D I R E C T I O N S I N D E V E LO P M E N T 68360 Infrastructure Geography of Growth Spatial Economics and Competitiveness Raj Nallari, Breda Griffith, and Shahid Yusuf Geography of Growth Geography of Growth Spatial Economics and Competitiveness Raj Nallari, Breda Griffith, and Shahid Yusuf © 2012 International Bank for Reconstruction and Development / International Development Association or The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org 1 2 3 4 15 14 13 12 This volume is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dis- semination of its knowledge, this work may be reproduced, in whole or in part, for noncom- mercial purposes as long as full attribution to the work is given. For permission to reproduce any part of this work for commercial purposes, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. ISBN (paper): 978-0-8213-9486-1 ISBN (electronic): 978-0-8213-9487-8 DOI: 10.1596/978-0-8213-9486-1 Library of Congress Cataloging-in-Publication Data Nallari, Raj, 1955- Geography of growth : spatial economics and competitiveness / by Raj Nallari, Breda Griffith, and Shahid Yusuf. p. cm. Includes bibliographical references. ISBN 978-0-8213-9486-1—ISBN 978-0-8213-9487-8 (electronic) 1. Space in economics. 2. Urban economics. 3. Economic geography. I. Griffith, Breda. II. Yusuf, Shahid, 1949- III. World Bank. IV. Title. HT388.N36 2012 330.9173’2 — dc23 2012008450 Cover photo: view of Sofia, Bulgaria by Boris Balabanov/World Bank. Contents Preface xi About the Authors xiii Abbreviations xv Chapter 1 Frameworks for Spatial Analysis 1 The Form of Urbanization 2 Agglomeration Economies 8 Conclusion 11 Notes 12 References 12 Chapter 2 Urbanization as a Typology of Space 15 Urbanization and Space 16 Classification of Cities 18 Size of Cities across Developing and Developed Regions 20 Criticism of the Data and Suggested Alternatives 23 Conclusion 26 Notes 27 References 28 v vi Contents Chapter 3 Urban Transition and Growth 29 Urbanization and Development 30 Urbanization in Developing Countries 38 Key Features of Cities in Developing Countries 43 Conclusion 46 Notes 47 References 48 Chapter 4 Spatial Concentration and Specialization 51 Specialization of Cities 52 Knowledge Cities 54 The Creative City 65 The Global City 73 Green Cities/Eco Cities 76 Conclusion 81 Notes 82 References 83 Chapter 5 The Attributes and Role of “Smart Cities� 89 Growth and Technology-Intensive Subsectors 91 What Makes Cities Smart 99 Toward an Urban Innovation Strategy 108 Policy Measures That Facilitate Technological Upgrading and Innovation 112 Identifying and Promoting Smart Cities 117 Annex: Technology Capability and Innovation Criteria 118 Notes 120 References 122 Chapter 6 Globalization, Urban Regions, and Cluster Development 127 A Holistic Approach to Development 128 The Role of Clusters 128 Policies to Support Clusters 130 Sustaining Clusters 130 Conclusion 132 Notes 134 References 134 Contents vii Chapter 7 Urban Development and Growth 137 Urbanization: From Canter to Gallop 139 The Metropolitan Powerhouse 143 Metropolitan Challenges 146 Wealth of Cities and National Policies 147 Notes 149 References 153 Chapter 8 Elements for Future Success of Metropolitan Regions 157 The Industrial Matrix 159 Connectivity 163 The Smarter Metropolis 167 Governing the Metropolitan Center 169 The Resilience Imperative 170 Metropolitan Futures 174 Annex: City Rankings 176 Notes 178 References 181 Box 4.1 Indexes Used in the Global City Indicators Program 76 Figures 1.1 Primacy and Economic Development, 1965–95 4 1.2 Concentration of Economic Activity in the United States 6 2.1 Total Population, by City Size, 1995, 2009, and 2025 19 2.2 Distribution of the World Urban Population, by Region, 1950, 2009, and 2050 22 2.3 Key Indicators of the Agglomeration Index 25 2.4 Agglomeration Index and UN Estimates of Urban Population, by Region, 2000 25 2.5 Sensitivity to Indicators Used: Example of Minimum Population Size of Large Cities 27 3.1 Relationship between GDP and Spatial Concentration 32 3.2 Density and GDP per Capita in Selected Countries, by Phase of Urbanization 34 3.3 Rural-Urban Disparities in GDP per Capita 35 viii Contents 3.4 Rural-Urban Disparities and Density in the Philippines, China, and India, Various Years 36 3.5 Change in Urban Population with and without China and India, 1985–2005 38 4.1 GCIF Membership in 2010, by Population Category 77 5.1 Shanghai R&D Public Service Platform 115 6.1 Ranking of Metropolitan Cities 133 7.1 Strong Correlates of Urban Productivity (City GDP per Capita) in China, 2007 142 7.2 Exposure of People to Cyclones and Earthquakes, 2000 and by 2050 144 7.3 Coastal Population of Selected Countries That Are Highly Vulnerable to Sea Level Rise 145 8.1 Industry Contributions to Productivity Growth in the United States, 1960–2007 160 8.2 R&D Intensity, by Industry Averaged across 10 OECD Countries 161 8.3 Contributions to Productivity Growth in the United States, by Industry, 1960–2007 163 8.4 Industry Contributions to Productivity in the United States, 1960–2007 164 Tables 1.1 Regional Development Policies Calibrated to Integrate Countries, by Density of Population 9 2.1 Population and Average Annual Rate of Change, by Group and Selected Years, 1950–2050 17 2.2 Number of Cities and Percentage of Total Population, by Size of City, 2009 and 2025 19 2.3 Size of Cities, by Region, Number of Inhabitants, and Share of Population, Selected Years, 1975, 2009, 2025 20 2.4 Percentage of Population Living in Urban Areas by Region, Selected Years, 1950–2050 21 2.5 Average Annual Rate of Change in Urban Population, 1950–2025 23 2.6 National and UN Data on Urbanization in Selected Countries 24 3.1 Regional Differences in Urbanization 39 3.2 A Dozen Economies of Scale 44 Contents ix 4.1 Manufacturing and Business Services in the United States, by Size of City, 1910 and 1995 53 4.2 Share of New York County (Manhattan) in Total Private Employment in the United States, 1997 54 4.3 Dimensions of Knowledge Base: Measures and Results 56 4.4 The Knowledge Base and Economic Performance in Select Cities 57 4.5 Metropolitan Area Regressions 59 4.6 Underpinnings of Knowledge Cities 63 4.7 Creativity Rankings in the United States, by City Size 67 4.8 Creative Class Occupations, Ranked by Percentage Change 69 4.9 Creative Workers: Consumers and Producers 71 4.10 Global City Indicators: City Services 75 4.11 Global City Indicators: Quality of Life 75 4.12 Role of Standardized Indicators for Cities 77 4.13 Summary of Results on Carbon Emissions per Home, 2006 78 5.1 Sources of GDP Growth in China, 1978–2004 92 5.2 Sources of GDP Growth in the Industrial and Services Sectors in China, 1978–2004 92 5.3 Fastest-Growing Manufactured Exports Worldwide, 1997–2007 93 5.4 Fastest-Growing Manufactured Exports from Asia, 1997–2007 94 5.5 Fastest-Growing Manufacturing Industries in China, 1996–2003 95 5.6 Top 10 Exports from China, 2006 95 5.7 Imports to China, 2002, 2005, 2008 96 5.8 Imports of High-Tech Products as a Percentage of Total Imports in China, 2002, 2005, 2008 96 5.9 Top USPTO Patents Worldwide, 2005–09 97 5.10 Share of WIPO Patents, by Sector, 2007–09 98 5.11 Top Five Patenting Industries in the United States, 2006 98 5.12 Top Five Industries Contributing to TFP Growth in the United States, 1960–2005 99 5.13 Science and Technology Occupations in the United States 103 x Contents 5.14 S&T Jobs in Select High-Tech Industries in the United States, 1997 104 5.15 High-Tech Jobs in Select Cities in the United States, 1997 104 5.16 IT Jobs in Select Cities in the United States, 1997 105 5.17 Key High-Tech Sectors in Seattle 105 7.1 Contribution of Manufacturing and Services to GDP, 1980–2008 140 8A.1 Mercer Quality of Living Ranking of Cities Worldwide, 2010 176 8A.2 Ranking of Creative Cities in the United States, by Arts Employees per Capita, 2008 177 8A.3 Ranking of Innovative Cities in the United States, 2008 177 8A.4 Top 10 Innovation Cities in the World, 2010 177 8A.5 Ranking of Innovation Cities in the Americas, 2010 178 Preface Economists have emphasized the importance of geography in growth and competitiveness, yet rarely has there been literature that identifies the cause of growth in some cities but not in others. Why was the city of Bangalore more attractive for industries than Karachi? What are the defining characteristics of successful cities? This book seeks to answer these questions through multiple consultations with leading experts and in-depth research on urban centers. Geography of Growth has been written for academics and practitio- ners; it combines the theoretical background on urban centers with concrete recommendations. The eight chapters move from providing background on the various models of urban centers to hypothesizing why growth and development are more prominent in some cities than in others. Chapter 1 addresses two questions: How has spatial concentration evolved with growth and development, and what are the efficiency implications of too much or too little spatial concentration? This chapter summarizes the various models that analyze growth by geographic concentration and sets the foundation for concepts discussed in later chapters. Chapter 2 focuses on urbanization in geographies. There is pressure to effectively measure the urban population, which is expected to grow by xi xii Preface 84 percent within the next 40 years. This chapter discusses how UN data measures growth, as well as the criticisms on the metrics used. Chapter 3 correlates urban presence with economic density in devel- oped and developing countries. It initially focuses on how urban transition and growth are blurring the rural-urban divide and the unprecedented volume of people who are moving to urban areas. The second part exam- ines regional trends in urban growth for the developing countries, then discusses some key features of cities in developing regions. Chapter 4 discusses how different industries inhabit and impact vari- ous urban sectors. For example, the chapter opens by describing how small and medium cities in Japan, the Republic of Korea, and the United States are highly specialized because of the requirements and influences of the types of industry there. This chapter expands on the economic cor- relations mentioned in chapter 3. Chapter 5 contextualizes urban growth in the current technological landscape as innovation, particularly in information technology, has become critical to increasing productivity and consequently growth. This chapter provides examples of “smart cities� and identifies common attri- butes that contribute to their success. This chapter also provides policy recommendations for practitioners on how to make cities “smarter.� Chapter 6 further analyzes urbanization in the current global context, specifically, the impact of globalization and industry clusters on urbaniza- tion. By citing examples of how globalization has had spillover effects in the urban sector, the chapter demonstrates the importance of globaliza- tion and the relevance of the growth of industry clusters in places such as Bangalore and Shenzhen. It segues to chapter 7. Chapter 7 addresses a current fundamental global trend: Why has urbanization been growing rapidly since the 1950s? Some theories suggest that it is industry that spurs urbanization and consequently growth in infra- structure, however this is not the case. Instead—the chapter concludes by looking at data across regions and cities—the municipalities are pivotal in influencing infrastructure development and growth in urban centers. Finally, chapter 8 deciphers why some cities are more successful than others. Why do Karachi and São Paulo have the human capital that qualifies them as urban centers but not as thriving cities? By citing examples of successful cities, this chapter provides policy recommenda- tions on how to make a city competitive in today’s economy. The authors would like to thank the policy experts and academics who helped identify key data, as well the administrative and publishing staff for helping in the successful production of this book. About the Authors Raj Nallari is the sector manager for the Growth and Competitiveness Practice at the World Bank Institute (WBI). He has worked at the World Bank for more than 20 years in various departments. Previously he worked at the International Monetary Fund. Raj has published on various topics, including growth adjustment systems, the labor market and gen- der, and macroeconomics. He has also edited several volumes of Development Outreach. He holds a PhD in economics from the University of Texas at Austin. Breda Griffith has worked as a consultant with WBI since 2005 in the areas of growth, poverty, gender, development, and labor markets. Her experience is deep and wide-ranging with publications in refereed jour- nals on development and language maintenance, entrepreneurship, and small business. Breda has also co-authored books on economic growth, poverty, gender and macroeconomic policy, new directions in develop- ment, and labor markets in developing countries. She holds a PhD in economics from Trinity College Dublin, Ireland and an MA in economics from the National University of Ireland. Shahid Yusuf joined the World Bank in 1974 as a Young Professional; while at the Bank, he spent more than 35 years tackling issues confronting xiii xiv About the Authors developing countries. He has written extensively on development issues, with a special focus on East Asia, and has also published widely in various academic journals. Shahid is currently Chief Economist for the Growth Dialogue at the George Washington University School of Business. He holds a PhD in economics from Harvard University and a BA in econom- ics from Cambridge University. Abbreviations AI agglomeration index FDI foreign direct investment GCIF Global Cities Indicators Facility GCIP Global Cities Indicators Program GDP gross domestic product GHG greenhouse gas GVIO gross value of industrial output ICT information and communications technology IT information technology MNC multinational corporation MSA metropolitan statistical area OECD Organisation for Economic Co-operation and Development PMSA primary metropolitan statistical area R&D research and development SAR Special Administrative Region SME small and medium enterprise S&T science and technology TFP total factor productivity UN United Nations USPTO U.S. Patent and Trademark Office WIPO World Intellectual Property Organization xv CHAPTER 1 Frameworks for Spatial Analysis Since the 1990s, the new economic geography has received a lot of atten- tion, as mainstream economists such as Krugman (1991a, 1991b) and others began to focus on where economic activity occurs and why. While economic geography has always been central to such questions, it has often been ignored, given the difficulties with modeling some of the relationships—for example, increasing returns and imperfect competition at the regional and urban levels. Using models to analyze industrial orga- nization, international trade, and growth theory has helped to spur the use of economic geography, which seeks to explain concentrations of population, economic activity, or both, such as agricultural areas, indus- trial areas, cities, and industry clusters. These concentrations of popula- tion or economic activity are subject to agglomeration economies and are thus self-reinforcing. The new economic geography seeks to understand why such concentrations arise and why they are self-reinforcing.1 Our concern here is not with the new economic geography per se but rather with the forces that give rise to spatial concentration of population and economic activity: How has spatial concentration evolved with growth and development, and what are the efficiency implications of too much or too little spatial concentration? 