam, 1987). The empirical research on convergence has often been based on cross-sectional analysis, for example, on the basis of concepts related to beta-convergence and sigma-convergence (see, e.g., Baumol, 1986; Barro, 1991). More recently, time-series analysis has also been used extensively, based on notions from stationary time processes (see, e.g., Bernard and Durlauf, 1995). The findings from these different strands of the literature are not always identical, however, and in recent years this has stimulated new research efforts inspired by endogenous growth theory. The convergence of regional disparities is clearly a complex phenomenon, as it refers to a variety of mechanisms through which differences in welfare between regions may vanish (see Armstrong, 1995). In the modern convergence debate, we observe increasingly more attention to the openness of spatial systems, reflected, inter alia, in trade, labour mobility, commuting and so on. (see, e.g., Magrini, 2004). In a comparative static sense, convergence may have various meanings in a discussion on a possible reduction

STIMSON PAGINATION (M2469).indd 303

20/12/2010 15:12

304

Endogenous regional development

in spatial disparities among regions (see also Baumol, 1986; Barro and Sala-i-Martin, 1992; Bernard and Durlauf, 1996; Boldrin and Canova, 2001). In particular, there is: b-convergence: a negative relationship between per capita income growth and the level of per capita income in the initial period (e.g., poor regions grow faster than initially rich regions); s-convergence: a decline in the dispersion of per capita income between regions over time. The convergence hypothesis in neoclassical economics has been widely accepted and discussed in the literature, but is critically dependent on two hypotheses (see Cheshire and Carbonaro, 1995; Dewhurst and MutisGaitan, 1995): (1) diminishing returns to scale in capital should prevail, which means that output growth will be less than proportional with respect to capital; (2) technological progress will generate benefits that also decrease with its accumulation (i.e., diminishing returns). A wealth of studies has been carried out to estimate the degree of b-convergence and s-convergence (see, e.g., Barro and Sala-i-Martin 1991, 1992). The general findings are that the rate of b-convergence is in the order of magnitude of 2 per cent annually, while the degree of s-convergence tends to decline over time, for both US states and European regions. Clearly there is still an ongoing debate worldwide on the type of socio-economic convergence, its speed, its multidimensional conceptualization and its causal significance in the context of regional policy measures (see, e.g., Fagerberg and Verspagen, 1996; Galor, 1996; Fingleton, 1999). Important research topics in the current literature appear to be: the role of knowledge and entrepreneurship; spatial heterogeneity in locational or socio-cultural conditions; and institutional and physical barriers. An important new topic in the field is group convergence (or club convergence), which means a convergence of sets of regions towards a more homogeneous cluster.1 We may conclude that the research field of spatial disparities is still developing and is prompting fascinating policy issues that deserve indepth attention in the years ahead. Besides the concern on policy issues, also in the academic arena too much interest has arisen over the last decade in spatial development. The degree of scientific consensus – as well as the degree of cross-fertilization of ideas – between regional economists and mainstream economists has led to interesting debates in regional science circles. In addition, in a period of globalization (including financial crises) and of the creation of broad single-currency areas, regions (and also nations) have to keep an eye on the competitiveness of their production systems, because no spontaneous or automatic adjustment mechanism is yet at work to counterbalance a lack (or an insufficient growth rate)

STIMSON PAGINATION (M2469).indd 304

20/12/2010 15:12

Regional growth and development theories revisited

305

of productivity. Local specificities and local material and non-material assets (including knowledge and learning mechanisms) become strategic elements upon which the competitiveness of regions is based. More than ever, theories of regional growth and development need to be able to interpret the way in which regions achieve a role in the international division of labour and, more important, the way in which regions can maintain this role over time. Finally, it should be recognized that regional growth trajectories are often confronted with sustainability issues and scarcity of environmental resources, such as environmental quality, natural resource security, urban quality of life, or spatial biodiversity. Such important issues are increasingly a critical element of a regional welfare function. Some sustainability elements may be interpreted in terms of scarce input factors, such as bioethanol, solar, or wind energy, while others are seen as welfare constituents, such as the quality of the living environment. This new scarcity has important spatial components and has to be analysed from the perspective of regional development as well. Consequently, regional development is not only a matter of regional competition for the use of scarce resources but also a matter of creating the environmental conditions for sustainable wealth creation and enjoyment. This calls increasingly for a broader welfare interpretation of regional growth, in which the triangle of efficiency, equity and sustainability form the driving forces of regional development. Our aim in this chapter is to present a selection of recent contributions that explain regional growth and local development, with the aim to highlight: ● ● ●

recent advances in theories; policy implications of these theories; cross-fertilization of ideas among regional economists and mainstream economists.

In the sections that follow we first discuss the theoretical advances on regional growth achieved in recent years in different parts of the world. Next we address sustainability issues from a regional perspective and then we provide an overview of future challenges in this field. That is followed with some concluding remarks.

DEVELOPMENTS IN REGIONAL GROWTH THEORY As mentioned above, the complex issue of regional development and growth has been a focal point of interest in recent decades. The great

STIMSON PAGINATION (M2469).indd 305

20/12/2010 15:12

306

Table 15.1

Endogenous regional development

Main tendencies in theories of regional economics

New Theoretical Tendencies

Regional Growth Theories

Regional Development Theories

Realism in theoretical approaches

Endogenous growth determinants A role in growth models for the complex non-linear and interactive behaviour and processes that take place in space Imperfect market conditions in growth models Growth as a long-term competitiveness issue Technological progress as an endogenous factor of growth Evolutionary trajectories of non-linear interdependencies of complex systems Non-linear, dynamic growth models

Intangible assets as sources of regional competitiveness Determinants of success and failure of clusters of SMEs, local districts, milieux An active role of space in knowledge creation

Dynamic approaches

Dynamic rather than static agglomeration economies Dynamics of local districts and local milieux

Source: Elaboration on Capello (2008).

number of relatively new and advanced contributions in the area of regional development/growth theories does not allow us to offer a detailed review of all individual achievements made; moreover, a detailed presentation of all new ideas would probably not be very stimulating. Our view is that selective and focused attempts to highlight general theoretical trends offers a more fruitful basis for a debate on current weaknesses and on possible future directions of regional economics. Inevitably, the set of ‘megatrends’ in regional growth analysis offered in this section is both selective and incomplete (Capello, 2008). Table 15.1 summarizes the two main ‘mega-trends’ that, in our view, largely characterize the theoretical developments over the last two decades in regional economics, and that are common to urban economics and to many other disciplines (Capello and Nijkamp, 2004):2 the need for more realism and the move towards dynamic rather than static approaches. These theoretical perspectives are presented in Table 15.1 for both regional growth and regional development theories, the former aiming at explaining the aggregate growth rate of income and employment in a

STIMSON PAGINATION (M2469).indd 306

20/12/2010 15:12

Regional growth and development theories revisited

307

formalized and quantitative way, the latter oriented towards the identification of all tangible and intangible qualitative elements of the growth process of regions. The Need for More Realism The first tendency that has accompanied the theoretical development in the field is the need for more realism in sometimes rather abstract conceptual approaches, by relaxing most of the glaringly unrealistic assumptions of the basic theoretical models. This tendency is justified by the need to broaden the interpretative capacity of the theoretical toolbox in this research field by searching for theories that are better able to reflect issues and policy strategies for the real world. In recent years, more realism has been required to insert into growth models the complex non-linear and interactive behaviour and processes that take place in space, and to understand regional competitiveness in terms of endogenous factors. The question of whether a region is intrinsically capable of growing as a result of endogenous forces has been a source of debate for decades. Industrial specialization, infrastructure endowment, central location, production factor endowment, or agglomeration economies have all been emphasized in the academic arena as driving forces of local economic success. The theory of industrial organization has also provided important contributions here. One important step forward in this field has been the focus on economies of scale in production, which, together with non-linear transportation costs, have been introduced into a (quantitative) inter-regional growth model; the final spatial distribution of activities critically depends on initial conditions, including the starting distribution of activities and the nature of the non-linearities embedded in the activity–transportation interactions, which give rise to multiple equilibria (Krugman, 1991). The additional value of this approach – sometimes referred to as the ‘new economic geography’ – rests in skilfully modelling the interaction between transportation costs and economies of scale in production, although the determinants of endogenous growth have already long been emphasized, starting with the Myrdal-Kaldor model (increasing returns, cumulative self-reinforcing growth patterns). The aim to incorporate agglomeration economies – in the form of increasing returns – into conventional models of a strictly macroeconomic nature was made possible by advances in more sophisticated mathematical tools for the analysis of the qualitative behaviour of dynamic non-linear systems (bifurcation, catastrophe and chaos theory), together with the advent of formalized economic models that abandoned the hypotheses of constant returns and perfect

STIMSON PAGINATION (M2469).indd 307

20/12/2010 15:12

308

Endogenous regional development

competition (Fujita and Thisse, 1996, 2002). Further links to complexity theory are also plausible in this context. In the spirit of Krugman’s efforts, in the field of endogenous determinants great emphasis has recently been put on knowledge as a driving force for development, and, what is really new, on the endogenous selfreinforcing mechanisms of knowledge creation. Macroeconomic models of endogenous growth, where knowledge is generally embedded in human capital (Romer, 1986; Lucas, 1988), have widely dominated the academic arena in the last decade. Their main aim was to insert more realism into growth models by relaxing the unrealistic assumption that technological progress is an exogenous process in an economic system. In the new growth theories, instead technological progress is regarded as an endogenous response of economic actors in a competitive environment. More specifically, increasing returns in factor productivity stemming from endogenous factors – such as innovation, scale economies and learning processes – are included in a neoclassical production function, where they offset the effect of the marginal productivity of the individual factors, which the traditional neoclassical approach assumes to be decreasing. These adjusted assumptions have led not only to interesting analytical findings but also to new far-reaching policy implications. Endogenous growth theory is already two decades old and has played a central role in the growth debate since the 1990s. The main idea of these new contributions is that technological progress is not exogenously given, but a self-organized response of individual agents in an entrepreneurial business environment. Consequently, in contrast to earlier macroeconomic explanatory frameworks, the emphasis is much more on firms’ individual economic behaviour (see, e.g., Barro and Sala-i-Martin, 1997; Aghion and Howitt, 1998). In this way, it can be demonstrated that regional growth is not the result of exogenous productivity-enhancing factors but rather is the result of deliberate choices of individual actors (firms and policymakers). Individual smart behaviour is thus at the centre of this debate. The strategic relevance of knowledge for innovation and entrepreneurship is thus increasingly recognized. The spatial distribution of knowledge and its spillovers are considered as important success factors for regional development in an open competitive economic system. Thus, the geographical patterns of knowledge diffusion, as well as the barriers to access to knowledge, are decisive for regional development in a modern global and open space economy. Consequently, knowledge policy – often instigated by ICT advances – is a critical success factor for regional welfare creation (see, e.g., Keeble and Wilkinson, 1999; Acs et al., 2002; Döring and Schnellenback, 2006), where knowledge refers to education, learning, training, creativeness and R&D.

