Innovation and the Growth of Cities To Annabel, Ashley and Jane Innovation and the Growth of Cities Zoltan. J. Acs Doris E. and Robert V. McCurdy Distinguished Professor of Entrepreneurship and Innovation, Robert G. Merrick School of Business, University of Baltimore and US Bureau of the Census Edward Elgar Cheltenham, UK • Northampton, MA, USA © Zoltan J. Acs 2002 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 Glensanda House Montpellier Parade Cheltenham Glos GL50 1UA UK Edward Elgar Publishing, Inc. 136 West Street Suite 202 Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Cataloguing in Publication Data Acs, Zoltán J. Innovation and the growth of cities/Zoltan J. Acs. p.; cm. 1. Technological innovations – Economic aspects. 2. Industrial management. 3. Urban economics. 4. Economic development. I. Title. HC79.T4 A26 2002 307.1′416—dc21 2002018836 ISBN 1 84064 936 4 (cased) Typeset by Cambrian Typesetters, Frimley, Surrey Printed and bound in Britain by Biddles Ltd, www.biddles.co.uk Contents List of figures vi List of tables vii Foreword ix Preface xii 1Technology and entrepreneurship 1 2 Knowledge, innovation and firm size 24 3 Local geographic spillovers 44 4 Sectoral characteristics 63 5 Innovation of entrepreneurial firms 74 6 Capital structure, innovation and firm size 98 7 Employment growth in metropolitan areas 115 8 Employment, wages and R&D spillovers 135 9Heterogeneity versus specialization 154 10 Regional innovation systems 167 11 Epilogue: towards a ‘new model of regional economic development’? 192 Appendix A: The innovation database 196 Appendix B: Innovations, R&D lab employment and university research by state 203 Appendix C: Innovation, R&D lab employment and university research by MSA 205 Appendix D: Innovation, private R&D lab employment and university research by MSA and industry sector 209 Appendix E: Industry groupings 214 Appendix F: List of variables 215 References 217 Index 237 v List of figures 1.1 Plot of the high-tech employment–population ratio, 1989 and university research expenditure, 1985 5 1.2 Plot of employment ratio against the proportion of scientists and engineers in each MSA, 1989 5 7.1 High-technology employment growth in US metropolitan areas: a shift-share analysis 124 8.1 Plot of aggregate high-technology employment, 1989 and university research expenditure, 1985 140 8.2 Plot of aggregate high-technology employment and the number of scientists and engineers per 100 workers, 1989 141 vi List of tables 1.1 Number of innovations by county 4 2.1 Which states are the most innovative? 27 2.2 How do states compare on various measures of innovative activity? 28 2.3 Distribution of three-digit industries by state 29 2.4 Comparison among patent, university research and innovation measures 31 2.5 A comparison between regression results using Jaffe’s patent measure and the innovation measure 33 2.6 Innovative output in large and small firms and R&D inputs by state 39 2.7 Tobit regressions of innovative activity by state and technological area 40 3.1 Research design characteristics in recent studies 49 3.2 Significance of local geographic spillovers in recent studies 50 3.3 Regression results for log(innovations) at the state level 53 3.4 OLS regression results for log(innovations) at the MSA level 57 3.5 Regression results for log(private R&D) at the MSA level 59 3.6 Regression results for log(university research) at the MSA level 60 4.1 Industry detailed regression results for log(innovations) at the MSA level (1982) – OLS results 69 4.2 Industry detailed regression results for log(innovations) at the MSA level (1982) 71 4.3 Industry detailed regression results for log(private research) at the MSA level (1982) 72 5.1 The rate of new product innovation 91 6.1 Long-term debt to common equity, innovation rate and total asset size ($ m.) for the 30 least leveraged firms in the sample, 1982 102 6.2 Mean short-term, long-term and total debt to common equity, innovations and total assets ($ m.) by innovation class for 1982 103 6.3 Descriptive characteristics of the regression sample 108 6.4 Regression results (OLS) for short-term, long-term and total debt to common equity equations for all firms in 1982 109 vii viii Innovation and the growth of cities 6.5 Regression results for debt to common equity for large and small firms in 1982 111 6.6 Regression results for debt to common equity for large and small firms in 1982 corrected for heteroskedasticity 113 7.1 US high-technology employment 117 7.2 Shift-share analysis 122 7.3 Shift-share analysis: net relative growth 125 7.4 Shift-share analysis: industry mix 127 7.5 Shift-share analysis: competitive