1 2 Geography of Growth The Form of Urbanization Urbanization is a transitory process (Henderson 2003) exemplified by movements of population from rural to urban areas predicated on the marginal product of labor being higher in the urban area. The Harris and Todaro (1970) model, the workhorse for most development economists, deals with issues concerning rural-urban migration, where workers decide to migrate or not based on expected income differentials between rural and urban areas rather than just wage differentials. Here, we assume that agriculture is the traditional sector, with low productivity, low wages, and no unemployment, as labor is perfectly competitive and mobile. This implies that rural-urban migration in a context of high urban unemploy- ment can be economically rational if expected urban income (defined as actual wages adjusted for the unemployment rate) exceeds expected rural income. In equilibrium, the expected urban wage is equal to the marginal product of an agricultural worker, and there is no migration. However, economic activity began to concentrate in certain places, leading to uneven growth and development. The assumption of constant returns to scale as in the Harris-Todaro model could not explain the unevenness or “spikes in economic maps.� Regional and urban develop- ment seemed to be better explained by monopolistic competition and increasing returns to scale of production, and this meant that the assump- tion of perfect competition in models of spatial economics had to be dropped. Around this time, Dixit and Stiglitz (1977) developed a model of monopolistic competition, and this proved useful in several fields, including spatial economics, new trade theory, and new growth theory: monopolistic competition gives rise to economic power, which in turn can lead to increasing returns. Henderson (2003) acknowledges the contribution that these dual- economy models make but points out critical defects. First, the dual start- ing point is assumed and not modeled. Second, there are no forces for agglomeration where we would expect to see industrial concentration in the urban sector. And third, there is little mention of spatial aspects of the economy. The core-periphery models, especially those configured in an economic growth context, and Krugman’s 1991 interpretation attempt to address these three issues. Core-periphery models are “more about urban concentration� (Henderson 2003, 5). Urban Concentration Urbanization involves the movement out of rural areas and into urban areas, and government policy can have an effect on this transition and on Frameworks for Spatial Analysis 3 the sectoral composition of national output. For example, central govern- ment policies that promote labor mobility, develop infrastructure, and remove impediments to internal trade will affect the mix of urban-rural population. Other policies of local governments such as the provision of public goods and amenities also get reflected in the urban cost-of-living curve and may influence the population mix. At initial and middle stages of development, government policies—through tariffs, price controls, and subsidies—directly affect the national composition of output and indi- rectly affect urbanization. Urban concentration is usually predicated on the existence of urban primacy and its robustness over time. For example, the convergence- divergence debate in the economic growth literature has also been applied to urban primacy—that is, it asks, Do the other urban centers converge on the primary city over time? What are the economic policies hindering or helping such a process? The models assume a dual-economy approach: the primary city and other urban places. Primacy is the simplest measure of urban concentration, and a common expression is the ratio of the population of the largest metro area to all of the urban population in the country (Henderson 2003).2 Results from the empirical studies suggest that there is an inverted U-shape relationship, where relative concentration first peaks and then declines with economic development (figure 1.1). The U shape is more relevant in earlier (1965–75) than in later (1985–95) periods (Henderson 2003). Optimal primacy is the level that maximizes national productivity growth, with large deviations in primacy strongly affecting productivity growth.3 Henderson notes a tendency toward primacy in Algeria, Argentina, Chile, Mexico, Peru, and Thailand. Political economy and government policies can be complicit in fostering excessive concentration and favoring a primary city (usually the national capital) over other cities. Favored cities tend to draw in enormous populations that create an “extremely congested high cost-of-living metro area� (Henderson 2003, 10). Local government can help to preempt this unsustainable situation, and Henderson notes previous efforts in China that did limit internal migration. Democratization and fiscal decentralization tend to disfavor the existence of primary cities. Ades and Glaeser (1995) find that if the primary city is a national capital, it is 45 percent larger; if the country is a dictatorship, then the primary city is 40–45 percent larger. Davis and Henderson (2003), using panel data from 1960–95 with instrumental variable estimation, find that moving from the extreme of most to least centralized government reduces primacy by 5 percent. Moving from the extreme of least to most democratic form of government reduces primacy by 8 percent (Henderson 2003). 4 Geography of Growth Figure 1.1 Primacy and Economic Development, 1965–95 a. 1965–75 0.75766 population of largest metro area as % of total population 0.050184 5.66988 In (real GDP per capita) 9.52384 b. 1975–95 0.962783 population of largest metro area as % of total population 0.035561 5.70044 In (real GDP per capita) 9.90231 Source: Henderson 2003. Note: GDP = gross domestic product. Urban infrastructure (roads, airports, well-functioning rental markets) can reinforce productive efficiency and reduce primacy. Increasing the amount of roads per square kilometer of national land or the amount of navigable inland waterways per square kilometer, ceteris paribus, by one standard deviation reduces urban primacy by 10 percent. Urban policy making in developing countries often has the twin objectives of making cities better by improving the delivery of local public goods, such as gar- bage collection, sewerage, and public transport and by limiting urbaniza- tion, which implies limiting the inflow of people from rural areas to crowded cities. Cities bring efficiency gains and, therefore, economic benefits that are derived largely from between- and within-city allocation of Frameworks for Spatial Analysis 5 resources. Therefore, limiting urbanization entails losses. The policy objec- tive should be to deal with imbalances (slums, congestion) rather than to stop the influx of people. The bottom line is to avoid concentrating resources in one place, which implies that governments should ensure competition in all markets and secure well-defined property rights and land markets if they want healthy urban areas. During the past few decades, the costs of transport and communica- tion have been reduced dramatically, and space as an impediment has been minimized. For example, lower transport and communication costs have enabled firms and their workers to compete globally across both developed and developing countries. Local markets, regional markets, and international markets are now seen as seamless “world markets� and open for competition. As such, all markets have to beat the “China price of low wage,� and this requires firms to innovate constantly to stay ahead of their competitors. Thus, China, Japan, and the Republic of Korea are able to produce steel in areas that are not known to have iron and coal mines. Locations now compete on the basis of providing secure property rights and law and order rather than strategic location or proximity to a seaport. Location becomes redundant for many services, as communications have become increasingly more accessible and mobile. Thus, numerous low- skill service jobs have begun to migrate from high-wage locations to low- cost locations overseas in places as diverse as Brazil or India or Ireland. Closeness and connectivity are important and concentrate goods and workers in the center. Core-Periphery Models The core-periphery models focus on the effect of transportation costs on spatial concentration. Krugman (1991a) introduces a basic trade model of “core-periphery� with two goods (agriculture and manufacturing) in two regions with two types of labor (farmers and workers), where increasing returns at the firm level interact with transport costs, and factor mobility can cause spatial economic structures to emerge and change. Using an approach originally suggested by Alan Turing (1952) for the analysis of morphogenesis in biology yields surprisingly clear results about this two-region economy. For instance, Krugman (1996) outlines a “racetrack economy� with 12 regions around the circumference of a cir- cle, like a clock, in which goods must be transported along the circumfer- ence. When one starts with any initial distribution of economic activity nearly equal across all space, the simulation always ends with all manu- facturing agglomerated in just two regions, which are located on the 6 Geography of Growth exact opposite side of each other. This self-organizing central place of activity can also be derived using Turing’s model, which is also called the reaction diffusion model. International trade and spatial economics are also linked. While trade theory deals with the immobility of factors of production (land and labor) between locations and the mobility of output or commodities, spatial eco- nomics is concerned about the mobility of factors (labor and capital) and the implications for the concentration of economic activity. The approach introduces the possibility that labor can move between agriculture and manufacturing and assumes that manufacturing firms use each other’s outputs as intermediate inputs. This yields backward and forward linkages, as in the core-periphery model, but here it causes international inequalities in wages and could explain why some nations prosper while others decline. If there were no countries in the world, just one continuous space across the globe, this framework could be used to show the emergence of regions of specialization in a borderless world with continuous space. At one extreme lies the core-periphery dichotomy at the global scale. In 2000, for example, the North American Free Trade Agreement yielded 35 percent of world gross domestic product (GDP), the European Union (15 countries) yielded 25 percent, and East Asia yielded 23 percent; thus, 83 percent of world GDP was concentrated in three regions. Furthermore, the concentration of world GDP in these three regions has been intensify- ing since 1980. At a national scale, about 50 percent of the output in the United States is produced on less than 2 percent of the country’s land (figure 1.2). Output is concentrated in a few counties such as the area along the Boston–New York–Washington, DC, corridor, in the state of Florida, in and around Galveston-Houston, in Silicon Valley in California, Figure 1.2 Concentration of Economic Activity in the United States Source: Easterly and Levine 2002. Note: Counties shown in black take up 2 percent of U.S. land area but account for half of U.S. GDP. Frameworks for Spatial Analysis 7 and in a few other places. A similar pattern is observed within cities such as Washington, DC, where rich neighborhoods are concentrated in the northwest of the city and poor neighborhoods are concentrated in the southeast. Economic maps of France, Germany, India, Japan, Poland, and most other countries show a similar pattern, with “spiky� economic density (GDP per square kilometer). Maps of economic density indicate that some areas are lagging in relation to others—for example, Bihar, Orissa, and Rajasthan in India. The existence of lagging regions motivates policy and the commitment of large amounts of resources in the form of taxes and subsidies to these regions. Several concerns arise from such uneven growth and development. First, at the local level, people are concentrated in cities, and this trend threatens to outstrip the concentration of economic activity. As a result, a billion people live in the world’s slums, and this problem has to be dealt with in policy and planning. At the national level, spatial disparities in living standards widen as economic mass concentrates in leading prov- inces and lagging regions are left behind: a billion people are now margin- alized in remote and lagging areas of the