STIMSON PAGINATION (M2469).indd 308

20/12/2010 15:12

Regional growth and development theories revisited

309

It is noteworthy that the identification of endogenous determinants of growth was the crucial scientific issue that explained the birth of regional development theories. Development is in fact by definition endogenous. It is fundamentally dependent on the concentrated organization of the territory, embedded in which is a socio-economic and cultural system whose components determine the success of the local economy: ● ● ●

entrepreneurial ability; local production factors (labour and capital); the relational skills of local actors.

All these generate cumulative knowledge acquisition and, moreover, a decision-making capacity that enables local economic and social actors to guide the development process, support it when it is undergoing change and innovation and enrich it with the external information and knowledge required to harness it to the general process of growth, and to the social, technological and cultural transformation of the world economy. The micro-behavioural nature of these approaches allowed a deep understanding of the sources of territorial externalities, and of increasing returns in the form of agglomeration economies, which are at the basis of industrial cluster formation. Within this approach, much emphasis is given to the role of entrepreneurship in regional development (Nijkamp and Stough, 2004). More realism in the study of clusters and their determinants has called for a better understanding of the success and failures of local productive systems, hardly explained in the first theories proposed. Dynamic agglomeration economies – defined as territorial advantages that act on the capacity of firms and regions to innovate – have become the centre of most recent theoretical reflections in this field, giving rise to neo-Schumpeterian approaches in regional development. A major debate dominates the academic arena, with the aim of identifying the role of space in innovative processes. This has prompted a new strand of growth literature on regions or cities. In the vast regional and urban literature created in this field,3 the endogenous determinants of innovation are increasing returns in the form of dynamic locational advantages deriving from the following: 1.

The spatial or geographical proximity of firms, which facilitates the exchange of tacit knowledge. This characterizes reflection by economic geographers concerned to explain the concentration of innovative activities. 2. The relational proximity of firms, defined as interaction and cooperativeness among local agents, the source of collective learning

STIMSON PAGINATION (M2469).indd 309

20/12/2010 15:12

310

3.

Endogenous regional development

processes and socialization to the risk of innovation (i.e., territorialized relations among subjects operating in geographical and social proximity). This was the approach taken by territorial economists in explaining the dynamics of local systems in terms of local innovative capacity. Institutional proximity, which takes the form of rules, codes and norms of behaviour that facilitate cooperation among actors and therefore the socialization of knowledge and assist economic actors (individual people, firms and local institutions) to develop organizational forms that support interactive learning processes. This aspect was emphasized by more systemic approaches that sought to understand the evolution of complex systems like the innovative system in a broad global competitive network.

Dynamic Approaches A second clear mega-trend in theoretical developments – typical only of regional development/growth theories – has been the attempts to move towards dynamic approaches. Time matters, as well as space, in regional science, and this also holds in regional economics. The effort to encapsulate time in spatial analyses has been accomplished in two different ways, according to two different meanings of time applied in the two fields of analysis: a more traditional chronological time; and time as the rhythm of innovative phenomena that occur in the territory that has been applied in regional growth models all over the world. The introduction of chronological time within spatial analysis is not at all a simple task, since it requires a mathematical and methodological toolbox, only recently available to regional scientists. Theories on nonlinear regional dynamics – framed in the context of chaos theory, synergetics theory, or predator–prey analysis – may be mentioned here (see Nijkamp and Reggiani, 1999). In growth models, until a few years ago, the vast majority of experiments and applications took for granted the existence of linear – and thus regular – growth processes. Linear models are certainly able to generate unstable solutions, but the solutions of such models are restricted to certain regular standard types. Such models may provide approximate replications of short-run and medium-run changes, but fail to encapsulate long-term developments characterized by structural shifts of an irregular nature. This limitation has recently been overcome with the adoption of non-linear models, which allow for a change in the dynamics of a system generated even by small perturbations in structural forms; structural instability means the possible existence of significant qualitative changes in the behaviour of the system (i.e., in the state variables) that are

STIMSON PAGINATION (M2469).indd 310

20/12/2010 15:12

Regional growth and development theories revisited

311

closely connected with bifurcation and catastrophe phenomena, which occur if the parameter values (i.e., the control variables) are changing (see Fujita and Thisse, 1996, 2002). The application of non-linear models to the well-known neoclassical and Keynesian models has shown that the deterministic and unique results achieved by the dynamic linear models are no longer guaranteed: inter-regional income convergence determined by the traditional neoclassical model collapses and opens the way to alternative possible trajectories, and equilibria solutions; non-linear Keynesian Myrdal-Kaldor models substitute the deterministic result of continuous growth or decline with new and opposite development trajectories, after catastrophe phenomena occur (Miyao, 1984, 1987a, 1987b). We thus observe a wealth of new theoretical contributions on the dynamics of spatial development. Such a theoretical improvement has also been useful in achieving a greater realism of these models, which are now able to incorporate the dynamic interactions between the components of a spatial system represented in a network constellation. The latter network approaches are functionally determined by interdependencies between the behaviour of actors and distance frictions. Such spatial interactions may be stable in nature (i.e., operating under fixed external conditions) or subject to change as a result of dissipative evolutionary processes in the external world. In the latter case, model parameters become time-dependent, so that non-linear complex dynamics may emerge (see Puu, 1991; Nijkamp and Reggiani, 1993 and 1999; Nijkamp, 2006). Next, we also note that, in the field of regional development, conceptually speaking a different concept of time has been developed and applied. Time à la Bergson-Heidegger is interpreted as duration and a continuous process of creation, characterized by discontinuity, irreversibility, sequentiality and cumulativity. Time has thus been conceived by an important part of regional studies as the pace of learning, innovation and creation processes. Local clusters (and industrial districts) are, by definition, the loci where learning and cumulative learning processes take place. The identification of the sources and the endogenous determinants of such processes, apart from simple physical proximity, represent a great challenge for regional economists. Knowledge spillovers, collective learning, learning regions (or learning space) and knowledge-based regions are all theories that embrace the most advanced perspectives in this direction. In these theoretical approaches, therefore, innovation has become the critical survival factor in a competitive space economy and determines the direction and pace of regional development (Nijkamp and Abreu, forthcoming). We may thus conclude that the introduction of micro-based endogenous growth components has led to an enrichment of regional growth theories,

STIMSON PAGINATION (M2469).indd 311

20/12/2010 15:12

312

Endogenous regional development

in which the role of public policy is no longer seen as a top-down ‘control and command’ approach, but as a partnership model, in which strategies are to be developed in cooperation with all stakeholders in space.

THE CHALLENGE OF REGIONAL SUSTAINABILITY Concern for the local environment is not a recent policy issue. Already in Ancient Rome we find examples of policy measures to mitigate the noise nuisance of horse-drawn carriages during the night. And several medieval cities in Europe had guidelines on where and when to park carriages (including priority rules for wealthy citizens). In the welfare economics literature, such non-market phenomena were called ‘externalities’. To guarantee again an equalization of marginal costs and benefits, a system of taxes (or subsidies) was foreseen, so that an optimal market equilibrium could be re-established. These Pigouvian welfare rules were seen as interesting exceptions to an otherwise perfectly working market mechanism. However, in the age of mass industrialization and large-scale mobility, such externalities became a dominant phenomenon rather than an exception. Already in the 1950s and 1960s, the first voices on environmental decay were heard, often inspired by concerns about declining water quality, noise nuisance, local health and air pollution. The real breakthrough and awareness took place in the first wave of environmental consciousness, viz. after the publication of the First Report to the Club of Rome in 1973. In that global study, much concern was expressed about environmental pollution, scarcity of natural resources, the threat of the ‘population bomb’, lack of food and the position of the developing countries. The next wave of increased interest emerged in 1987, with the publication of the Brundtland Report. This document was greatly concerned about the issue of sustainable development, in particular from the viewpoint of the developing world and from the viewpoint of the next generations. Since then, ‘sustainability’ has become a fashionable word but, unfortunately, it lacked an operational definition, so that in practice this concept was uncritically used for sectoral development, for regional or local developments, or for development with regard to new generations. However, this concept has prompted an avalanche of research – including applied modelling research – on various elements of local, urban or regional sustainability, from the perspective that our world calls for a holistic view on future development. And, finally, we witness the third wave, where the interest in global developments – such as climate change or the rise in sea level – is anchored in local or regional settings. This embeddedness of the global change debate in local and regional development has prompted new types of research in

STIMSON PAGINATION (M2469).indd 312

20/12/2010 15:12

Regional growth and development theories revisited

313

the area of regional science (see Batabyal and Nijkamp, 2008). From the wide range of issues, we now select five important focal points that serve to illustrate the importance of the recognition of sustainability issues in regional development. Regional Economic Development Regional economic development comprises more elements than regional economic growth: for example, access to social facilities, a healthy living environment, a high quality of education, a sustainable living environment, and so on. Such conditions manifest themselves in many facets of the local or regional environment. We mention here three examples: 1.

2.

3.