component 129 7.6 Shift-share analysis by size of metropolitan area 133 8.1 Linking university departments to industrial sectors 143 8.2 Summary statistics by variable 143 8.3 Aggregate high technology employment estimates 146 8.4 Disaggregated high technology employment estimates 148 8.5 Industry OLS high technology employment function estimates 151 9.1 High-technology employment by MSA and industry cluster, 1988 160 9.2 Mean industrial and university R&D by cluster 161 9.3 Summary statistics by variable 163 9.4 High-technology employment estimates 164 10.1 Systems of innovation 183 A.1 Distribution of large- and small-firm innovations according to significance levels 197 A.2 Number of innovations for large and small firms in the most innovative industries 199 A.3 Innovation, R&D lab employment and university research expenditure for US SMSAs 200 A.4 Most innovative firms, sales and R&D expenditure 201 Foreword Zoltan Acs, as my students might say, ‘gets it’. He is the kind of scholar who does not get hemmed in by disciplinary boundaries. He does not get bound up in conceptual mumbo-jumbo. He takes on real-world problems: what kinds of firms innovate? Where do they do it? And, what does this mean for cities and regions? When the real world looks different than theory, Acs is likely to think that maybe it’s the theory that’s got it wrong. To help get it right, he goes out and takes a good hard look at what’s really going on, collects new and unique data, and then tries to figure out just what causes what. Such is the case with Innovation and the Growth of Cities. The geography of the United States is being reshaped, Acs argues, and innovation holds the key. The book opens with the story of Dayton Ohio’s rise and decline – a metaphor for the rise and decline of the once-great American industrial heart- land and the seismic shift in the landscape of innovation and economic growth. These shifts are more than mere facts for Acs who hails from Cleveland and lived the very transformations of which he writes. Acs has long been a disciple of Schumpeter. In this book, he brings Schumpeter to geography, while bringing geography to Schumpeter. Innovation and the Growth of Cities situates Acs within a long and distin- guished intellectual tradition of thinkers who care about the connection of innovation and geography – from Alfred Marshall to Jane Jacobs. The great contribution of this tradition is that it marries geography to the long-held notion, established by both Schumpeter and Marx, that innovation is the driving force of economic growth. Innovation is not just an abstract economic process, nor one that is purely the province of firms. It does not emerge out of nowhere. In a very real sense, it comes from ‘somewhere’. The ‘new combi- nations’ that lie at the heart of innovation do not come from thin air; rather they are the product of pools of resources and interactions that are themselves concentrated in particular places. The great Jane Jacobs – who Robert Lucas has rightly suggested should be nominated for a Nobel Prize – showed long ago that innovation results from the creativity and diversity that concentrate in particular places. Cities, as Wilbur Thompson used to say, are the ‘incubators’ of innovation. And as the long sweep of economic history has shown, creative places – from Athens and Florence, to Manchester and Detroit, and more recently the Silicon Valley – are the cauldrons of innovation and economic growth. ix x Innovation and the growth of cities Acs understands this. The new geography, he argues, is not the result of natural endowments of land, labor and capital, as economists have long thought. Rather, he suggests, it is powered by innovation and entrepreneur- ship; and this in turn is the product of real people acting in real places. In other words, the factors that really matter are the ones we create for ourselves. We do this by recombining the knowledge and other resources in new and novel ways. What is more, some places are better than others at doing this. That is because they are able to attract, mobilize and connect the factors that really matter – innovative people and creative entrepreneurs. Innovation is indeed a very concentrated activity, as Acs shows. In the 60- year period between 1935 and 1995, the industrial heartland saw its share of patents drop from 85 percent to less than half; California overtook New York in innovative activity, while Texas surged past Illinois. Innovative centers like Silicon Valley, Seattle and Austin overtook older industrial cities like Cleveland, Buffalo and Pittsburgh as new centers for technology, entrepre- neurship and economic growth.