world. At the global level, poor people are trapped in isolated countries that are not developing at all (Paul Collier calls them the “bottom billion�). Should rural labor move to jobs, which are more likely to be in cities and leading regions, or should jobs be provided in remote parts of a coun- try? The general policy advice is that governments should remove imped- iments to the flow of capital and labor and reinforce agglomeration economies by abolishing national minimum wages, reducing unemploy- ment and social benefits, and abolishing rent control to increase the sup- ply of housing, among other policies. Do agglomeration economies accrue at the plant, industry, city, or regional level? Krugman (2010) believes that they accrue at the plant level, whereby firms are located in a single area nearer to consumer demand (in large cities with large populations), which minimizes transport costs. In contrast, Michael Porter (1990) is of the view that intraindustry economies lead to clusters. This debate has implications for policy: Should governments target industrial policy to facilitate individual firms or to facilitate a whole industry? The World Development Report 2009 (World Bank 2009) emphasizes that the most potent instruments for integrating leading and lagging regions are spatially blind improvements in institutions or, put more simply, the provision of essential services such as education, health, and 8 Geography of Growth public security. The main findings of research on economic growth are that (a) growth will remain unbalanced and, to try to spread out economic activity—too much, too far, or too soon—will discourage it (Gill 2010), and (b) development can still be inclusive, in that even people who start their lives far from economic opportunity can benefit from the growing concentration of economic activity in a few places. The way to get both the benefits of uneven growth and inclusive development is through economic integration and mobility of labor and capital to fast-growing areas, so (c) leading and lagging regions need to be connected not only by infrastructure that connects places but also by institutions (for example, political and fiscal decentralization, property rights for land, financial integration, education and training, and assured product markets) that connect people (table 1.1). Agglomeration Economies Agglomeration or clustering of activities (shops, restaurants, movie the- aters) can happen when a new road or a new factory is built. Agglomeration occurs at many geographic levels and can take many forms. It is most intense at the level of the city, where close spatial proximity makes the prospect of agglomeration spillovers—both scale externalities and knowl- edge economies—most relevant and a factor in promoting economic growth (Henderson 2003). Important forces affect geographic concentration and dispersion of economic activity. These are the centripetal forces, such as backward and forward linkages, thickness of markets, and knowledge spillovers. They are countered by or opposed to agglomeration by centrifugal forces such as immobile factors, transport costs, land rents, congestion, pollution, and other pure diseconomies. External effects are exerted on both centripetal and centrifugal forces, creating market failures that abound in urban eco- nomics. These forces determine the size of the city—in other words, the agglomeration may be too big. Alfred Marshall ([1890] 1920) suggested over a century ago a three- fold classification of the reasons for industrial concentration. Concentration arises because of (a) knowledge spillovers, (b) the advantages of thick markets for specialized skills, and (c) the backward and forward linkages associated with large local markets.4 Most of the theoretical and empirical work has focused on the backward and forward linkages. These linkages lead to increasing returns to production at the level of the individual firm. Increasing returns introduce self-reinforcing and multiplier effects and Table 1.1 Regional Development Policies Calibrated to Integrate Countries, by Density of Population Sparsely populated Densely populated lagging Densely populated lagging Indicator lagging regions regions in united countries regions in divided countries Example (countries) Chile, China, Ghana, Honduras, Bangladesh, Brazil, Colombia, Arab India, Nigeria Pakistan, Peru, Russian Federation, Rep. of Egypt, Mexico, Thailand, Sri Lanka, Uganda, Vietnam Turkey Dimensions of the integration Economic distance Economic distance; high population Economic distance; high population challenge densities in lagging areas densities; internal divisions What policies should facilitate Labor and capital mobility Labor and capital mobility; market Labor and capital mobility; market integration for goods and services integration for goods and services; selected economic activities in lagging areas Policy priorities Spatially blind institutions Fluid land and labor markets; Fluid land and labor markets; Fluid land and labor markets; security; education and health security; education and health security; education and health programs; safe water and programs; safe water and programs; safe water and sanitation sanitation sanitation Spatially connective infrastructure Interregional transport infrastructure; Interregional transport infrastructure; information and communication information and communication services services Spatially targeted incentives Incentives to agriculture and agro- based industry; irrigation systems; workforce training; local roads Source: World Bank 2009. Note: Countries are classified based on a typology outlined in World Bank (2009). Three types of countries are identified: those with sparsely populated lagging regions; those with densely populated lagging regions that are united based on ethnolinguistic and little or no political fragmentation; and those with densely populated lagging regions and within- country divisions such as ethnolinguistic differences and political fragmentation. 9 10 Geography of Growth coexist with imperfect competition, which has implications for how firms set prices and wages. von Thünen’s model (1826) explained the economic effects of falling “space-bridging� costs. Owners of mobile fac- tors of production, such as capital and technical knowledge, need to be paid the same return whether their assets are employed in the center of market activity or in the periphery. Otherwise there is an incentive to engage in “locational arbitrage.� For example, in a city or region, real estate rents drop as the distance from the center of activity grows. In the center, enterprises use a lot of capital to build high-rises, thus saving on land costs, and only space-saving offices are located there. Cheap land on the periphery is devoted to space-intensive uses, such as manufacturing plants, logistics centers, and dumps. If landowners on the periphery were to raise their rents, they would soon be out of business. Why do certain cities such as Detroit for automobiles, Hollywood for movies, and Silicon Valley for high technology specialize in a narrow range of industries? A small modification of Von Thünen’s model shifts the focus from agglomeration of resources to the geographic concentra- tion of particular industries. When production is vertically integrated, with both upstream and downstream sectors producing inputs for each other, and when both sectors have higher returns and transport costs, there are both backward and forward linkages, and production could be in a single location because both have an incentive to be closest to the largest markets for each other. What happens to spatial concentration if manufacturing is more dynamic and mobile, but agriculture is immobile? What happens to eco- nomic mass if both manufacturing and agriculture are mobile? A model that combines a von Thünen–style explanation of land rent with a linkage explanation of manufacturing concentration can show how a spatial pat- tern in which a single city is surrounded by an agricultural hinterland can be self-sustaining as long as the urban population is not too large. If the population does become too large, it will be in the interest of a small group of workers to move to some other location; by using the criterion of sustainability, it is possible to develop a model of the emergence of new cities and hence a multicity structure. If several manufacturing industries exist with different costs of transportation or economies of scale, the process of city formation can yield a hierarchy of cities of different types and sizes. A larger city allows for a more efficient sharing of indivisible facili- ties (such as local infrastructure), risks, and the gains from variety and Frameworks for Spatial Analysis 11 specialization. Furthermore, a larger city also allows for a better matching between employers and employees, buyers and suppliers, and entrepre- neurs and financiers. Finally, a larger city can facilitate learning about new technologies, market evolutions, or new forms of organization. More frequent direct interactions between economic agents in a city can thus favor the creation, diffusion, and accumulation of knowledge. Hence, many different mechanisms can generate increasing urban returns. Moreover, sources of increasing urban returns may also be sources of urban inefficiencies. The wage in a city increases with the size of the urban labor force, reflecting the existence of urban agglomeration externalities. The inten- sity of increasing urban returns is measured by the slope of the wage curve. Since the nature and intensity of increasing returns are expected to differ across activities, so will the exact shape of the wage curve. This upward-sloping wage curve stands in sharp contrast to “neoclassical� wage curves that slope downward. Increasing urban returns have received a considerable amount of theoretical attention. The wage curve masks the many distortions beneath it. For example, most developing countries distort agricultural prices (rural) and manufac- turing prices (urban). The “urban bias� of most developing countries is well documented, as reflected in higher urban wages and thus a higher wage curve. In turn, this bias should lead to larger cities. More generally, national technology and public policies are reflected in the wage curve of any particular city, affecting its level and, sometimes, its slope. The cost-of-living curve in urban areas is linked to the wage curve and traffic congestion, among other determinants. Governments can address many components of the cost-of-living curve, from sewerage to public transport, and thus reduce the costs. Conclusion The spatial economy was given new impetus in the 1990s with the work on the new economic geography, which provided economists with new tools to examine why and where population or economic activity is located. It also tied in with the theoretical and empirical work of the new economic growth literature at that time. The chapter has discussed the two principal ways in which urbanization is organized at a spatial scale. Central to this are the study of urban concentration and the study of agglomeration economies. 12 Geography of Growth Notes 1. “By modeling the sources of increasing returns to spatial concentration, we learn something about how and when these returns may change—and then explore how the economy’s behavior will change with them� (Fujita, Krugman, and Venables 1999, 4). 2. Henderson (2003) cites the empirical work of Ades and Glaeser (1995), Junius (1999), and Davis and Henderson (2003). 3. “A 33 percent increase or decrease in primacy from a typical best level of 0.3 reduces productivity growth by 3 percent over five years� (Henderson 2003, 9). 4. Concentration minimizes transport costs, but also creates other externalities, such as congestion and overcrowding. References Ades, Alberto F., and Edward L. Glaeser. 1995. “Trade and Circuses: Explaining Urban Giants.� Quarterly Journal of Economics 110 (1): 195–227. Collier, P. 2007).The Bottom Billion. Why the Poorest Countries are Failing and What Can Be Done About It. Oxford, England: Oxford University Press. Davis, James, and J. Vernon Henderson. 2003. “Evidence on the Political Economy of the Urbanization Process.� Journal of Urban Economics 53 (1, January): 98–125. Dixit, Avinash K., and Joseph E. Stiglitz. 1977. “Monopolistic Competition and Optimum Product Diversity.� American Economic Review 67 (3): 297–308. Easterly, William, and Ross Levine. 2002. “It’s Not Factor Accumulation: Stylized Facts and Growth Models.� In Economic Growth: Sources, Trends, and Cycles, ed. Norman Loayza and Raimundo Soto, 61–114. Santiago: Central Bank of Chile. Fujita, Masahisa, Paul Krugman, and Anthony J. Venables. 1999. The Spatial Economy: Cities, Regions, and International Trade. Cambridge, MA: MIT Press. Gill, Indermit. 2010. “Regional Development Policies: Place-Based or People- Centred?� Vox.org, October 9. http://www.voxeu.org/index.php?q=node /5644. Harris, John R., and Michael P. Todaro. 1970. “Migration, Unemployment, and Development: A Two-Sector Analysis.� American Economic Review 60 (1): 126–42. Henderson, J. Vernon. 2003. Urbanization, Economic Geography, and Growth. Providence, RI: Brown University Press. Junius, Karsten. 1999. “Primacy and Economic Development: Bell Shaped or Parallel Growth of Cities?� Journal of Economic Development 24 (1): 1–22. Frameworks for Spatial Analysis 13 Krugman, Paul. 1991a. Geography and Trade. Cambridge, MA: MIT Press. ———. 1991b. “Increasing Returns and Economic Geography.� Journal of Political Economy 99 (3, January): 483–99. ———. 1996. Self-Organizing Economy. Cambridge, MA: Blackwell. ———. 2010. “The New Economic Geography, Now Middle-Aged.� Presentation for the Association of American Geographers, April 16. Marshall, Alfred. [1890] 1920. Principles of Economics: An Introductory Volume, 8th ed. London: Macmillan. Porter, Michael E. 1990. The Competitive Advantage of Nations. New York: Macmillan. Turing, Alan. 1952. “The Chemical Basis of Morphogenesis.