First, business location is nowadays not a simple cost-minimization question, but a multidimensional trade-off of many factors, in which the (image of the) quality of the local environment or access to recreational or sports facilities plays a critical role. Second, public environmental policies may increase the locational costs of new or incumbent firms, and hence influence their locational decisions. Thus, the way externalities are treated by the public sector will impact on the willingness to invest by business firms. Third, the local institutional setting – for example, the presence of public–private partnerships – may create a sense of common responsibility, also for the local quality of life, so that the institutional structure may be a driving force for local or regional sustainable development.

The welfare of regions and the use of the physical resource base of these regions are clearly mutually interwoven phenomena. It is thus clear that regional economic development and sustainability strategies may be seen as mutually complementary forces that may reinforce each other. Natural Resources Natural resources may have a productive meaning (e.g., mines, oil reserves) or a consumption meaning (e.g., recreation parks or lakes for sailing or fishing). They may be decisive for balanced regional development, as will now be exemplified by means of three selective illustrations on natural resource use: 1.

First, deforestation may be a wise strategy from a short-term productive perspective, but calls for a careful and balanced policy in view of

STIMSON PAGINATION (M2469).indd 313

20/12/2010 15:12

314

2.

3.

Endogenous regional development

long-term sustainability. An equilibrium policy that attains a balance between long-term wealth and short-term revenues may therefore be a wise strategy. A second example concerns water provision and use. The situation regarding the quality of water in many regions and cities is rather critical nowadays, and requires intensified policy initiatives. Proper management of scarce water reserves – in terms of both quality and quantity – is a condition sine qua non for healthy and sustainable local and regional development. A final example concerns waste management. Waste is a necessary by-product of any production and consumption process; it is often a cost factor, but it may be turned into a revenue factor (e.g., in the case of district heating, waste incineration, and so on.). Thus, the challenge will be to turn ‘environmental bads’ into ‘environmental goods’ through the proper usage of sticks and carrots. And this calls for operational environmental research at a local scale.

Smart local resource use may improve local welfare conditions. We may conclude, therefore, that proper waste management at local or regional level will create favourable conditions that stimulate balanced local or regional development. Environmental Regulation Local or regional economic policy is not only a matter of taxes, subsidies or prohibitions but also calls for transparent environmental regulations. We will give a few examples here. 1.

2.

3.

Environmental regulations have often been criticized because they might cost jobs. But this calls for a careful empirical analysis, as there are also many counterfactual findings based on several experiences all over the world. There is an increasing body of literature that demonstrates that strict environmental regulations by no means have a negative effect on economic activity, but most likely a negligible or even positive effect in the long run. Pigouvian taxation is seen as a proper response to the existence of externalities. This may certainly be true from an economic perspective, but the way such taxes are then allocated is decisive for their political support and acceptability. This is also of great importance for local taxation schemes (e.g., for parking, waste treatment, and so forth.). Finally, the local or regional dimension of environmental policy is of great importance. This has a particular relevance for the local tax base

STIMSON PAGINATION (M2469).indd 314

20/12/2010 15:12

Regional growth and development theories revisited

315

and the distribution of public money in a system of fiscal federalism. Such spatial demarcation issues may greatly impact on local sustainable development. Environmental regulations may act as competitiveness vehicles in regional development. All in all, we may conclude that spatial heterogeneity in environmental regulation systems will significantly influence the economic and environmental aspects of any sustainable development policy. Regional Climate Change The relationship between global change and local development has often been neglected, but is receiving increased attention. Global warming will certainly have a far-reaching impact on local or regional development conditions, as illustrated by the following examples. Changes in our ecosystems – with more extremes and outliers – will call for adjusted ecosystems management at local or regional level so that these systems can be better protected against floods or other nature disasters. This may call for other types of land use policy. Another example concerns the interrelationship between local and global climatic conditions (e.g., in terms of building regulations, or water or energy supply). This prompts increasingly proactive policies (e.g., in the EU). Many cities are to be found along the coast or river flood plains. The rise in sea level calls for new design and security principles in urban planning in the decades to come. We may clearly draw the conclusion that sustainability policy will call for drastic adjustments of local and regional land use, environmental and resource policy. Modelling Local and Regional Environments Much quantitative research has been undertaken in the past few decades to map out the complex policy dynamics of the modern space economy. Computable General Equilibrium modelling and simulation modelling have proven to be appropriate tools. More recently, we have seen the emergence of geographic information systems (GIS) as an example of operational model applications. A great many useful applications can be found in the literature. For instance: (1) Several studies have been published on housing planning and demand, with respect not only to the quality of the dwellings in a narrow sense, but also to the neighbourhood and accessibility effects in a broader sense (Kain and Quigley, 1970). (2) Another example concerns location analysis in the context of land use and land values.

STIMSON PAGINATION (M2469).indd 315

20/12/2010 15:12

316

Endogenous regional development

GIS has been very instrumental in dealing with large spatial data sets, for example, in evaluating commercial buildings and densities, in making spatial assessment of land values, or in identifying risk-prone areas (Geoghegan et al., 1997; Weng, 2002). GIS has indeed become an important toolbox for regional science research, but its potential for local or regional socio-economic development policy is still underutilized. There is still a long way to go before geo-science modelling will be an integral part of local or regional development policy. The interwoven nature of the ecology of our planet – at different geographic-scale levels – with local and regional development thus prompts many new research endeavours. There is yet one final question: ‘Has sustainability had an impact on regional development policy, and is there a link to endogenous growth policy?’ The answer is in the affirmative. Environmental and resource conditions at local or regional scales are no longer permanently given factors. They can be influenced and adjusted by smart and appropriate decisions and investments in both the public and the private sector. In other words, regional sustainable development from an endogenous perspective is the result of deliberate choices and actions of local and regional stakeholders, including public policy-making bodies.

NEW RESEARCH CHALLENGES IN REGIONAL GROWTH AND DEVELOPMENT The previous observations have clearly demonstrated that, nowadays, fascinating new theoretical challenges are being faced by regional scientists, and have to be addressed. A first challenge emerged from the attempt to obtain advantage from a future convergence in different theoretical approaches, a convergence only partially obtained by the new regional growth theories. New growth theories make a commendable effort to include space in strictly economic models. Also to be commended is the implicit merging in their theoretical structure of the various conceptions of space put forward over the years: that is, the merging of the physical-metric space represented by transport costs with the diversified space, which assumes the existence of certain territorial polarities where growth cumulates. However, the new economic geography is still unable to combine the economic laws and mechanisms that explain growth with territorial factors that spring from the intrinsic relationality present at local level. Any approach that did achieve this would represent the maximum cross-fertilization between location theory, development theory and macroeconomic growth theory – a synthesis that would bring out the territorial micro-foundations of macroeconomic growth models.

STIMSON PAGINATION (M2469).indd 316

20/12/2010 15:12

Regional growth and development theories revisited

317

What is needed, therefore, is a convincing ‘umbrella model’ that comprises the micro-territorial, micro-behavioural and intangible elements of the development process. What is required for this purpose is the definition of patterns, indicators and analytical solutions to be incorporated into formalized models necessarily more abstract and synthetic in terms of their explanatory variables, variables besides the cost of transport, which remove the role of territory in the development process. A move in this direction is the quantitative sociology that embraces the paradigm of methodological individualism and seeks to ‘measure’ the social capital of local communities. It is obviously necessary to bring out territorial specificities within a macroeconomic model. Or, in other words, it is necessary to demonstrate the territorial micro-foundations of macroeconomic growth models. Serious risks both from disciplinary barriers and from adherence to interdisciplinary perspectives on strategic problems can be mentioned here. They are the result not only of regional scientists’ narrow perspective, as mentioned by Bailly and Coffey (1994), but also of some rather idiosyncratic attitudes of mainstream disciplines towards a clearly multidisciplinary science like regional science. In this respect, the following examples may be illuminating. The first concerns the theory on social capital developed by quantitative sociology: the concept could take advantage of, and provide advantage to, all reflections on local synergies and milieu effects developed by regional and urban economists, and by the strategic planning studies in the field of urban planning. The reflections in the field of knowledge spillovers developed by industrial economists could take advantage of regional science concepts of collective learning and relational proximity, in which the endogenous spatial development patterns of knowledge are not left to simple probabilistic contacts, but are explained through territorial processes (Camagni and Capello, 2002). Last but not least, the theoretical reflections that characterize the ‘new economic geography’ seem to be the result of a skilful effort of a group of mainstream economists, driven, however, by a somewhat inexplicable attitude that denies the importance of well-known spatial concepts (i.e., technological spatial externalities), or (re-)invents important spatial concepts (i.e., cumulative self-reinforcing processes of growth; transportation costs vs. agglomeration economies in location choices). The inevitable consequence of such an attitude is to mix the important and undeniable steps forward made by the ‘new economic geography’ school with already well-known knowledge in the field of regional science. Especially in the case of economics, we hope that, after the (re-) discovered interest by mainstream economists in space, and in spatial phenomena, the attitude towards regional science will change in favour of a more cooperative and pronounced interest.

STIMSON PAGINATION (M2469).indd 317

20/12/2010 15:12

318

Endogenous regional development

Related to the interdisciplinary challenge, a last important point is worth mentioning. An interdisciplinary approach should lead scientists to explore new frontiers and achieve new interpretative analytical frameworks. However, the tendency shown in this respect is quite different, being merely inclined to exploit passively the new ideas suggested by complementary disciplines. A case worth mentioning is the enthusiastic way in which regional scientists accepted the spatial spillover theory as a theory that could provide a new interpretation of the role of space as a channel for knowledge transfer. Instead, a more critical approach to this theory shows that to some extent it has gone some steps backwards in the interpretation of space in spatial knowledge creation. Space is purely geographical, a physical distance among actors, a pure physical container of spillover effects that come about – according to the epidemiological logic adopted – simply as a result of physical contact among actors. Important consequences ensue from this interpretation of space. First, this view is unable to explain the processes by which knowledge spreads at local level, given that it only envisages the probability of contact among potential innovators as the source of spatial diffusion. Second, it concerns itself only with the diffusion of innovation, not with the processes of knowledge creation. It thus imposes the same limitations as did Hägerstrand’s (1967) pioneering model in regard to the spatial diffusion of innovation: the diffusion of knowledge means adoption, and adoption means more innovation and better performance. This ignores, however, the most crucial aspect of the innovation process: how people (or the context) actually learn. This calls for a more thorough and innovative investigation of cognitive processes in a regional context (Capello, 2009). This is an aspect of overriding interest not only for researchers but also, and especially, for policy-makers should they wish to explore the possibilities of normative action to promote local development.

RETROSPECT AND PROSPECT In the globalization process of creating an open economy, local factors and local specificities are fundamental elements on which the competitiveness of countries depends. They are, therefore, important areas where practitioners and policy-makers require a sophisticated and advanced toolbox to intervene. Regional economics has been subject to varied and creative advances in theoretical economic contributions. Various core tendencies in the development trajectories of the discipline have been stressed in this

STIMSON PAGINATION (M2469).indd 318

20/12/2010 15:12

Regional growth and development theories revisited

319

review, in particular the attempt to introduce more realism into the theoretical approaches, combining rigorous theoretical reflections with an understanding of the reality of place. The force field of regional welfare is varied and sometimes unpredictable, as the region is part of a complex global system. An important contributor to regional development is technological progress, an extensively studied topic in the recent economic growth literature. From a geographic (regional, urban, or local) perspective, in recent years much attention has been paid to the spatial conditions that induce technological progress (e.g., entrepreneurial climate, availability of venture capital, incubator facilities, institutional transformations, etc.) (see also Longhi and Musolesi, 2007). Furthermore, the spatial diffusion of technology has also received considerable attention, in particular in the geography literature. The development of regions in an open world is also strongly influenced by their environmental and resource base. And, therefore, sustainability is a critical task for regional policy as well. This concept is, however, not only a burden but opens the door for new and creative regional growth strategies. In fact, in the last few years, spatial development has been put vigorously on the agenda of policy-makers who foresee economic competitiveness as highly dependent on an efficient territorial system of regions and cities. At the European level, the concept of territorial sustainability has come to the fore, meaning the normative aim of complementing economic equity aims with social, environmental and territorial ones. It is noteworthy that regional scientists increasingly address, in their scientific agenda policy, issues that have a strong societal interest; convergence problems, on the one hand, and endogenous determinants of regional growth (like knowledge creation), on the other. These two themes both have a practical interest and reflect a need for a multidisciplinary approach, providing regional scientists with all the challenges to identify new pathways. Whether this will actually happen is a matter of willingness to grasp the opportunities that are provided in the present era, and to respond to the plea of policy-makers for a more locally-oriented understanding of the real world. The history of regional growth theory covers almost half a century. We have seen a deepening of our understanding of the complex backgrounds of, and impediments to, balanced regional development. But regions are not islands that can be isolated from the rest of the world. They are subjected to the same forces as nations and continents, and, consequently they have to be positioned in a similar conceptual explanatory framework, complemented with regional socio-economic specificities. Regional

STIMSON PAGINATION (M2469).indd 319

20/12/2010 15:12

320

Endogenous regional development

growth is thus a race without a finish; and that also applies to regional growth theory.

NOTES 1. For further discussion on club convergence, we refer to Baumont et al. (2003); Chatterji (1992); Chatterji and Dewhurst (1996); Fischer and Stirbock (2006); Islam (2003); López-Bazo et al. (1999); Quah (1996); Rey and Montouri (1999); Sala-i-Martin (1996). 2. For an extensive review of regional growth theories, see, among others, Johansson et al. (2001). 3. For the literature on spatial spillovers, see, amongst others, Anselin et al. (1997 and 2000); Audretsch and Feldman (1996); Aydalot (1986); Feldman (1994); Feldman and Audretsch (1999); de Groot et al. (2001); Jaffe (1989); Jaffe et al. (1993); Maier and Sedlacek (2005); on collective learning, see Bellet et al. (1993); Camagni (1991); Capello (1999) and (2001); Crevoisier and Camagni (2000); Maillat et al. (1993); Rallet (1993); Rallet and Torre (1995); Ratti et al. (1997); on learning regions, see Lundvall (1992); Lundvall and Johnson (1994); Maskell and Malmberg (1999); on knowledge-based regions, see Florida (1995); Malecki (2000); Nijkamp and Stough (2004); Simmie (1997).

REFERENCES Acs, Z.J., Anselin, L. and Varga, A. (2002), ‘Patents and Innovation Counts as Measures of Regional Production of New Knowledge’, Research Policy, 31(7), 1069–85. Aghion, P. and Howitt, P. (1998), Endogenous Growth Theory, MIT Press, Cambridge, MA. Anselin, L., Varga, A. and Acs, Z. (1997), ‘Local Geographic Spillovers between University Research and High Technology Innovations’, Journal of Urban Economics, 42(3), 422–48. Anselin, L., Varga, A. and Acs, Z. (2000), ‘Geographic and Sectoral Characteristics of Academic Knowledge Externalities’, Papers in Regional Science, 79(4), 435–43. Armstrong, H.W. (1995), ‘Convergence among the Regions of the European Union, 1950–1990’, Papers in Regional Science, 74(2), 143–52. Audretsch, D. and Feldman, M. (1996), ‘R&D Spillovers and the Geography of Innovation and Production’, American Economic Review, 86(3), 630–40. Aydalot, Ph. (ed.) (1986), Milieux Innovateurs en Europe, GREMI, Paris. Bailly, A. and Coffey, W. (1994), ‘Regional Science in Crisis: A Plea for a More Open and Relevant Approach’, Papers in Regional Science, 73(1), 3–14. Barro, R. (1991), ‘Economic Growth in a Cross-section of Countries’, Quarterly Journal of Economics, 106(6), 407–43. Barro, R.J. and Sala-i-Martin, X. (1991), ‘Convergence across States and Regions’, Brookings Papers on Economic Activity, 1, 107–82. Barro, R.J. and Sala-i-Martin, X. (1992), ‘Convergence’, Journal of Political Economics, 100 (2), 223–51. Barro, R. and Sala-i-Martin, X. (1997), ‘Technology Diffusion, Convergence and Growth’, Journal of Economic Growth, 2(1), 1–27. Batabyal, A.A. and Nijkamp, P. (2008), ‘Sustainable Development and Regional

STIMSON PAGINATION (M2469).indd 320

20/12/2010 15:12

Regional growth and development theories revisited

321

Growth’, in R. Capello and P. Nijkamp (eds), Handbook of Regional Growth and Development Theories, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Baumol, W.J. (1986), ‘Productivity Growth, Convergence, and Welfare: What the Long-run Data Show’, American Economic Review, 76(5), 1072–85. Baumont, C., Ertur, C. and LeGallo, J. (2003), ‘Spatial Convergence Clubs and the European Regional Growth Process, 1980–1995’, in B. Fingleton (ed.), European Regional Growth, Springer-Verlag, Berlin. Bellet, M., Colletis, G. and Lung, Y. (1993), ‘Introduction au Numéro Spécial sur Economie et Proximité’, Revue d’Economie Régionale et Urbaine, 3, 357–61. Bernard, A. and Durlauf, S. (1995), ‘Convergence in International Output’, Journal of Applied Econometrics, 10(2), 97–108. Bernard, A. and Durlauf, S. (1996), ‘Interpreting Tests of the Convergence Hypothesis’, Journal of Economics, 71(1–2), 161–74. Boldrin, M. and Canova, F. (2001), ‘Europe’s Regions: Income Disparities and Regional Policy’, Economic Policy, 16, 207–53. Borts, G.H. (1960), ‘The Equalisation of Returns and Regional Economic Growth’, The American Economic Review, 50, 319–47; reprinted in D. McKee, R. Dean and W. Leahy (eds) (1970), Regional Economics: Theory and Practice, The Free Press, New York, pp. 147–76. Borts, G.H. and Stein, J.L. (1964), Economic Growth in a Free Market, Columbia University Press, New York. Borts, G.H. and Stein, J.L. (1968), ‘Regional Growth and Maturity in the United States: A Study of Regional Structural Change’, in L. Needleman (ed.), Regional Analysis, Penguin Books Ltd., Harmondsworth, pp. 159–97, original edition (1962), in Schweizerische Zeitschrift für Volkswirtscahft und Statistik, 98, 290–321. Camagni, R. (1991), ‘Local Milieu, Uncertainty and Innovation Networks: Towards a New Dynamic Theory of Economic Space’, in R. Camagni (ed.),  Innovation Networks: Spatial Perspectives, Belhaven-Pinter, London, pp. 121–44. Camagni, R. and Capello, R. (2002), ‘Apprendimento Collettivo, Innovazione e Contesto Locale’, in R. Camagni and R. Capello (eds), Apprendimento Collettivo e Competitività Territoriale, Franco Angeli, Milano, pp. 11–26. Cameron, R. (2005), ‘Spatial Economic Analysis’, Journal of Development Perspectives, 1(1), 146–63. Capello, R. (1999), ‘Spatial Transfer of Knowledge in High-technology Milieux: Learning vs. Collective Learning Processes’, Regional Studies, 33(4), 353–65. Capello, R. (2001), ‘Urban Innovation and Collective Learning: Theory and Evidence from Five Metropolitan Cities in Europe’, in M.M. Fischer and J. Froehlich (eds), Knowledge, Complexity and Innovation Systems, Springer, Berlin, Heidelberg, New York, pp. 181–208. Capello, R. (2008), ‘Regional Economics in its Fifties: Recent Theoretical Directions and Future Challenges’, Annals of Regional Science, 42(1), 747–67. Capello, R. (2009), ‘Spatial Spillovers and Regional Growth: A Cognitive Approach’, European Planning Studies, 17(5), 639–58. Capello, R. and Nijkamp, P. (2004), ‘The Theoretical and Methodological Toolbox of Urban Economics: From and Towards Where?’, in R. Capello and P. Nijkamp (eds), Urban Dynamics and Growth: Advances in Urban Economics, Elsevier, Amsterdam, pp. 1–30.