� Philosophical Transactions of the Royal Society of London, Series B 237 (641): 37–72. von Thünen, J. H. 1826. Der Isolierte Staat in Beziehung auf Landtschaft und Nationalökonomie. Hamburg: Perthes. English translation: von Thünen’s Isolated State, trans. C. M. Wartenberg. Oxford: Pergamon Press, 1966. World Bank. 2009. World Development Report 2009: Reshaping Economic Geography. Washington, DC: World Bank. CHAPTER 2 Urbanization as a Typology of Space World urban population is expected to increase 84 percent by 2050, rising from 3.4 billion in 2009 to 6.3 billion in 2050. Almost all of this increase will take place in the developing regions. By 2050, the rate of urbanization is expected to reach 66 percent (from 45 percent currently) in the less developed regions and 86 percent (from 75 percent currently) in the more developed regions. The developing countries are experiencing rapid urbanization and are expected to become predominantly urban societies over the coming four decades. The transformation of the urban system has rendered obsolete the distinction between rural and urban areas, and the advances in transportation and telecommunications have facilitated this. Settlement systems have increased in complexity. An example is the change from a monocentric system of cities to a polycentric one, whereby clusters of smaller cities surround a larger one—in effect, a form of territorial specialization. The chapter begins by examining the data on urbanization and projec- tions over the coming four decades, primarily using urbanization data from the United Nations (UN). The remainder of the chapter discusses the main criticism of these data and suggests an alternative measure of urbani- zation that provides a more robust measure of spatial concentration. 15 16 Geography of Growth Urbanization and Space Classifying cities by population size is a comprehensive way of identifying various types of cities. The UN compiles data on urban population and its share of total national population for various countries. The countries report the data to the UN, and as a result, there is no standard definition, which makes cross-country comparisons problematic. The most recent UN publication on urbanization suggests the following: • Half of the world’s 6.7 billion people will live in urban areas by 2010. • Not all of the world’s regions are equally urbanized. • Asia and Africa are the least urbanized regions but account for most of the urban population. • Asia’s urban population, currently 1.6 billion, is expected to double over the coming four decades, adding another 1.8 billion persons by 2050. • China is expected to become 70 percent urban (from 40 percent pres- ently), accounting for 1 billion people by 2050. • India is expected to urbanize the least over the coming four decades. Currently, 30 percent of its population live in urban areas, a rate that is expected to reach 55 percent by 2050, or 900 million people. • Dhaka, Karachi, and Lahore are expected to grow the fastest and will acquire megacity status—cities with more than 10 million inhabit- ants—by 2050. • Africa’s urban population is likely to triple over the next 40 years, increasing from 340 million to more than 900 million. • The fastest-growing cities in Africa are not yet megacities, but Kinshasa and Lagos are expected to surpass 10 million inhabitants by 2050. • Urbanization in Latin America and the developed world will remain largely the same over the coming four decades. • Natural increase accounts for the majority of urban growth, some 60 percent. An exception is China, where increases in urbanization are primarily due to changes in the number of areas considered urban and to migration. Projections to 2050 depend on a continuing decline in the fertility rate in the developing world. If fertility rates continue at their current levels and urbanization continues at the predicted pace, the global urban popu- lation will reach 8.1 billion by 2050 instead of the projected 6.3 billion. Table 2.1 illustrates these points. The world’s urban population is expected to reach 6.3 billion by 2050, with growth coming from the urban areas in the less developed regions. While rural population in the Table 2.1 Population and Average Annual Rate of Change, by Group and Selected Years, 1950–2050 Population (billions) Average annual rate of change (%) Group 1950 1975 2009 2025 2050 1950–75 1975–2009 2009–25 2025–50 Total population World 2.53 4.06 6.83 8.01 9.15 1.89 1.53 1.00 0.53 More developed regions 0.81 1.05 1.23 1.28 1.28 1.02 0.48 0.22 −0.01 Less developed regions 1.72 3.01 5.60 6.73 7.87 2.25 1.82 1.16 0.63 Urban population World 0.73 1.51 3.42 4.54 6.29 2.91 2.40 1.76 1.31 More developed regions 0.43 0.70 0.92 1.01 1.10 1.97 0.82 0.58 0.33 Less developed regions 0.30 0.81 2.50 3.52 5.19 3.96 3.30 2.15 1.55 Rural population World 1.80 2.55 3.41 3.48 2.86 1.39 0.85 0.12 −0.77 More developed regions 0.39 0.35 0.31 0.26 0.18 −0.39 −0.35 −1.01 −1.62 Less developed regions 1.41 2.20 3.10 3.21 2.69 1.77 1.01 0.22 −0.71 Source: UN 2010. 17 18 Geography of Growth more developed regions has been declining for some time, rural popula- tion in the developing regions is expected to continue increasing until 2025, when it will begin to contract. Urban population growth at the global level is slowing down. Between 1950 and 2009, urban population grew at an annual average rate of 2.6 percent, increasing from 0.7 billion to 3.4 billion (UN 2010). Contraction of the rural population and sus- tained increase of the urban population will result in increasing propor- tions of the global population living in urban areas. Classification of Cities The United Nations classifies various types of cities by the size of their population. Cities with less than 1 million people are classified as small cit- ies, and those with greater than 1 million but less than 5 million are classi- fied as medium-size cities. Large cities are those with populations between 5 million and 10 million, and megacities have a population of 10 million or greater. Figure 2.1 shows total urban population by size of city for 1995, 2009, and 2025. Figure 2.1 illustrates the uneven distribution of world population among cities of different sizes. Almost 52 percent of urban occupants live in cities of less than half a million people. These small cities are expected to account for 45 percent of the projected increase in the world’s urban population up to 2025. Small cities with less than half a million people account for 53.2 percent of the urban population in the more developed regions, marginally higher than the share in the less developed regions: 51.4 percent (UN 2010). At the other end of the spectrum are the megacities, often comprising urban agglomerations of “several cities or urban localities that are func- tionally linked� (UN 2010, 6). The largest megacity, Tokyo, the capital of Japan, with an estimated population of 36.5 million in 2009, comprises 87 surrounding cities and towns, including Chiba, Kawasaki, and Yokohama. There were just three megacities in 1975—Mexico City, New York, and Tokyo. By 2009, 21 cities had attained this size, and 8 more are expected to become megacities by 2025, all in developing countries (table 2.2).1 Large cities, or those with a population of between 5 million and 10 million, numbered 32 in 2009 and are expected to number 46 by 2025. They accounted for 6.6 percent of the total population in 2009. Three-quarters of these large cities are “megacities in waiting� and are located in the developing regions (UN 2010, 8). Large cities account for a greater proportion of the urban population in the less developed Urbanization as a Typology of Space 19 Figure 2.1 Total Population, by City Size, 1995, 2009, and 2025 1,593 1,600 1,400 1,200 1,146 total population (millions) 1,004 1,000 914 800 749 684 629 600 534 526 465 469 400 352 321 320 237 225 143 186 200 0 0 0 0 0 00 e or 00 00 00 00 ,0 m 0, 0, 0, 0, 00 10 50 00 00 d ,0 an 0– 1, 5, 10 an 0– 0– 00 00 0– th 00 00 ,0 0, 00 ss 00 10 0, 0, le 0, 50 00 ,0 00 10 1, 5, 1995 2009 2025 Source: UN 2010. Table 2.2 Number of Cities and Percentage of Total Population, by Size of City, 2009 and 2025 Urban population Number of cities (% of total population) Size of city 2009 2025a 2009 2025a Mega 21 29 9.4 10.3 Large 32 46 6.6 7.1 Medium 374 506 21.9 22.1 Small 509 667 10.3 10.3 Source: UN 2010. Note: Megacities = more than 10 million; large cities = between 5 million and 10 million; medium-size cities = between 1 million and 5 million; small cities = between 500,000 and 1 million. a. Projection. regions, 8.8 percent in 2009, than in the more developed regions, 4.9 percent in 2009 (table 2.3). Medium-size cities, with more than 1 million inhabitants but fewer than 5 million, are numerous—374 in 2009, rising to 506 in 2025 (table 2.2). 20 Geography of Growth Table 2.3 Size of Cities, by Region, Number of Inhabitants, and Share of Population, Selected Years, 1975, 2009, 2025 Region and size of Population (millions) Percentage distribution urban settlement 1975 2009 2025 1975 2009 2025 World Total urban area 1,511 3,421 4,536 100.0 100.0 100.0 Mega 53 320 469 3.5 9.4 10.3 Large 109 225 321 7.2 6.6 7.1 Medium 292 749 1,004 19.3 21.9 22.1 Small 157 352 465 10.4 10.3 10.3 Very small 900 1,775 2,277 59.6 51.9 50.2 More developed region Total urban area 698 924 1,014 100.0 100.0 100.0 Mega 42 101 104 6.1 10.9 10.3 Large 50 45 70 7.1 4.9 6.9 Medium 137 202 207 19.6 21.9 20.4 Small 73 84 92 10.5 9.1 9.0 Very small 396 491 541 56.7 53.2 53.4 Less developed region Total urban area 814 2,497 3,522 100.0 100.0 100.0 Mega 11 219 365 1.3 8.8 10.4 Large 60 180 251 7.3 7.2 7.1 Medium 155 546 797 19.1 21.9 22.6 Small 83 268 374 10.3 10.7 10.6 Very small 505 1,284 1,736 62.0 51.4 49.3 Source: UN 2010. Note: Megacities = more than 10 million; large cities = between 5 million and 10 million; medium-size cities = between 1 million and 5 million; small cities = between 500,000 and 1 million; very small-size cities = less than 500,000. They account for 22 percent of the urban population. Medium-size cities in the less developed and more developed regions account for the same percentage of the urban population—21.9 percent in 2009 (table 2.3). The number of small cities, those with a population between 500,000 and 1 million, was 509 in 2009, increasing to 667 by 2025. These smaller cities account for just 10.3 percent of the urban population (table 2.2). Size of Cities across Developing and Developed Regions Across the regions, city size varies tremendously. Roughly 67 percent of the urban population in Europe reside in cities with less than 500,000 inhabitants, while just 8 percent live in cities of 5 million or greater. The urban distribution by city size is similar in Africa—58 percent of urban inhabitants live in smaller cities and 9 percent live in cities of 5 million Urbanization as a Typology of Space 21 or more. Roughly one in five people lives in a large city in Asia, Latin America and the Caribbean, and North America. The proportion of people living in smaller cities is 49 percent in Asia, 48 percent in Latin America and the Caribbean, and 37 percent in North America. Oceania is a special case, as none of its cities is larger than 5 million people and relatively few—38 percent—of its inhabitants live in cities with fewer than half a million people (UN 2010). Table 2.4 shows the rapid urbanization that has taken place since 1950 and is expected to continue. The process began in the more devel- oped regions, which were approximately 30 percent urban in 1920. By 1950, more than half of the population was living in an urban area. North America, Australia, and New Zealand led the group in 1950, with more than 60 percent of the population living in an urban area, while Europe was the least urbanized,2 with more than 50 percent living in an urban area (UN 2010). The ranking is expected to hold until 2050, when more than 90 percent of North America, New Zealand, and Australia and 84 percent of Europe will be urban (UN 2010). Latin America and the Caribbean has a high level of urbanization, surpass- ing Europe in 2009 and expected to increase until 2050, when 89 percent of its inhabitants will reside in urban areas. By contrast, Asia and Africa are mostly rural, with only 42 and 40 percent, respectively, of the population living in urban areas in 2009. These regions are expected to urbanize rapidly over the coming four decades, when 65 and 62 percent of the population, respectively, will reside in urban areas (table 2.4). As a whole, in 2009, Asia was home to more than half of the world’s urban population (figure 2.2). Together with Africa, Asia will experience Table 2.4 Percentage of Population Living in Urban Areas by Region, Selected Years, 1950–2050 Major area 1950 1975 2009 2025 2050 World 28.2 37.2 50.1 56.6 68.7 More developed regions 52.6 66.7 74.9 79.4 86.2 Less developed regions 17.6 27.0 44.6 52.3 65.9 Africa 14.4 25.7 39.6 47.2 61.6 Asia 16.3 24.0 41.7 49.9 64.7 Europe 51.3 65.3 72.5 76.9 84.3 Latin America and the Caribbean 41.4 60.7 79.3 83.8 88.8 North America 63.9 73.8 81.9 85.7 90.1 Oceania 62.0 71.5 70.2 70.8 74.8 Source: UN 2010. 