STIMSON PAGINATION (M2469).indd 321

20/12/2010 15:12

322

Endogenous regional development

Chatterji, M. (1992), ‘Convergence Clubs and Endogenous Growth’, Oxford Review of Economic Policy, 8(4), 57–69. Chatterji, M. and Dewhurst, J.H.L. (1996), ‘Convergence Clubs and Relative Economic Performance in Great Britain: 1977–1991’, Regional Studies, 30(1), 31–40. Cheshire, P. and Carbonaro, G. (1995), ‘Convergence-Divergence in Regional Growth Rates: An Empty Black Box?’, in H.W. Armstrong and R.W. Vickerman (eds), Convergence and Divergence among European Regions, Pion, London, pp. 89–111. Crevoisier, O. and Camagni, R. (eds) (2000), Les Milieux Urbains: Innovation, Systèmes de Production et Ancrage, EDES, Neuchâtel. de Groot, H., Nijkamp, P. and Acs, Z. (2001), ‘Knowledge Spillovers, Innovation and Regional Development’, Papers in Regional Science, 80(3), special issue. Dewhurst, J.H.L. and Mutis-Gaitan, H. (1995), ‘Varying Speeds of Regional GDP Per Capita Convergence in the European Union, 1981–91’, in H.W. Armstong and R.W Vickerman (eds), Convergence and Divergence among European Regions, Pion, London, pp. 22–39. Döring, Th. and Schnellenback, J. (2006), ‘What Do We Know About Geographical Knowledge Spillover and Economic Growth?’, Regional Studies, 40(3), 375–95. Fagerberg, J. and Verspagen, B. (1996), ‘Heading for Convergence? Regional Growth in Europe Reconsidered’, Journal of Common Market Studies, 34(3), 431–48. Feldman, M. (1994), The Geography of Innovation, Kluwer Academic, Boston, MA. Feldman, M. and Audretsch, D. (1999), ‘Innovation in Cities: Science-based Diversity, Specialisation and Localised Competition’, European Economic Review, 43(2), 409–29. Fingleton, B. (1999), ‘Estimates of Time to Economic Convergence: An Analysis of Regions of the European Union’, International Regional Science Review, 22(1), 5–34. Fischer, M. and Stirbock, C. (2006), ‘Pan-European Regional Income Growth and Club-convergence’, Annals of Regional Science, 40(4), 693–721. Florida, R. (1995), ‘Toward the Learning Region’, Futures, 27(5), 527–36. Fujita, M. and Thisse, J.-F. (1996), ‘Economics of Agglomeration’, Journal of the Japanese and International Economies, 10(4), 339–78. Fujita, M. and Thisse, J.-F. (2002), Economics of Agglomeration: Cities, Industrial Location and Regional Growth, Cambridge University Press, New York. Galor, O. (1996), ‘Convergence? Inferences from Theoretical Models’, Economic Journal, 106(437), 1056–69. Geoghegan J., Wainger, L.A. and Bockstael, N.E. (1997), ‘Spatial Landscape Indices in a Hedonic Framework: An Ecological Economics Analysis Using GIS’, Ecological Economics, 23(3), 251–64. Gilles, S.-P. (1998), ‘The Political Consequence of Unemployment’, Working Paper, No. 343, Department of Economics, Universitat Pompeu, Fabra. Hägerstrand T. (1967), Innovation Diffusion as a Spatial Process, University of Chicago Press, Chicago. Islam, N. (2003), ‘What Have We Learnt from the Convergence Debate?’, Journal of Economic Surveys, 17(3), 309–62. Jaffe A. (1989), ‘Real Effects of Academic Research’, American Economic Review, 79(5), 957–70. Jaffe A., Trajtenberg, M. and Henderson, R. (1993), ‘Geographic Localisation of

STIMSON PAGINATION (M2469).indd 322

20/12/2010 15:12

Regional growth and development theories revisited

323

Knowledge Spillovers as Evidenced by Patent Citation’, Quarterly Journal of Economics, 63(4), 577–98. Johansson, B., Karlsson, C. and Stough, R. (eds) (2001), Theories of Endogenous Regional Growth, Springer-Verlag, Berlin. Kain, J.F. and Quigley, J. (1970), ‘Measuring the Value of Housing Quality’, Journal of the American Statistical Association, 65(330), 532–48. Kaldor, N. (1970), ‘The Case for Regional Policies’, Scottish Journal of Political Economy, 17(3), 337–48. Keeble, D. and Wilkinson, F. (1999), ‘Collective Learning and Knowledge Development in the Evolution of Regional Clusters of High-technology SMEs in Europe’, Regional Studies, 33(4), 295–303. Krugman, P. (1991), Geography and Trade, MIT Press, Cambridge, MA. Longhi, S. and Musolesi, A. (2007), ‘European Cities in the Process of Economic Integration: Towards Structural Convergence’, Annals of Regional Science, 41(2), 335–51. López-Bazo, E., Vayá, E., Mora, A. and Suriñach, J. (1999), ‘Regional Economic Dynamics and Convergence in the European Union’, Annals of Regional Science, 33(3), 343–70. Lucas, R. (1988), ‘On the Mechanics of Economic Development’, Journal of Monetary Economics, 22(1), 3–42. Lundvall, B.A. (1992), ‘Introduction’, in B.A. Lundvall (ed.), National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning, Pinter Publishers, London, pp. 1–19. Lundvall, B.-A. and Johnson, B. (1994), ‘The Learning Economy’, Journal of Industry Studies, 1(2), 23–42. Magrini, S. (2004), ‘Regional (Di) Convergence’, in V.J. Henderson and J.-F. Thisse (eds), Handbook of Regional and Urban Economics, vol. IV, Elsevier, Amsterdam, pp. 2742–96. Maier, G. and Sedlacek, S. (eds) (2005), Spillovers and Innovations – Space, Environment and the Economy, Springer Verlag, Wien. Maillat, D., Quévit, M. and Senn, L. (eds) (1993), Réseaux d’Innovation et Milieux Innovateurs: un Pari pour le Développement Régional, EDES, Neuchâtel. Malecki, E. (2000), ‘Creating and Sustaining Competitiveness’, in J.R. Bryson, P.W. Daniels, N. Henry and J. Pollard (eds), Knowledge, Space, Economy, Routledge, London, pp. 103–19. Maskell, P. and Malmberg, A. (1999), ‘Localised Learning and Industrial Competitiveness’, Cambridge Journal of Economics, 23(2), 167–85. Miyao, T. (1984), ‘Dynamic Models of Urban Growth and Decay: A Survey and Extensions’, paper presented to the Second World Conference of Arts and Sciences, Rotterdam, 4–15 June. Miyao, T. (1987a), ‘Dynamic Urban Models’, in E. Mills (ed.), Urban Economics; Handbook of Regional and Urban Economics, vol. 2, North-Holland, Amsterdam, pp. 877–925. Miyao, T. (1987b), ‘Urban Growth and Dynamics’, in E. Mills (ed.), Urban Economics. Handbook of Regional and Urban Economics, North-Holland, Amsterdam, pp. 1–41. Myrdal, G. (1957), Economic Theory of Under-developed Regions, Gerald Duckworth & Co, London. Nijkamp, P. (2006), ‘Ceteris Paribus: Spatial Complexity and Spatial Equilibrium. An Interpretative Perspective’, mimeo, available from the author.

STIMSON PAGINATION (M2469).indd 323

20/12/2010 15:12

324

Endogenous regional development

Nijkamp, P. and Abreu, M. (forthcoming), ‘Regional Development Theory’, Encyclopedia of Geography. Nijkamp, P. and Reggiani, A. (eds) (1993), Nonlinear Evolution of Spatial Economic Systems, Springer-Verlag, Berlin. Nijkamp, P. and Reggiani, A. (1999), The Economics of Complex Spatial Systems, Elsevier, Amsterdam. Nijkamp, P. and Stough, R. (eds) (2004), Entrepreneurship and Regional Economic Development, Edward Elgar, Cheltenham, UK and Northampton, MA, USA. Okun, A.M. (1970), The Political Economic of Prosperity, Brookings Institution, Washington, DC. Paldam, M. (1987), ‘How Much Does One Percent of Growth Change the Unemployment Rate?’, European Economic Review, 31(1–2), 306–13. Patuelli, R. (2007), ‘Regional Labour Markets in Germany’, PhD Dissertation, Free University, Amsterdam. Puu, T. (1991), Non-linear Economic Dynamics, Springer-Verlag, Berlin. Quah, D.T. (1996), ‘Empirics for Economic Growth and Convergence’, European Economic Review, 40(6), 1353–75. Rallet, A. (1993), ‘Choix de Proximité et Processus d’Innovation Technologique’, Revue d’Economie Régionale et Urbaine, 3, 365–86. Rallet, A. and Torre, A. (eds) (1995), Economie Industrielle et Economie Spatiale, Economica, Paris. Ratti, R., Bramanti, A. and Gordon, R. (eds) (1997), The Dynamics of Innovative Regions, Ashgate, Aldershot. Rey, S.J. and Montouri, B.D. (1999), ‘US Regional Income Convergence: A Spatial Econometric Perspective’, Regional Studies, 33(2), 143–56. Romer, P. (1986), ‘Increasing Returns and Long-run Growth’, Journal of Political Economy, 94(5), 1002–37. Sala-i-Martin, X. (1996), ‘The Classical Approach to Convergence Analysis’, Economic Journal, 106(437), 343–70. Simmie, J. (ed.) (1997), Innovation, Networks and Learning Regions, Jessica Kingsley, London. Solow, R. (1956), ‘Contribution to the Theory of Economic Growth’, Quarterly Journal of Economics, 70(1), 65–94. Stimson, R.J., Stough, R.R. and Roberts, H.B. (2006), Regional Economic Development, Springer-Verlag, Berlin. Weng, Q. (2002), ‘Land Use Change Analysis in the Zhujiang Delta of China Using Satellite Remote Sensing, GIS and Stochastic Modeling’, Journal of Environmental Management, 64(3), 273–84.

STIMSON PAGINATION (M2469).indd 324

20/12/2010 15:12

Index absolute advantage principle 206, 209 absorption capacity see European Union structural assistance and regional disparities: absorption capacity agglomeration(s) 5, 6, 43–4, 114–15, 154–5, 179, 309 Indonesia 177–8 see also Silicon Valley agriculture 65, 213, 277, 278–9, 280 see also employment growth and decline in Australian capital city statistical divisions aménagement du territoire 217 appreciative intelligence see Silicon Valley Area Development annual survey 288–9 area-based approach 64 Arrow, K. 26–7, 28, 30 Arrow-Romer model 28 Asian BRIICS countries see foreign direct investment, knowledge assets and economic geography of growth attraction (economic development) 288 Australia 14, 49–50, 138 Australia: geographer’s perspective 12, 39–54, 57–8 analytical capabilities and policy advice 47–53 enterprising human capital and regional engagement 51–2 qualitative information and facilitated workshops 49–51 theoretically informed empiricism 48–9 case study regions and their characteristics 50–51 data set 57–8 endogenous regional theory 41–4

new regionalism and embededdness 44–7 Australia: national economy perspective 161, 162 planner’s perspective 77 vocational education and training 47, 48–9, 52–3, 54 see also employment growth and decline in Australian capital city statistical divisions; model determinants; Victoria Australian and New Zealand Regional Science Association International (ANZRSA) 269 Autocracy (human resource practice blueprint) 93 backward removal stepwise (regression) approach 136 Bangladesh: Grameen Bank 81 Basque Madrigon region 81 behavioural/management science perspective see Silicon Valley: appreciative intelligence Belgium 191, 206 Blakely, E.J. 2 bottom-up approach 64 Brazil 160 Brundtland Report 312 Bulgaria 182, 186, 187, 188, 190, 191, 192 Bureaucracy (human resource practice blueprint) 93 business development (economic development) 288 Canada: First Nations 274 capital outflow from advanced regions 208 causation, circular and cumulative 30

325

STIMSON PAGINATION (M2469).indd 325

20/12/2010 15:12

326

Endogenous regional development

centralized management 197 China see foreign direct investment, knowledge assets and economic geography of growth in Asian BRIICS countries cities, scale of see employment growth and decline in Australian capital city statistical divisions civic spirit 80 see also Victoria climate change 315 coalitions of people 53 Cobb-Douglas production function 24, 26–7 cognitive approach to district economies and synergies 214–15, 216 cohesion 185 horizontal and vertical 183 Cohesion Funds (EU) 184, 193 commercialization process 151 Commitment (human resource practice blueprint) 93–4, 97, 101 Common Agricultural Policy 65 comparative statics 22–3 competency building 52 competitiveness 185, 187–8, 272 and regional policies 217–20 see also European Union: macroeconomic and territorial policies for regional competitiveness Computable General Equilibrium modelling 315 Constant Elasticity of Substitution production function 152 Constant Share Approach 246 consumers 148–9 contemporary challenges 6–9 contracts theory 215–16 convergence 208–10 beta-convergence 31–2, 303–4 conventional gap (mean reversion) econometric model 49 economic 31–2 group/club 304 hypothesis 304 income 303 neoclassical conditional 208 nominal 183

STIMSON PAGINATION (M2469).indd 326

objective 184, 187, 193–4 real 183 sigma-convergence 303–4 spatial socio-economic 301 cooperation between areas 65 cost parameter 155 creative capital 11 creative destruction 42 creativity 9, 209, 219–20, 223, 228–9, 231, 308 critical success factors 10 cross-fertilization 30 cultural assets, potential 70–72 cultural attributes 80 cultural factors 5, 114 culture economy approach 62–3 currency devaluation 206 Cyprus 187, 190, 192 Czech Republic 187, 190, 191, 192, 196, 197, 198 Joint Operational Programme 197 decision-making under conditions of uncertainty 215 deforestation 313–14 Delbecq, A.L. 91 demand-supply interaction 207 demography 120 development: dependent 61 destructive 61 dictated 61 distorted 61 from below model 219 initiatives in small rural towns see Victoria in regional growth theory 305–12 dynamic approaches 306, 310–12 realism 306, 307–10 see also economic development differential growth, regional 221 differential/regional shift effect 117, 247–8, 249 see also model determinants: Australia and United States diminishing returns to scale 304 diversity and endogeny in explanatory paradigms 84–7 Dixon-Thirwall model 207 ‘do nothing’ approach 296, 298

20/12/2010 15:12

Index Dutch disease 200 dynamic approaches 306, 310–12 Eastern and Central Europe 182, 223, 229 ecological capital 11 economic cohesion 185 economic development 273, 313 planning, endogenous 77–9 regional 113–15, 116, 119 see also United States: economic development incentives and measurement of endogenous growth economic factors 212 economic geography see foreign direct investment, knowledge assets and economic geography of growth; new economic geography economic performance 271 Economic and Social Cohesion Policy see European Union economic success 275 economies of scale 160, 307 economist’s perspective 11–12, 20–36 comparative statics 22–3 economic convergence, empirical testing for 31–2 exogenous growth models 23–4 policy implications 32–3 precursors to modern endogenous growth theory 25–7 recent models 28–30 specific regional context 30–31 economy 120 economy-wide aggregate production function 27 education 8, 9, 29, 30–31, 303, 308 efficiency objectives 301 efficiency wages 207 embeddedness 12, 44–7 institutional 7, 9, 114 Empathic Knowledge Management 104 employment growth and decline in Australian capital city/cities statistical divisions 15, 237–66 accommodation and food services 239, 242, 244 Adelaide 264

STIMSON PAGINATION (M2469).indd 327

327 Brisbane 261 Melbourne 258, 259 Perth 263 Sydney 257 Adelaide 237, 240, 262, 264, 265 divergent rates of growth and change 241–5 employment and population 240 industrial explanations for regional shift effects 252, 254–6 shift-share analysis 249–52 administrative and support and safety 239, 242, 244, 253 Adelaide 264 Brisbane 261 Melbourne 259 Perth 262, 263 Sydney 257, 258 agriculture/forestry and fishing 239, 241, 242, 244, 253 Adelaide 264, 265 Brisbane 261 Melbourne 259, 260 Perth 262, 263 Sydney 257 arts and recreation services 243, 245, 252 Adelaide 264, 265 Brisbane 261 Melbourne 259, 260 Perth 262, 263 Sydney 257 Brisbane 237, 238, 260–61, 265 divergent rates of growth and change 241–5 employment and population 240 industrial explanations for regional shift effects 252, 254–6 shift-share analysis 249–52 Bureau of Statistics (Australian) 248–9 capital city dominance 238 Census of Population and Housing Time Series Profile (2006) 249 city shares of state population and employment 239–40

20/12/2010 15:12

328

Endogenous regional development

construction 239, 241, 242, 244, 252–3, 255–6 Adelaide 264, 265 Brisbane 261 Melbourne 259 Perth 263 Sydney 257, 258 education and training 242, 244, 253 Adelaide 262, 264 Brisbane 261 Melbourne 259 Perth 263 Sydney 257 electricity/gas/water and waste services 239, 242, 244, 252–3 Adelaide 264 Brisbane 261 Melbourne 259, 260 Perth 263 Sydney 257 financial and insurance services 239, 242, 244, 253 Adelaide 264 Brisbane 261 Melbourne 259 Perth 262, 263 Sydney 257, 258 healthcare and social assistance 241, 242, 244, 265 Adelaide 262, 264 Brisbane 260, 261 Melbourne 259 Perth 263 Sydney 257, 258 industry structure of employment 242–5 information, media and telecommunications 239, 242, 244, 258 Adelaide 264 Brisbane 260, 261 Melbourne 259 Perth 263 Sydney 257, 258 manufacturing 239, 242, 244, 252–4, 265 Adelaide 264 Brisbane 261 Melbourne 259, 260

STIMSON PAGINATION (M2469).indd 328

Perth 262, 263 Sydney 257, 258 Melbourne 237, 258–60 divergent rates of growth and change 241–5 employment and population 240 industrial explanations for regional shift effects 252–6 shift-share analysis 249–52 mining 239, 242, 244, 252–3 Adelaide 264, 265 Brisbane 261 Melbourne 259, 260 Perth 262, 263 Sydney 257, 258 national shares, changing 238–9 New South Wales 238, 239 Perth 237, 262, 263, 265 divergent rates of growth and change 241–5 employment and population 240 industrial explanations for regional shift effects 252, 254–6 shift-share analysis 249–52 professional/scientific and technical services 239, 242, 244 Adelaide 264 Brisbane 261 Melbourne 259 Perth 263 Sydney 257 public administration and safety 242, 244 Adelaide 264, 265 Brisbane 261 Melbourne 259, 260 Perth 262, 263 Sydney 257 Queensland 238, 239 regional shift effects 247, 251, 265 Adelaide 264 Brisbane 261 Melbourne 259 Perth 263 Sydney 257 retail/hiring and real estate services 239, 241, 242, 244, 265 Adelaide 264, 265 Brisbane 260, 261

20/12/2010 15:12

Index Melbourne 259, 260 Perth 263 Sydney 257, 258 share of national employment by industry 239 shift-share results of city Statistical Divisions 250 South Australia 238, 239 Sydney 237, 256–8, 265 divergent rates of growth and change 241–5 employment and population 240 industrial explanations for regional shift effects 252–6 shift-share analysis 249–52 Tasmania 238, 239 transport/postal and warehousing 239, 242, 244 Adelaide 264 Brisbane 261 Melbourne 259 Perth 263 Sydney 257 Victoria 238, 239 Western Australia 239 wholesale/trade 239, 241, 242, 244, 253, 254–5 Adelaide 262, 264, 265 Brisbane 261 Melbourne 259, 260 Perth 263 Sydney 257, 258 employment scale weighted location quotient 117 endogenous shift effect see differential/ regional shift effect Engineering (human resource practice blueprint) 93 Enns, S. 275 enterprising skills 52 entrepreneurial rents 14, 142–58 consumers 148–9 equilibrium and propositions 152–4 final goods producers 149–50 intermediate goods producers 150–51 knowledge creation/spillovers 144–7, 154–6 new economic geography models 146