22 Geography of Growth Figure 2.2 Distribution of the World Urban Population, by Region, 1950, 2009, and 2050 60 54 % of world’s urban population 50 50 40 38 31 30 20 20 16 15 13 12 9 9 10 10 86 4 111 0 a ia pe No be nd ica a ric ni As ro rib a a er rth an ea Af Eu Am Oc Ca eric Am tin e th La 1950 2009 2050 Source: UN 2010. a significant increase in its urban population over the coming four decades. By 2050, roughly 54 percent of the global urban population will reside in Asia, and 20 percent will reside in Africa. The world’s urban population is highly concentrated, with roughly 75 percent of the 3.4 billion urban dwellers worldwide living in 25 countries. Urban population in these countries ranges from 31 million in South Africa to 620 million in China. Together, China, India, and North America account for 36 percent of the global urban population. Furthermore, the increase in world urban population is concentrated in a few countries. China and India are expected to account for roughly one-third of the total increase in urban population over the coming four decades, with nine additional countries expected to contribute 26 per- cent of the urban increase. These countries, which will increase their urban population between 15 million and 51 million, are the Democratic Republic of Congo and Nigeria in Africa; Bangladesh, Indonesia, Pakistan, and the Philippines in Asia; Brazil and Mexico in Latin America; and the United States in North America. Table 2.5 shows the urban growth rates for the world regions. The rates in Africa and Asia are of particular note—a projected 3 and 2 percent a year, respectively, between 2009 and 2025. Urbanization as a Typology of Space 23 Table 2.5 Average Annual Rate of Change in Urban Population, 1950–2025 average annual rate of change (%) Major area 1950–75 1975–2009 2009–25 Africa 4.77 3.85 3.14 Asia 3.66 3.24 2.04 Europe 1.81 0.55 0.34 Latin America and the Caribbean 4.17 2.51 1.22 North America 1.96 1.37 1.11 Oceania 2.60 1.44 1.20 Source: UN 2010. Criticism of the Data and Suggested Alternatives The main criticism of the UN data on population is the subjectivity in their compilation. Each participating country reports on the level of urbanization in the country according to its classification system, which may change over time. The national classification system does not always coincide with that used by the UN. Uchida and Nelson (2008) cite sev- eral examples of where and how country-level data on urbanization differ from the official data reported by the UN (table 2.6). Differing country classifications of urbanization make cross-country analyses difficult, if not impossible. For example, Sweden defines urban as settlements of 200 inhabitants, whereas Nigeria and Syria define urban as settlements of 15,000 (Uchida and Nelson 2008). Furthermore, the concept of “urban� is becoming more difficult to classify in an environ- ment where improvements in transportation systems and communica- tions render obsolete the traditional divide between urban and rural. The literature on the form of urbanization has thus moved away from the simple urban-rural dichotomy to embrace a more dynamic form that explores the relationship between urbanization and economic develop- ment (see Cohen 2004, 2006). Dichotomy has been expressed in the core-periphery models, two-region models, dual-economy models, and urban primacy models. Theoretical and empirical studies examine the evolution of spatial concentration with development and the efficiency implications of too much or too little spatial concentration. Focusing on spatial concentration as opposed to a simple dichotomy emphasizes prob- lems in the measurement of urban concentration. Various methods are used to measure urban concentration, including the Hirschman-Herfindahl index of concentration, the Pareto parameter, and primacy.3 All of these measures depend on how a city and an urban 24 Geography of Growth Table 2.6 National and UN Data on Urbanization in Selected Countries % of population living in urban areas National Country and year data UN data Note India, 1991 39 26.0 National data suggest a higher rate of urbanization and include 113 million inhabitants residing in areas with populations of 5,000 and more. Mauritius, 2000 66 42.7 Reclassifying the population residing in settlements of between 5,000 and 20,000 inhabitants suggests a much higher rate of urbanization than reported by the UN. Mexico, 2000 67 74.4 UN data suggest a higher rate of urbanization when settlements of 2,500 are included. Source: Compiled from Uchida and Nelson 2008. area are defined, and, given the vagaries across time and space, the mea- sures may not be systematic or consistent. Uchida and Nelson (2008) suggest an alternative measure of urban concentration, which they term an agglomeration index (AI). Their agglomeration index consists of population density, the size of popula- tion in a “large� urban center, and travel time to that urban center. They define their index for 2000 and note that it “creates a global definition of settlement concentration that could be used to conduct cross-country comparative analyses� (Uchida and Nelson 2008, 2) (figure 2.3). Locations that satisfy all three indicators are included in the agglomera- tion index and are delineated by them. The concept of the AI is, as a result, more fluid and transcends discrete entities (cities, administrative boundaries, and rural and urban space [see Uchida and Nelson 2008]). The AI is not influenced by country definitions of urban. Furthermore, a key advantage of the index is that population counts are not used to calculate it but serve instead to define and locate cities for the purposes of measuring travel time. For this reason, the accuracy of population counts is far less important here than in the primacy measure (Uchida and Nelson 2008). Uchida and Nelson (2008) compare their agglomeration index, which uses a minimum threshold of 150 people per square kilometer for den- sity, a maximum travel time of 60 minutes, and a minimum of 50,000 inhabitants to define a large city, with the UN estimate for the share of urban population in world regions (figure 2.4). Urbanization as a Typology of Space 25 Figure 2.3 Key Indicators of the Agglomeration Index Population size of large city AI Travel time to Population nearest large city density Source: Uchida and Nelson 2008. Figure 2.4 Agglomeration Index and UN Estimates of Urban Population, by Region, 2000 100 urban population as % of total population 80 60 40 20 0 ia a d Af and tin al A nd es ou h- OE b e d ric M Pa an As n gr hig tri a dl cific a Am sia CD a n ps rib a Af ric un e a rth ast pe h Ca ca th i m D ut As n Ce uro co e eri No e E co EC ra So st r ha in n-O nt E e Ea Sa id no b- th La Su AI UN Source: Uchida and Nelson 2008. Note: OECD = Organisation for Economic Co-operation and Development. The agglomeration index defines urban as a minimum threshold of 150 people per square kilometer for density, a maximum of 60 minutes travel time, and a minimum of 50,000 inhabitants. 26 Geography of Growth The index and the UN estimate differ markedly for South Asia, Middle East and North Africa, and Latin America and the Caribbean, where the agglomeration index indicates a higher urban concentration in the first two regions and the UN estimates indicate a higher share in the third region. Uchida and Nelson (2008) suggest that the higher AI share in South Asia is picking up the higher density of population in that region and that Latin America and the Caribbean may not be as heavily urban- ized as originally thought. The results are, however, dependent on the threshold levels chosen, and it is therefore important to have good justification for those levels. Uchida and Nelson (2008) find that the results are quite robust when considering changes in the population density threshold—that is, from 150 in the base case to 300 and 500 people per square kilometer. However, the results change remarkably when considering differences in travel time, from 30 minutes in the base case to 60 and 90 minutes, respectively. According to Uchida and Nelson (2008), with a moderate increase in the threshold (from 60 to 90 minutes), the AI for South Asia rises to 60.8 percent, more than double the UN figure. Similarly, a change in the minimum size for classification as a city from 50,000 in the base case to 100,000 and 500,000, respectively, alters the results for the AI, causing a steep drop relative to the UN estimates (figure 2.5). The agglomeration index is a superior measure when examining issues of urbanization. Its components—density, travel time, population size—differentiate among cities of various sizes and in so doing provide greater information on urban settlements and the impact of, for example, environmental footprints, congestion, and provision of public infrastruc- ture. The UN data alone cannot differentiate between “one city growing ever-larger and numerous small cities sprouting in what was a sparsely populated area� (Uchida and Nelson 2008, 10). While there are data difficulties in compiling the index, the authors note that developments in satellite technology make available a wide variety of data that are cur- rent and accurate, facilitating a better-informed index. Conclusion The chapter examined urbanization data by first focusing on the data collected by the United Nations. These data provide information on past, current, and future trends in urbanization for cities categorized by size and location. Although the UN data are the most comprehensive avail- able, they are often criticized for not comparing cities adequately over Urbanization as a Typology of Space 27 Figure 2.5 Sensitivity to Indicators Used: Example of Minimum Population Size of Large Cities 100 % of cities in size category 80 60 40 20 0 Af aran cif d Af and A d OE bea nd es ou h- Pa n tin ral an gr hig tri e aa rib a a Ea rica ic Eu ica th Am sia n ps un rth ast La nt pe h th Asi Ca ic r Sa m D Ce ro co e er No e E co EC b- st CD in n-O dl Su e id no M 50,000 100,000 500,000 UN Source: Uchida and Nelson 2008. Note: OECD = Organisation for Economic Co-operation and Development. time and place. As urbanization continues, the traditional dichotomy between urban and rural areas is not tenable, yet the official data continue to be organized at these levels. The text considered an alternative mea- sure of spatial concentration—an agglomeration index that takes into account population density, the amount of population in the large urban center, and travel time to that urban center. The index is sensitive to the threshold values chosen for these variables, and timely, up-to-date data availability may be an issue. Despite these caveats, the index has much to recommend it; its main advantage is that it facilitates a consistent cross- country comparison of urbanization and differentiates among different city sizes. Notes 1. Asia is expected to gain five megacities, while Latin America is projected to gain two, and Africa is projected to gain one. 2. Oceania includes developing regions as well as the developed countries of Australia and New Zealand. Oceania includes Melanesia (Fiji, New Caledonia, Papua New Guinea, the Solomon Islands, and Vanuatu), Micronesia (Guam, 28 Geography of Growth Kiribati, the Marshall Islands, the Federated States of Micronesia, Nauru, Northern Mariana Islands, and Palau), and Polynesia (American Samoa, Cook Islands, French Polynesia, Niue, Pitcairn, Samoa, Tokelau, Tonga, Tuvalu, and the Wallis and Futuna Islands). 3. The Hirschman-Herfindahl index of concentration is the sum of the square of the share of population in each city relative to the national urban popula- tion. The Pareto parameter is a measure of how cities decline in size as they move from the largest to the smallest (Uchida and Nelson 2008). Primacy is the share of the population contained in the largest city relative to the national urban population. References Cohen, Barney. 2004. “Urban Growth in Developing Countries: A Review of Current Trends and a Caution Regarding Existing Forecasts.� World Development 32 (1): 23–51. ———. 2006. “Urbanization in Developing Countries: Current Trends, Future Projections, and Key Challenges for Sustainability.� Technology in Science 28 (1): 63–80. Uchida, Hirotsugu, and Andrew Nelson. 2008. “Agglomeration Index: Towards a New Measure of Urban Concentration.� Background Paper for World Development Report 2009: Reshaping Economic Geography. World Bank, Washington, DC. http://siteresources.worldbank.org/ INTWDR2009/Resources/4231006-1204741572978/Hiro1.pdf. United Nations. 2010. “World Urbanization Prospects: The 2009 Revision; Highlights.� Department of Economic and Social Affairs (DESA), Population Division, United Nations, New York, March. CHAPTER 3 Urban Transition and Growth There is a close association between urbanization—a growing share of urban population—and economic growth and development. This associa- tion suggests various typologies that are based on population size, eco- nomic density, and territorial specialization.1 Economic density refers to the economic mass or output generated on a unit of land. It is the defin- ing characteristic of urban settlements. Density is measured by the value added or gross domestic product (GDP) generated per square kilometer of land and is higher the more concentrated these factors—capital and labor—are. Thus, economic density is highly correlated with employment and population density. Economic density is examined here for developing and developed countries as a key part of urban transition and growth. Further aspects of urban transition and growth are the blurring of the rural-urban divide and the unprecedented volume of people moving to urban areas. The second part of the chapter examines regional trends in urban growth for the developing countries and concludes with a discussion of some of the key features of cities in developing regions. 29 30 Geography of Growth Urbanization and Development Urbanization is the pathway to development: no country has developed without the growth of its cities (World Bank 2009). The path to urban- ization is not linear, and cities come in many different sizes. The economic landscape has a similar picture. The World Development Report 2009 con- trasts the economic and population density of Brussels, where an average square kilometer hosts 2,000 workers producing US$350 million of goods and services and population density is 6,000 people per square kilometer, with the agricultural areas of Belgium, where an average square kilometer hosts 7 workers producing US$330,000. In between is a con- tinuum of density. At the head is the primary or leading city, followed by a spectrum embracing secondary cities, small urban centers, towns, and villages. In some countries, the size difference between the leading or primary city and the next city is considerable. Paris has 10 million inhab- itants; Marseille, the next largest city, has just 1.5 million inhabitants. By contrast, the primary city in the United States—New York, with 22 million—is not that different from Los Angeles, with 18 million. Mumbai and New Delhi in India both have 22 million inhabitants (World Bank 2009). In addition to the continuum of density that transcends a simple rural-urban dichotomy, it is possible to identify a symbiosis of places. Cities of different sizes complement one another. The primary city forms the core of a country’s metropolitan area with adjacent cities. Furthermore, secondary cities may act as regional hubs, providing eco- nomic functions for the areas around them—finance, commerce, public health, education, and culture (World Bank 2009). Using an agglomeration index that derives from the close association between the economic and urban landscape, the World Development Report 2009 examines the relationship between city size (a proxy for economic density) and development. The agglomeration index facilitates cross- country comparisons. Using this agglomeration index, the report identifies three main findings for developed and developing countries. • Economic density rises with development. The proportion of inhabit- ants in an urban area rises rapidly as the city or town transforms from an agrarian to an industrial economy. This geographic transformation coincides with the urban area’s move from low to middle income. While urbanization may slow after this, economic density continues to increase, given that services (the next transformation) are even more geographically concentrated than industry.2 Urban Transition and Growth 31 • Rural-urban and within-urban disparities in welfare narrow with devel- opment. Rural-urban gaps in income, living standards, and poverty begin to converge as economies begin to grow and development takes hold. Within-city gaps may take longer to narrow and may kick in only at advanced stages of development. • Neither the pace of urbanization nor its association with economic growth is unprecedented. What is unprecedented is the sheer number of inhabitants being added each year to the urban population in the developing regions (World Bank 2009). Economic Density Rises with Development The urban landscape at early stages of development is likely to consist of small towns and cities that evolved to fulfill various functions, such as a port city or a market town. As industrialization takes hold, urban- ization takes off, with new cities emerging and current cities expanding, leading to a hierarchy of places. As cities multiply and expand, popula- tion and economic density increase. The World Development Report 2009 identifies two transitions that lead to a transformation of the geographic landscape. First, the move from an agrarian to an industrial base coin- cides with urbanization and a transition from a rural to an urban econ- omy. Second, the transition from a manufacturing to a service economy occurs at a much higher level of development and the process of urban- ization is much slower. “In most countries, these transformations hap- pen at the same time, but in different areas� (World Bank 2009, 57). The concentration of economic activity suggests that the richer the area, the more economically dense it is. Using the primary city and an area of 1° longitude by 1° latitude, the World Development Report 2009 shows a historically rapidly rising concentration, followed by a leveling off for primary cities and cities in the densest grid cells (figure 3.1). Dublin, London, Paris, Singapore, and Vienna ranked at the top of the densest cities in the world in 2005, with more than US$200 million in GDP per square kilometer. Tokyo-Kanagawa, New York–New Jersey, Oslo-Akershus-Vestfold, and Vienna-Mödling were the densest grid cells of 1° longitude by 1° latitude, producing GDP per square meter in excess of US$30 million (World Bank 2009). Primary cities in both developing and developed countries account for a disproportionate share of national GDP. For example, Mexico City contrib- uted 30 percent of Mexico’s GDP in 2005 and occupied just 0.1 percent of its land. Similarly, Budapest, Casablanca, Lagos, Nairobi, and Riyadh 32 Figure 3.1 Relationship between GDP and Spatial Concentration a. Primary city b. Area of 1° longitude by 1° latitude Gross product (US$ millions) per km2 90 Gross product (US$ millions) per km2 300 London Singapore 70 GDP per square kilometer 250 GDP per square kilometer Seoul Tokyo-Kanagawa Madrid Dublin 200 50 (US$ millions) (US$ millions) Vienna New York–New Jersey 150 Paris Ontario 30 Oslo-Akershus-Vestfold 100 Vienna-Mödling 10 50 0 –10 0 5 10 15 20 25 30 0 10 20 30 40 50 GDP per capita (constant US$ thousands) GDP per capita (constant PPP US$ thousands) Source: World Bank 2009. Note: PPP = purchasing power parity. Urban Transition and Growth 33 contributed around 20 percent of the country’s GDP, while occupying less than 1 percent of its land. Faster urbanization, which occurs at early stages of development and is now taking place in the developing world, is associated with higher total GDP growth that then levels off (figure 3.2). Furthermore, the geographic concentration of population, GDP, and household consumption rises sharply with development, then levels off (World Bank 2009). Rural-Urban Dichotomies and Development The “bumpy� nature of the economic and urban landscape has implica- tions for economic welfare, living standards, and poverty that become more pronounced as countries develop (World Bank 2009). Rising income inequality accompanies urbanization initially, but then declines as urbanization takes hold. Rural-urban disparities among today’s developed countries have largely disappeared (figure 3.3). However, rural-urban disparities in productivity and income are very much part of the economic landscape in developing countries, which are still in the first phase of urbanization. As occurred for the developed countries before them, the disparities in consumption, social services, and productivity diminish with urbanization. World Bank (2009, 65) notes, “Most developing countries have passed the peak in their rural-urban disparities� and have made considerable progress toward achieving the Millennium Development Goals. Convergence takes place more rapidly when the economy is more urbanized, as has occurred in China, India, and the Philippines (figure 3.4). According to Cohen (2004, 37), “Ease of transportation and commu- nication has blurred the distinction between urban and rural areas.� In some areas of the world, this has led to the emergence of new types of settlement systems that do not belong in the rural or urban classification. An example, from Pacific Asia, is the extended metropolitan areas called desakota. In these settlements, the geography is rural, with much of the land under cultivation, but the income derives from nonagricultural sources;3 for example, inhabitants are employed in nonfarm jobs or com- mute to the city for work. Cohen also notes that the nature of agricultural work in these regions has shifted to higher-value production. In areas that are already highly urbanized, such as countries in Latin America, there is little point in continuing to differentiate human settlement into urban and rural areas. What is needed is an appreciation of the changing spatial context in a predominantly urban environment. Cohen (2004) refers to the experience in Mexico, where a highly monocentric system of cities 34 Figure 3.2 Density and GDP per Capita in Selected Countries, by Phase of Urbanization a. Developing countries b. Developed countries Santiago, 1800–2000 35 35 Athens, 1800–2000 30 30 Vienna, 1800–2000 Lisbon, 1800–2000 In city population to In city population to natural population natural population 25 25 Seoul, 1800–2000 Dublin, 1800–2000 20 20 Sydney, 1800–2000 Budapest, 1850–2000 Toronto, 1800–2000 15 15 Cairo, 1800–2000 Zurich, 1800–2000 10 São Paolo, 1850–2000 10 Brussels, 1800–2000 5 Kuala Lumpur, 1900–2000 5 Warsaw, 1850–2000 0 0 0 5 10 15 0 5 10 15 GDP per capita (constant international US$, thousands) GDP per capita (constant international US$, thousands) Source: World Bank 2009. Figure 3.3 Rural-Urban Disparities in GDP per Capita a. GDP per capita b. Agglomeration index, 2000 3.50 3.50 ratio of urban to rural GDP per capita ratio of urban to rural GDP per capita 3.00 3.00 2.50 2.50 2.00 2.00 1.50 1.50 1.00 1.00 0.50 0.50 0 0 5 10 15 20 25 30 40 50 60 70 80 90 100 GDP per capita (constant 1990 international $, thousands) agglomeration index, 2000 (%) Source: World Bank 2009. 35 36 Geography of Growth Figure 3.4 Rural-Urban Disparities and Density in the Philippines, China, and India, Various Years a. Philippines, 2000 b. China, 1999 and 2006 ratio of urban disposable 4.0 6 to rural net incomes 3.5 ratio of urban to 5 rural incomes 3.0 2.5 4 2.0 3 1.5 2006 2 1999 1.0 0.5 1 0 0 0 20 40 60 80 100 0 20 40 60 80 100 urban share (%) urban population share (%) c. India, 1983 and 1994 disparity in life expectancy, urban-rural ratio (by state) 1.25 1.20 1.15 1983 1.10 1.05 1994 0 0 0.1 0.2 0.3 0.4 0.5 state-specific urban share (%) Source: World Bank 2009. became, over two decades between 1980 and 2000, a polycentric system with nine large metropoles. Greater use of satellites and geo-coding of census and survey data enable different conceptualizations and measure- ments of spatial concentration. One spatial form that has emerged over the past 20–30 years is the city-region. According to Cohen (2004, 38), “The ‘city-region’ can be identified loosely by the extent and nature of economic activity within an extended economic zone surrounding the city proper.� Bangkok is one such example, containing more than 17 mil- lion people and expected to extend more than 200 kilometers from its center by 2010. These city-regions have evolved to house large-scale capital investments, such as airports and manufacturing plants, located on the urban fringe, and the core has become “the command center� for regional or global businesses, offering telecommunications, banking, law, financial management, information services, and management consulting services, for example (Cohen 2004). Income disparity within a city is a phenomenon that many developing cities experience. This disparity is visible in the slums—chronically over- Urban Transition and Growth 37 crowded dwellings of poor quality in underserved areas within cities (World Bank 2009).4 Slums are a part of rapid urbanization—as labor markets expand to fulfill demand from industries and services, labor moves to cities. Underdeveloped land markets are unable to cope with increasing urbanization, and slums emerge. For many slum-dwellers, the proximity to the large city provides opportunities for economic gain. Many residents in the Dharavi slum in Mumbai started their own businesses after the state government provided limited rights over their dwellings and access to water and power (World Bank 2009). Henderson (2010) is less sanguine about the emergence of slums and suggests that the favela- or slum-style development of Latin American cities is a result of local government policy that favors certain areas by restraining inmigration.5 Unprecedented Volumes of Population The take-off in urbanization originated in nineteenth-century Great Britain. The urban share of population in 1800 was 19.2 percent, increas- ing rapidly to 40.0 percent by 1820. Urbanization spread to the new world—Canada and United States—by the mid-nineteenth century. The pace of urbanization in Europe and North America at the end of the nineteenth century was 7.7 percentage points over the 20 years from 1880 to 1900. This was not unlike the pace of urbanization for the devel- oping countries between 1985 and 2005, which experienced median and mean absolute increases of 7.1 and 8.0 percentage points, respectively. There are two ways in which the urbanization story for the developing world today differs from the experience of the developed world during the mid-to-late nineteenth century. The first is the sheer size of popula- tion. As one example, China added 225 million people to its towns and cities between 1985 and 2002, almost the entire population of the United States (World Bank 2009). Over the 10 years beginning in 1985, the developing world experienced an increase in its urban population of 8.3 million, close to three times the increase in population witnessed by many countries in Europe and North America in the final two decades of the nineteenth century (World Bank 2009). Furthermore, the megacities in the developing world are unprecedented in their size and number (figure 3.5). The second way in which the developing countries are expe- riencing a different pattern of urbanization is in public services. Cities in developing countries today are benefiting from advances in public health and medicine and improvements in water systems. Cities in developed countries at similar stages of urbanization in the nineteenth century had 38 Geography of Growth Figure 3.5 Change in Urban Population with and without China and India, 1985–2005 a. All countries including China and India 140,000 change in urban population 120,000 (thousands), 1985–2005 100,000 80,000 United States, 1880–1900 60,000 average for developing countries, 1985–2005 40,000 average for developing countries, 1880–1900 20,000 0 countries b. All countries excluding China and India 30,000 United States, 1880–1900 25,000 change in urban population (thousands), 1985–2005 20,000 15,000 average for developing countries 10,000 (excluding China and India)1985–2005 average for developing countries, 1880–1900 5,000 0 countries Source: World Bank 2009. poorer health and lower life expectancy than rural areas (World Bank 2009). Urbanization in Developing Countries Cohen (2004) summarizes the major differences in urbanization experi- enced by the developing regions and identifies some of the challenges in the coming decades. The discussion is summarized in table 3.1. Enormous interregional and intraregional differences are evident in the pattern of Table 3.1 Regional Differences in Urbanization Rate of urbanization (%) Region 1950 2000 Characteristics of urbanization Latin America and 42 75 Urban primacy; many cities with more than 1 million people; reverse polarization; spatial polarization the Caribbean between rich and poor; long tradition of urbanization in some countries;a most Caribbean countries have a high rate of urbanization, with Haiti as an exception. South Asia 18 27 Population becoming increasingly urban; extreme poverty and depravation, creating enormous urban management challenges; region is home to 5 of the world’s 30 largest cities: Mumbai, Kolkata, and Dehli (India), Dhaka (Bangladesh), and Karachi (Pakistan); improved modes of transportation have extended the reach of urban areas and blurred the distinction between rural and urban—desakota; the majority of land is under cultivation, but nonfarm jobs are an important source of employment and income; region will be home to three of the world’s five largest urban agglomerations;b most urban growth in the future will take place in smaller cities and towns. East Asia and — 38 Areas that have integrated into the global economy experienced rapid urban transformation that is now the Pacific being repeated in the newly industrializing economies;c special economic zonesd along China’s coastal regions were a catalyst for industrialization and urbanization; population projections suggest that the 1.25 billion additional people by 2030 will be absorbed into urban areas and that 54% of the population will be urban by 2030. Former Soviet — — Historically, government determined the nature, scale, and spatial distribution of economic activities; cities republics were planned in concentric circles with large industrial plants and large-scale housing estates; the end of the Cold War and collapse of the Soviet Union had huge economic, social, and demographic consequences that were most apparent in the cities. Middle East and 27 58 Less diversity in urbanization among countries in the region compared to other regions; the need for access North Africa to water, rapid industrialization, and high levels of international labor migration to oil-rich Gulf states resulted in urban population of more than 50% and more than 85% in many countries;e socioeconomic and political heterogeneity within the region results in a wide variety of urban problems and challenges. 