STIMSON PAGINATION (M2469).indd 329

329

new growth theory 144–6 policy implications 156–7 systems of innovation framework 145–7 entrepreneurial talent 156 entrepreneurship 7, 114, 116, 119, 120, 129, 131, 137–8, 309 capital 135–6, 137 European Union 223, 227–8, 231 United States 129, 131, 137 environmental policies and regulation 313, 314–15 equity issue 301 Estonia 187, 190, 192, 196, 198 Europe 147 European Commission 184 European Community Strategic Guidelines (2007–13) 186 European Paradox 145 European Regional Development Fund 183 European Social Fund 183, 198 European Territorial Cooperation Objective 188 European Union 28, 32, 304 European Union: macroeconomic and territorial policies for regional competitiveness 15, 204–33 competitiveness and regional policies 217–20 impact of structural funds on regional growth rates 225 per-capita income disparities 213 persistence of regional disparities as rationale for policies 205–11 convergence process 208–10 peripheral regions, difficulties encountered by 206–7 spatial development policies derived from globalization debate 210–11 regional growth and differential spatial effect of macroeconomic policies 211–14 spatial effects of revaluation of exchange rates 224 see also MASST model European Union: sociological perspective 60 territorial capital 226–30

20/12/2010 15:12

330

Endogenous regional development

territorial sustainability 319 see also ’LEADER’ programme European Union Budget 186 European Union structural assistance and regional disparities: absorption capacity 15, 182–201 absorptional capacity methodology 194–6 administrative absorption capacity 184–5, 199 Belgium 191 Bulgaria 182, 186, 187, 188, 190, 191, 192 Cohesion Funds 184, 193 Cohesion Policy 183–9 Community Strategic Guidelines (2007–13) 186 Convergence Objective 187, 193–4 Cyprus 187, 190, 192 Czech Republic 187, 190, 191, 192, 196, 197, 198 disparities 189–90 Eastern and Central Europe 182 economic cohesion 185 Estonia 187, 190, 192, 196, 198 EU Budget 186 EU-8 188 EU-9 196 EU-10 186, 190 EU-12 186 EU-15 185, 186, 189, 190, 191, 193, 197 EU-25 186, 193, 194 EU-27 186, 187, 188, 189, 190, 191, 193, 212 financial absorption capacity 185 horizontal cohesion 183 human resources 195 Hungary 187, 190, 192, 196, 198 implications 193–4 Ireland 186 Latvia 187, 190, 192, 198 Lithuania 187, 190, 192, 198 macroeconomic absorption capacity 184 Malta 187, 192 n+2 rule 185 Netherlands 186 NUTS 2 regions 190, 193 phasing-in regions 194

STIMSON PAGINATION (M2469).indd 330

phasing-out regions 194 Poland 187, 190, 192, 197, 198 Regional Competitiveness and Employment Objective 187–8 regional development dynamics 190–91 Romania 182, 186, 187, 188, 190, 191, 192, 193, 195, 196 Single Market and Single Currency programmes 183 Slovakia 187, 190, 192, 196, 197, 198 Slovenia 187, 190, 192, 196, 198 social cohesion 185 Structural Funds 184, 189, 193, 194–5, 197–8, 199–200 structure 195 subsequent evaluations of absorption capacity 196–9 systems and tools 195 Territorial Cooperation Objective 188 United Kingdom 190, 191 vertical cohesion 183 within-country disparities 191–3 excellence policies 213 exogenous development: and endogenous development, comparison of 61–2 sociological perspective 59–60 exogenous factors 2, 4 exogenous forces 4, 7 exogenous growth models 23–4, 25 exogenous incentives 294, 295–7 exogenous shift effect 247–8 exploratory model 117 external funds injection 199 external networks 282 externalities 312, 313 falsification 85–6 final goods producers 149–50 First Report to the Club of Rome (1973) 312 Florida, R. 90 foreign direct investment, knowledge assets and economic geography of growth in Asian BRIICS countries 14–15, 160–79 China 160–61, 163–6, 168–73, 174–5, 179

20/12/2010 15:12

Index inter-regional inequality 169–70, 172 National Economy and Social Development Plan (2005) 169 urban–rural ratio 170–72 greenfield projects 163–4 India 160–61, 163, 164–6, 168, 173–5, 179 economic growth city-regions 174 Foreign Investment Board 173 GDP composition by industry 174 Investment Commission 173 Indonesia 160–61, 164–6, 168, 175–9 agglomeration economies 177–8 economic growth city-regions 176 trade protectionism 177 inward FDI and R&D 163–8 manufacturing 163 primary industries 163 service sector 163 forward/backward linkages 207 France 164, 206, 223, 229 proximity approach 216 Fu, X.-C. 275 G-factor 94 geographer’s perspective see Australia geographic information systems 315–16 geographical clusters 147, 155, 309, 311 see also agglomeration Germany 147, 206, 223 Gini Inequality Index 191 global labour force expansion 160 global trade 160 globalization 5–6, 7, 39, 83, 115, 160, 161 Google 95–6, 99 government industry assistance see United States: economic development incentives and measurement of endogenous growth government intervention 49 government policy 24–5 Greece 223, 227, 229 greenfield projects 163–4 GREMI 216, 219

STIMSON PAGINATION (M2469).indd 331

331

growth econometrics model 48–9 growth stage 113 growth theory see new growth theory Hamel, G. 106–7 Harrod-Domar model 23–4 Hirschmann-Herfindahl index 297 historical society 278 Honda 291 housing planning and demand 315 Hsung, R.-M. 275 human capital 5, 10–11, 29, 42, 49 Australia 51–2 planner’s perspective 80 regional shift component 114 Human Development Index 303 human resources (indicator) 195 Hungary 187, 190, 192, 196, 198 Hyundai 291 immiserizing growth phenomena 200 increasing returns to scale 207 India see foreign direct investment, knowledge assets and economic geography of growth in Asian BRIICS countries individual agency 40 Indonesia see foreign direct investment, knowledge assets and economic geography of growth in Asian BRIICS countries industrial structure 114 industry mix effect 247, 248 industry specialization 49 infrastructure: provision 278, 279, 281 support and institutional thickness 58 innovation 9, 64, 113, 165 innovative milieu 5, 114 institutional arrangements 8 institutional embeddedness 7, 9, 114 institutional factors 7, 116, 138, 212, 297 institutional proximity of firms 310 institutionalist perspective 48, 215 institutions 5, 40, 119, 120, 129, 131, 137 integrated approach 65 inter-regional level 210, 212

20/12/2010 15:12

332

Endogenous regional development

inter-regional performance see European Union structural assistance and regional disparities: absorption capacity inter-regional trade 58 inter-sectoral relationships 207 intermediate markets 150–51, 152, 153–4 internal community networks 282 intra-regional level 210 investment 160 see also foreign direct investment Ireland 186, 223 Italian school 216 Italy 206, 223, 225, 227, 229 Japan 168 Kaldor, N. 26 Kaldor regional models 31 Keynesian approaches 3, 4 Keynesian economics 21 Keynesian models 311 knowledge 5, 6, 29, 44, 143, 149, 308–9 accumulation 27 acquisition 309 Asian BRIICS 165, 167 assets see foreign direct investment, knowledge assets and economic geography of growth base 219 commercialization 146, 156–7 creation 57, 144–7, 153, 156–7 disembodied 28 production function 42 spillovers 8, 42, 144–7, 150, 154–7, 311, 317 Korea 168 labour 288 force survey employment data 247 mobility 154 productivity 207 land use: as endogenous planning base 76–7 and land values location analysis 315 Latvia 187, 190, 192, 198 ‘LEADER’ programme 60, 63–6 components 64–5 phases 63–4 significance 65–6

STIMSON PAGINATION (M2469).indd 332

leadership 5, 7, 271, 272, 282–3, 297 regional shift component 114, 116, 118–19, 120, 129, 137, 138 technological 49, 57 learning 26, 308 -by-doing 25, 27, 28–30, 42 collective 311 region 9 Lin, N. 275 linear models 310 Lisbon Strategy 187 Lithuania 187, 190, 192, 198 local action group 64 local control 61 local environment quality 313 local financing and management 65 local human resource base 58 local institutional setting 313 local integration of small firms 57–8 local level of scale see Victoria local sectoral specialization 58 lodging and food services 293 Lucas models 33 Lucas, R. 29–30 macroeconomic long-run growth theory 22 macroeconomic policies see European Union Malta 187, 192 Malthusian effect 28 manufacturing 294 see also employment growth and decline market fit 118, 119, 131, 136–7, 138 market imperfections 6, 115 MASST Model (macroeconomic, sectoral, social and territorial) 205, 212, 221–30 regional impact of macroeconomic policies 222–3, 224 structural funds 223, 225 territorial capital and regional growth 223, 226–30 measurement of endogenous growth see United States: economic development incentives and measurement of endogenous growth Mercedes Benz 290–91, 292, 296

20/12/2010 15:12

Index Metzker, C. 98 migration, simplified theories of 23 mixed-method research design 47 model determinants: Australia and United States 14, 111–39 Australia 120–28 concentration of jobs in broad industry sectors 128 income level 128 industry sector specialization 128 jobs in broad occupation categories 128 Local Government Areas 120–21, 123, 127 location on or near coast or to metropolitan area 128 model results 123–8 model variables 121 patterns of endogenous regional growth performance 121–3 population growth 128 unemployment 128 university and technical qualifications 128 evolution of regional economic development theory 113–15 measurement and modelling 115–20 exploratory model applications 119–20 new model framework 115–16 potential independent and intervening variables 118–19 proxy measure for dependent variable 116–17 United States 129–37 educational attainment level 131, 137 employment growth/decline 129 entrepreneurship capital 135–6, 137 exploratory model 129 general model results 131–6 government, regional 131, 135 method 129 Metropolitan Statistical Areas 120, 129, 130, 131, 132–4, 136, 137, 138 occupational structure of routine production 135