39 (continued next page) 40 Table 3.1 (continued) Rate of urbanization (%) Region 1950 2000 Characteristics of urbanization Sub-Saharan Africa 15 38 Least developed and least urbanized region of the world; most cities are small by international standards; Kinshasa and Lagos are exceptions; colonialism influenced the structure and pattern of economic growth— cities displaced traditional networks of trade and influence and attracted migrants; colonial urbanization also affected the physical structure and layout of many cities into two highly uneven zones—a European space with a high level of infrastructure and an indigenous space with marginal services; postcolonial cities grew quickly as a result of high population growth and high spatial mobility; cities are economically marginalized in the new global economy;f challenges for urban authorities are to provide low-income housing, high-quality urban services, and employment.g Source: Compiled from Cohen 2004. Note: — = Not available. a. Argentina, Chile, and Uruguay. b. Delhi, Dhaka, and Mumbai. c. Indonesia, Malaysia, Philippines, and, the Thailand. d. For examp,le, Shantou, Shenzhen, Xiamen, and Zhuhai, which were established as testing grounds for a more open, export-oriented development strategy. e. Bahrain, Kuwait, Lebanon, Libya, Qatar, and Saudi Arabia. The Republic of Yemen is an exception, with just 25 percent of the population classified as urban. f. “Since the 1970s, urban growth in Africa has been most affected by the region’s economic crisis. A current list of ailments includes declining productivity in agriculture and industry, a lack of foreign exchange, increasing indebtedness, worsening balance-of-payments position, and declining real wages� (Cohen 2004, 45). g. In addition to economic mismanagement, several countries have suffered from long civil wars, with large numbers moving to cities. Cities are growing despite poor macroeconomic performance and without significant foreign direct investment. Urban Transition and Growth 41 urbanization from the 1950s to the present. In each case, the socioeco- nomic and political history interacts with the geographic landscape to determine the urbanization experience and is affected by globalization, democratization, and decentralization. As each region’s and, indeed, each country’s socioeconomic, political, and geographic landscape is different, so too has been their path of urbanization. Latin America and the Caribbean is primarily an urban region with a long history, at least in some countries. The overall rate of urbanization masks intraregional differences, whereby Argentina, Brazil, Chile, French Guiana, Mexico, Uruguay, the República Bolivariana de Venezuela, and several Caribbean countries or territories (Anguilla, The Bahamas, Cuba, Guadeloupe, Martinique, Puerto Rico, Trinidad and Tobago) are more than three-quarters urban, while the Dominican Republic, Guatemala, Guyana, and Jamaica are around 50 percent urban. Haiti has the lowest rate of urbanization in the region, at around 30 percent. Large-scale rural- to-urban migration, coupled with industrial policies that targeted cities and were predicated on import substitution and protection of infant industries, led to a high rate of urban primacy. Cities with more than 1 million people in the region increased from 6 in 1950 to more than 50 in 2000, and “the four largest cities—Buenos Aires, Mexico City, Rio de Janeiro, and São Paulo—have grown to previously unimaginable sizes� (Cohen 2004, 40). Rural-to-urban migration is expected to continue. However, industrial policy has shifted to favoring areas outside of the megacities. This shift, coupled with congestion costs, economic recession, and programs of structural adjustment in the 1990s, has adversely affected many Latin American cities and the megacities in particular. Urbanization will continue, but at a slower pace. At the other end of the spectrum is Sub-Saharan Africa, one of the least urbanized regions and where urbanization has been largely predi- cated on demographic factors and not always for benign reasons (see the notes to table 3.1). The region’s colonial past affected the manner in which the physical aspects of urban space developed, incorporating two zones—a European zone “enjoying a high level of urban infrastructure and services and an indigenous space that was marginally serviced� (Cohen 2004, 45). Data problems make it difficult to describe urban trends. And while fertility rates are falling, population momentum sug- gests a fast pace of urbanization, such that “before 2023, African society will become predominantly urban� (Cohen 2004, 45). The Middle East and North Africa region is home to some of the world’s oldest cities, yet urbanization was slow to take off, with just 42 Geography of Growth 27 percent of the population classified as urban in 1950. By 2000, there had been an eightfold increase in urban inhabitants. Cairo, Istanbul, and Tehran are among the largest urban agglomerations in the world, home to more than 7 million inhabitants each. Most countries within the region are at least 50 percent urban (Cohen 2004). The challenges for urban planners are many, given the socioeconomic and political heterogeneity within the region. For example, rural-to-urban migration and population growth in the Arab Republic of Egypt have led to many slums and hous- ing shortages in Cairo. Urban issues in postconflict Iraq “have to do with establishing the infrastructure of urban government and other issues of rehabilitation and reconstruction� (Cohen 2004, 44). China and India dominate the East Asia and the Pacific region, and the combined urban population in the region contains just under half of the world’s urban population. The vastness of the region makes generaliza- tions difficult. Cohen (2004) classifies countries based on their experi- ence with economic development and urbanization (table 3.1). Some countries have opened up to the world economy, benefiting from global- ization and experiencing increasing rates of urbanization. Many of the coastal cities have undergone rapid urban transformation as a result. The designation of special economic zones in China also led to rapid urban transformation in cities such as Dalian, Guangzhou, Qingdao, Shenzhen, Tianjin, and Xiamen. The pace of urban change in South Asia has been relatively modest due largely to the more rural nature of the countries. Significant urban challenges abound in an environment with extreme poverty and inade- quate physical infrastructure. Increasing industrialization has benefited urbanization, as have improved modes of transportation that have extended the reach of the urban areas, blurring the distinction between urban and rural areas. Nonfarm jobs have become a feature of the desakota zones around the cities, contributing positively to employment and income. Urbanization in the former Soviet republics was dictated by govern- ment decisions relating to the nature, scale, and spatial distribution of economic activities. The absence of a land market in cities led to a large amount of unused land throughout the city. Large-scale industrial pro- duction was favored over the service or retail sectors. Moreover, many of these industrial production plants were kept in operation long after they had ceased to be profitable. The fallout once the Soviet Union collapsed was felt most strongly in the cities, with enormous declines in output, “rapid impoverishment of large sections of society, great uncertainty Urban Transition and Growth 43 about the future, and a fundamental reevaluation of the location, func- tioning, and organization of productive activity� (Cohen 2004, 43). The social, economic, and demographic consequences were also unprece- dented, with declining rates of marriage, birth, and male life expectancy. Key Features of Cities in Developing Countries Duranton (2008) provides a theoretical framework in which he examines the key features of cities in developing countries, first from a static effi- ciency perspective and then from a dynamic growth perspective. He aug- ments the theoretical discussion with examples from the empirical literature, where relevant and where available. The literature on cities in developing economies, although much thin- ner in volume, is nevertheless consistent with the more voluminous lit- erature on cities in developed economies in supporting an upward-sloping wage curve, costs of living that rise with city size, a bell-shape net wage curve, and some labor mobility driven by net wage differentials (Duranton 2008).6 Various studies on agglomeration economies identify both local- ization effects (agglomeration effects take place within sectors) and urbanization effects (agglomeration effects take place between sectors) for cities in developing countries (table 3.2).7 Localization economies foster specialized cities, while urbanization economies foster diversified cities. However, cities in developing countries are rarely specialized, espe- cially when compared with cities in the developed countries. Despite some caveats with regard to the data, the evidence suggests the existence of both diversified and specialized cities in the developing regions.8 Turning to the labor supply curve and mobility, the literature suggests that internal migration flows in developing countries are consistent with an upward-sloping labor supply curve but that mobility is not perfect (Henderson 2008). Brueckner (1990) and Ravallion and Wodon (1999) find that the direction of migration flows is consistent with existing dif- ferences in net wages. Henderson (2008) concludes that movements in the labor supply curve are driven largely by what is happening in the countryside. For example, Barrios, Bertinelli, and Strobl (2006) note that climate change that affects living standards in rural areas gives rise to mobility to the cities. Fay and Opal (1999) suggest that worsening rural conditions that lower the labor supply curve lead to urbanization without growth. These “negative correlations may explain why many developing countries attempt to restrain urbanization,� yet Henderson finds “scant evidence about cities being too large in general� (Henderson 2008, 18, 19). 44 Geography of Growth Table 3.2 A Dozen Economies of Scale Type of economy Static or of scale dynamic Example Internal 1. Pecuniary Firms are able to purchase intermediate inputs at volume discounts. Technological 2. Static Static Average costs fall because of fixed costs of operating a technological plant. 3. Dynamic Dynamic Firms learn to operate a plant more efficiently over time. technological External or agglomeration Localization 4. “Shopping� Static Shoppers are attracted to places where there are many sellers. 5. “Adam Smith� Static Outsourcing allows both the upstream input suppliers specialization and downstream firms to profit from productivity gains because of specialization. 6. “Marshall� labor Static Workers with industry-specific skills are attracted to a pooling location where there is greater concentration. 7. “Marshall-Arrow- Dynamic Reductions in costs arise from repeated and continuous Romer� learning production activity over time and spill over between by doing firms in the same place. Urbanization 8. “Jane Jacobs� Static The more different things are done locally, the more innovation opportunity there is for observing and adapting ideas from others. 9. “Marshall� labor Static Workers in an industry bring innovations to firms in other pooling industries; this is similar to number 6, but the benefit arises from the existence of diverse industries in one location. 10. “Adam Smith� Static Similar to number 5 above, the main difference being that division of labor the division of labor is made possible by the existence of many different buying industries in the same place. 