STIMSON PAGINATION (M2469).indd 333

333

population size 129, 131, 136, 137 spatial patterns of endogenous employment growth performance 129–31 specific model results 136–7 modelling local and regional environments 315–16 modern endogenous growth theory, precursors to 25–7 monetarism 3, 4 Moran’s I statistic 124 Myrdal-Kaldor model 307, 311 n+2 rule 185, 197 national economy perspective see foreign direct investment, knowledge assets and economic geography in Asian BRIICS countries national growth 221 national and regional dimensions of development 197 national savings 209 national shift effect 247–8 natural resources 313–14 nature of regional development 2–4 neo-endogenous development 62 neoclassical conditional convergence theory 208 neoclassical economics 23–4, 26, 32, 40, 207, 208, 304 neoclassical framework 302, 303 neoclassical models 23–4, 25, 30, 31, 33, 311 neoclassical production function 308 Netherlands 186, 223 networking 65 new economic geography 39, 41, 146, 307, 316, 317 new growth theory 1, 4–6, 30, 42, 144–6, 316 regional shift component 113–15 new regionalism 39–40, 41, 44–7, 54 new research challenges 316–18 New Zealand 161, 162 non-competition clauses 157 non-linear models 310–11 North American Industry Classification System code 287 NUTS 2 regions 190, 193

20/12/2010 15:12

334

Endogenous regional development

Okun’s Law 303 oligopoly 114 on-the-job training 29, 31 opportunity costs 209 ordinary least squares regression 117, 120, 121, 123, 131, 135, 138 Organizational Learning Laboratory 104 organizing metaphors 84 out-migration process of unemployed people from lagging regions 208 outcome measures 277 output 277, 287–8 paradigms, changing 3–4 partnership approach 64, 198 PayPal 97, 98–100 performance differentials see European Union: macroeconomic and territorial policies for regional competitiveness Pigouvian taxation 314 Pigouvian welfare rules 312 planning profession and rural, urban or regional endogenous development 13, 73–82 economic development planning 77–9 evolution of planning as indigenous/ endogenous assessment 74–6 land and its uses as endogenous planning base 76–7 social space, endogenous plan as 80–81 pluralism 47 Poland 187, 190, 192, 197, 198 political border effects 155 political costs 209 politicization 197–8 population super-concentration in urban areas 209 Portugal 225 position generators 274–5, 283 post-Fordism 59 post-modernism 59 Power Advantage Index 297 power of large corporations 58 price flexibility 206 process measures 277 product cycle theory 4–5, 113–14

STIMSON PAGINATION (M2469).indd 334

product cycle trap 78–9 production functions 24, 26–7, 152, 308 productive capital 10 productivity change 287–8 profit cycles 114 programming process 196–7 progress association 280 proxy variables 43 public services modernization 198 qualitative approach 54 qualitative information 47, 49–51 qualitative measures 2 qualitative methodologies 40–41 quantitative element 48 quantitative information 47 quantitative measures 2 quantitative methodologies 40–41 quantitative sociology 317 R&D 8, 9, 142–3, 145, 303, 308 Asian BRIICS countries 163–8 China 169, 170 costs 29 Indonesia 179 R-squared values 131, 136 racial segregation 76 random shocks 49 rationalism 3, 4, 7 rationality 217–18 bounded 215 realism 306, 307–10 receptivity 223, 226, 227, 231 recreational/sports facilities 313 Regional Competitiveness and Employment Objective 187–8 regional competitiveness performance cube 272 relational proximity of firms 309–10 rent-seeking 199 resource endowments 118, 119, 131, 136, 138, 271 retail industry 294 see also employment growth and decline in Australian capital city statistical divisions retention strategies (economic development stool) 288 revaluation 231

20/12/2010 15:12

Index reverse simulation 103–4 Romania 182, 190, 191, 192, 193, 195, 196 population and GDP per capita 186, 187, 188 Romer model 12 Route 128 87–94, 100, 106 rules of the game 85 Rural Transaction Centre 281 Russia 160 Saxenian, A. 87 Schumpeter, J. 20, 25–6 self-determination 61 shift-share analysis 117, 287–8, 291–2, 294, 295 Silicon Valley 5, 114 Silicon Valley: appreciative intelligence 13, 83–107 Asian immigrants 91–2 components and qualities of appreciative intelligence 96–8 discrimination 92 diversity and endogeny in explanatory paradigms 84–7 ecology of Silicon Valley versus Route 128 87–94 employee attachment 89–91 human resource practices 93–4 ‘old boys’ network 92 small worlds phenomenon 91 social networks 93 transferable lessons 100–107 simulation modelling 315 Single Currency programme 183 Single Market programme 183 Slovakia 187, 190, 192, 196, 197, 198 Slovenia 187, 190, 192, 196, 198 social attributes 80 social capital 5, 11, 42, 80–81, 114, 317 local level of scale in Victoria, Australia 271, 272–3, 274, 275, 282 social cohesion 185 social costs 209 social elements 221 social networks 93 social overhead capital 223, 226, 231 social planners 28–9 social regulation 40

STIMSON PAGINATION (M2469).indd 335

335

social space, endogenous plan as 80–81 sociological perspective 12–13, 59–67, 70–72 characteristics and values of ERD 60–61 cultural assets, potential 70–72 culture economy approach 62–3 definition 59–60 European Union ‘LEADER’ programme 63–6 exogenous development and endogenous development, comparison of 61–2 Solow model 1, 4, 11, 30, 32, 113 Solow residual 26 Solow-Swan formulation 42 South Africa 160 Spain 223, 225, 227, 229 spatial autocorrelation 138 spatial concepts 317 spatial development policies derived from globalization debate 210–11 spatial efficiency issue 301 spatial equilibrium 146 spatial patterns of endogenous employment growth performance (United States) 129–31 spatial proximity of firms 309 spatial socio-economic disparities or convergence 301 spatial spillover/proximity effects 138, 318 spatial sustainability issue 302 spatial-economic convergence idea 302 spatial/urban development policies, integrated 217 specific regional context 30–31 spin-off and spin-out phenomena 154 stability and growth 114 standardization stage 113 Star (human resource practice blueprint) 93 step-wise regression 120, 136, 138 Stimson, R.J. 3–4 strategic planning 279 structural assistance see European Union structural change and adjustment 115 structural characteristics 49

20/12/2010 15:12

336

Endogenous regional development

Structural Funds 184, 189, 193, 194–5, 197–8, 199–200, 223, 225 structure (indicator) 195 stylized facts using abstract mathematical reasoning 41–2, 43–4, 46–7 subsidiarity 61 sustainability 61, 302, 305, 312–16, 319 sustainable development 6, 10–11, 312 see also virtuous circle Sweden 145 systems of innovation framework 145–7 systems and tools (indicator) 195 Taiwan 168 tax burdens 290 technical augmentation factor 27 technological change 4, 24, 42, 113, 207 technological leadership 49, 57 technological progress 21, 24, 26, 113, 304, 308, 319 technology 8, 303 -based theories 5 high 49 -led economic growth 144 outflow from advanced regions 208 TENs 213 territorial capital 214–17, 226–30, 231 territorial policies see European Union territorial sustainability 319 Thatchenkery, T. 98 Third Wave Theory 77 third-way growth econometrics 42–3 time, concept of 311 time-series analysis 303 timing-related problems 199–200 total shift effects 247 trade analysis 25 trade protectionism 177 traditional management theories 101 trans-nationality openness 161 transfer-generating authority/principal agent problems 200 transition dynamics 49 transportation costs 307 trends in economic growth theory 301–5

STIMSON PAGINATION (M2469).indd 336

triple helix scenario 9 trust 5, 114 United Kingdom 9, 190, 191, 206, 223, 229 greenfield projects 164 United States 14, 26, 32, 269–70, 304 United States: economic development incentives and measurement of endogenous growth 16, 286–98 Bureau of Economic Analysis 286 Bureau of Labor Statistics 287 Department of Commerce 287 employment and establishments in Grayson County and Sector 311 food manufacturing and totals 293 federal system 289–90 Grayson County (Texas) 293, 297 incentives, influence of in local economic growth 288–91 Alabama 290–91 Texas 291 lodging and food services 293 manufacturing 294 Mercedes Benz 290–91, 292, 296 potential problems 294–7 retail trade 294 Ruiz Foods 291, 292–4, 297 Sherman-Denison Metropolitan Area (Texas) 292–4 shift-share calculations Grayson County (Texas) 294, 295 site selection factors according to corporate executives 289 Texas Enterprise Fund 291, 296, 297 United States: Emeryville (California) 79 federal deficit 33 greenfield projects 164 Jim Crow Laws 76 New Orleans 80 planner’s perspective 77 R&D 168 Route 128 86, 87–94, 97, 100, 106 San Jose (California) 80 Youngstown (Ohio): deindustrialization process 79 see also model determinants; Silicon Valley

20/12/2010 15:12

Index untraded interdependencies 216 urban scale 114 value 5, 114 Variance Inflation Factors 136 vector-autoregressive 30 verification 85–6 Vernon, R. 79 Victoria: local development initiatives in small rural towns 15–16, 268–84, 295 Bendigo Bank Community Bank 278, 281 case study shire councils 275–82 shire A 276–9 shire B 279–82 Community Connect Model 276 community contact team 277 Community Development Program 279–80, 281, 282

STIMSON PAGINATION (M2469).indd 337

337

community engagement 277 community plan development process 277, 278, 279, 283 purpose and method 269–70 ‘Sponsor a Cow’ campaign 281 theoretical views on endogenous development 271–5 virtuous circle for sustainable regional development 111–12, 269, 272 waste management 314 water provision and use 314 Weiss, J. 91 Williamson Law 205 Woolcock, M. 273 World Trade Organization 66 Yelp 99 YouTube 99

20/12/2010 15:12

STIMSON PAGINATION (M2469).indd 338

20/12/2010 15:12

STIMSON PAGINATION (M2469).indd 339

20/12/2010 15:12

STIMSON PAGINATION (M2469).indd 340

20/12/2010 15:12