11. “Romer� Dynamic The larger the market, the higher the profit; the more endogenous attractive the location to firms, the more jobs there are, growth the more labor pools there are, the larger the market, and so on. 12. “Pure� Fixed costs of infrastructure are spread over more agglomeration taxpayers; diseconomies arise from congestion and pollution. Source: World Bank 2009. Urban Transition and Growth 45 He refers to the strong barriers to labor mobility that have made Chinese cities too small, as examined in Au and Henderson (2006a, 2006b), but other examples are difficult to find. Henderson (2008) questions whether restrictions on labor mobility, either because of a misguided application of the Harris and Todaro (1970) model9 or because of a policy of restric- tions, may be part of a political-economy equilibrium and therefore difficult to change. What is evident in developing economies is a large preponderance of urban primacy, where the largest city is often disproportionately bigger than the second largest city. Henderson (2008) disagrees with the trade policy arguments that have been used to explain urban primacy on the grounds of theoretical ambiguity and available empirical evidence.10 He suggests that urban primacy is due to political and institutional factors. These factors are difficult to quantify in terms of their magnitude and direction. Ades and Glaeser (1995) and Davis and Henderson (2003) suggest a positive relationship between urban primacy and unstable and undemocratic regimes. Henderson (2008) suggests that the many regula- tions and permits that govern economic activity in developing countries may also favor urban primacy, as those nearer to the center of power may find it easier to obtain permits or circumvent regulations. The political economy associated with urban primacy may be very difficult to break. Henderson (2008) suggests that administrative deregulation, as in the Republic of Korea, may be an effective tool to limit primacy, but at the end of the day, the issue appears to be one of political economy. What one also finds in some cities in developing economies is the existence of a dual housing sector. This is manifest by a division between formal housing and squatter settlements. Henderson (2008, 29) notes that in “some large developing country cities, more than half the popula- tion live in squatter settlements.� In examining the reasons for squatter settlements, he concludes the following: • If squatter settlements were simply a matter of some of the poor opting out of formal housing, policy choices would revolve around issues of redistribution and the most effective way to do this. • Squatter settlements arise because of a lack of effective, formal prop- erty titles (which also have knock-on effects for enterprise develop- ment and female labor supply). • Squatter settlements may be the outcome of policy distortions—for example, binding minimum lot size and rent controls. 46 Geography of Growth • Squatter settlements may not be so cheap if the inhabitants have to pay high prices for necessities such as water, for example, or, in the absence of titles, have to pay for some form of protection (Henderson 2008). Based on these points, Henderson concludes that regulatory constraints in the formal housing sector, while limiting the size of the city, will be crowded out by the growth of squatter settlements. Removing regulatory constraints is socially desirable. Titling policies are desirable, and these should be grounded in the legal and taxation system. There is a political economy of squatter settlements, with many vested interests that benefit from this undesirable status quo (Henderson 2008). In summary, cities in developing countries would benefit from (a) eliminating primary city favoritism, (b) solving the biases that lead to squatter settlements by implementing a titling policy, and (c) not discouraging the internal migration of people. These policy suggestions also hold when examining the features of cities from a dynamic growth perspective. However, pursuing such policies is demanding,11 and the political economy of these issues may be difficult to address (Henderson 2008). Conclusion Urbanization—the movement of people into cities—has implications for the economic development of the land area that this population will occupy or inhabit. The chapter examined several features of urbanization associated with economic development. Economic density—the output produced per unit of land—is a principal feature of urban transition and growth. It rises with urbanization and becomes ever more concentrated as the structure of output changes from manufacturing to services. The economic landscape is bumpy, with some areas exhibiting higher rates of economic density. This has implications for economic welfare, living stan- dards, and poverty. Initially, inequality rises with economic density and urbanization, but over time, convergence occurs between the rural and urban areas, blurring the dichotomy between them. Urbanization in the developing world differs from that in the developed. First, it differs in its history: the volume of people moving into cities in the developing world today is unprecedented. Furthermore, advances in security, health, and sanitation suggest that the experience is different than what occurred in the developed world when urbanization was beginning to take place. The Urban Transition and Growth 47 chapter also examined regional differences in urbanization for the devel- oping countries and noted some of the facets of urbanization—primacy, squatter settlements, discouragement of city growth—that characterize many developing cities today. Notes 1. The population size of cities is examined in chapter 2. Chapter 4 examines territorial specialization of cities. 2. One may argue that developments in transportation and in information and communication technology dispute this claim. However, the raison d’être of cities and urban areas is the face-to-face communication they facilitate. 3. Remittances from family members working in the city also constitute income to the deskota (Cohen 2004). 4. Examples include Dharavi in Mumbai, Kibra and Huruma in Nairobi, Washington in Abidjan, Majboor Nagar and Kanchan Puri in Delhi, San Fernando in Buenos Aires, and Rocinha in Rio de Janeiro (World Bank 2009). 5. China provides an example. An explicit policy limits migration to certain key cities by making living conditions unpleasant for migrants. This supersedes a less explicit policy that used the hukou (the system of household registration) to constrain rural people to live in rural areas but take industrial jobs and controlled migrants to ensure that urbanization was localized and diffuse (Henderson 2010). 6. Empirical studies on the cost-of-living and bell-shape net wage curve for developing economy cities are scant and not examined here. 7. Duranton (2008) cites Henderson (1988) for localization economies in Brazil; Henderson, Lee, and Lee (2001) for localization economies for Korean indus- tries, particularly traditional industries; Lall, Shalizi, and Deichmann (2004) for India; and Deichmann et al. (2005) for Indonesia. Henderson (2009) also finds urbanization economies for advanced sectors in Korea. Furthermore, Lall, Funderburg, and Yepes (2004) find evidence, albeit weak, of urbaniza- tion economies in India, while Lall, Koo, and Chakravorty (2003) find stron- ger evidence for India. Deichmann et al. (2005) find mild evidence of urbanization effects for several sectors in Indonesia, and Au and Henderson (2006a, 2006b) find both urbanization and localization effects for several cit- ies in China. Henderson (2005) and Overman and Venables (2005) review the literature on agglomeration economies in developing countries. 8. Duranton (2008, 17) suggests two main criticisms. First, the presence of agglomeration economies may reflect the sorting of “more productive work- ers in bigger and more specialized cities, rather than true agglomeration 48 Geography of Growth economies.� Second, the data represent the formal sector. Case study evi- dence suggests that the informal sector is a strong contributor to agglomera- tion economies. 9. The argument for dual labor markets “was first developed by Harris and Todaro (1970) and has been extremely influential in policy circles.� Workers in the formal sector are paid higher wages in urban than in rural areas. “The initial earnings gap between the rural and formal urban sector causes workers to move to the city� (Henderson 2008, 27). 10. Trade liberalization does reduce urban primacy in the model of Krugman and Livas Elizondo (1996), in which all cities are able to import differentiated goods. This equalization of market potential reduces the tendency for agglomeration of manufacturing in a single core city. Henderson (2008) questions the assumption of equalization and suggests that trade liberaliza- tion is more likely to benefit cities in coastal regions or cities close to trading partners, thus reinforcing their dominance. Empirical studies find weak evi- dence for urban primacy and are beleaguered by other variables that are correlated with trade. According to Henderson (2008, 20), “Ades and Glaeser (1995); Nitsch (2006) suggest that trade plays no systematic role with respect to urban primacy.� 11. Henderson (2008, 44) writes, “Such an agenda is rather demanding since it includes raising the efficiency of public good provision, lowering barriers to mobility, improving market access to allow secondary cities to develop, and so forth.� References Ades, Alberto F., and Edward L. Glaeser. 1995. “Trade and Circuses: Explaining Urban Giants.� Quarterly Journal of Economics 110 (1): 195–227. Au, Chun-Chung, and J. Vernon Henderson. 2006a. “Are Chinese Cities Too Small?� Review of Economic Studies 73 (3): 549–76. ———. 2006b. “How Migration Restrictions Limit Agglomeration and Productivity in China.� Journal of Development Economics 80 (2): 350–88. Barrios, Salvador, Luisito Bertinelli, and Eric Strobl. 2006. “Climatic Change and Rural Urban Migration: The Case of Sub-Saharan Africa.� Journal of Urban Economics 60 (3): 357–71. Brueckner, Jan K. 1990. “Analyzing Third World Urbanization: A Model with Empirical Evidence.� Economic Development and Cultural Change 38 (3): 587–610. Cohen, Barney. 2004. “Urban Growth in Developing Countries: A Review of Current Trends and a Caution Regarding Existing Forecasts.� World Development 32 (1): 23–51. Urban Transition and Growth 49 Davis, James C., and J. Vernon Henderson. 2003. “Evidence on the Political Economy of the Urbanization Process.� Journal of Urban Economics 53 (1): 98–125. Deichmann, Uwe, Kai Kaiser, Somik V. Lall, and Zmarak Shalizi. 2005. “Agglomeration, Transport, and Regional Development in Indonesia.� Policy Research Working Paper 3477, World Bank, Washington, DC. Duranton, Gilles. 2008. “Cities: Engines of Growth and Prosperity for Developing Countries?� Working Paper 12, World Bank, Washington, DC, on behalf of the Commission on Growth and Development. Fay, Marianne, and Charlotte Opal. 1999. “Urbanization without Growth: A Not- So-Uncommon Phenomenon.� Policy Research Working Paper 2412, World Bank, Washington, DC. Harris, John R., and Michael P. Todaro. 1970. “Migration, Unemployment, and Development: A Two-Sector Analysis.� American Economic Review 60 (1): 126–42. Henderson, J. Vernon. 1988. Urban Development: Theory, Fact, and Illusion. Oxford: Oxford University Press. ———. 2005. “Urbanization and Growth.� In Handbook of Economic Growth, vol. 1B, ed. Philippe Aghion and Steven N. Durlauf, 1543–91. Amsterdam: North- Holland. ———. 2008. “Cities: Engines of Growth and Prosperity for Developing Countries?� Commission on Growth and Development Working Paper 12, commissioned by the World Bank, Washington, DC. ———. 2009. “Urbanization in China: Policy Issues and Options.� Report pre- pared for CERAP (China Economic Research and Advisory Programme). http://www.cairncrossfund.org/download/ /Background%20 Papers/Henderson%20-%20Final_Report_2009.11.10%5BUrbanization%5D. pdf. ———. 2010. “Cities and Development.� Journal of Regional Science 50 (1): 515–40. Henderson, J. V., T. Lee, and Y. J. Lee. 2001. “Scale Externalities in Korea.� Journal of Urban Economics 49 (3): 479–504. Krugman, Paul, and Raul Livas Elizondo. 1996. “Trade Policy and the Third World Metropolis.� Journal of Development Economics 49 (1): 137–50. Lall, Somik V., Richard Funderburg, and Tito Yepes. 2004.“Location, Concentration, and Performance of Economic Activity in Brazil.� Policy Research Working Paper 3268, World Bank, Washington, DC. Lall, Somik V., Jun Koo, and Sanjoy Chakravorty. 2003. “Diversity Matters: The Economic Geography of Industry Location in India.� Policy Research Working Paper 3072, World Bank, Washington, DC. 50 Geography of Growth Lall, Somik V., Zmarak Shalizi, and Uwe Deichmann. 2004. “Agglomeration Economies and Productivity in Indian Industry.� Journal of Development Economics 73 (3): 643–73. Nitsch, Volker. 2006. “Trade Openness and Urban Concentration: New Evidence.� Journal of Economic Integration 21 (2): 340–62. Overman, Henry G., and Anthony J. Venables. 2005. “Cities in the Developing World.� Report for the Department for International Development, London. Ravallion, Martin, and Quenton Wodon. 1999. “Poor Areas, or Only Poor People?� Journal of Regional Science 39 (4): 689–711. World Bank. 2009. World Development Report 2009: Reshaping Economic Geography. Washington, DC: World Bank. CHAPTER 4 Spatial Concentration and Specialization The chapter presents a further typology of space based on today’s specialization of cities. Urban specialization arises from the trade-off between scale economies—internal and localized external economies that include knowledge economies1 arising from the education level of city inhabitants—and diseconomies in own industry and in living (hous- ing costs including rents, crime, overcrowding, land use).2 Industries that benefit from local agglomeration economies as well as internal economies are more likely to reside in larger cities. Industries that depend on their own activity for productivity improvements are likely to locate in small, specialized cities where own industry economies of scale are maximized (Henderson 2010).3 Cities in developing countries tend to host many specializations. Territorial specialization has not yet taken place, as urbanization is still progressing. Furthermore, “The important inheritance of colonially cre- ated port cities, the economic necessity to concentrate the first major efforts in infrastructure, the availability of skilled labour in a situation of great skills scarcities, the dependence upon imported manufacturing inputs and services� are suggestive of why developing cities lack territo- rial specialization (Harris 1991, 26). The largest cities are the most accessible to foreign investors and are “incubators for new firms trying 51 52 Geography of Growth to discover the best product lines and production methods� (Henderson 2010, 524). Cities in developed economies are highly specialized, given their advanced stage of economic development and history of urbanization. Small and medium-size cities in Japan, the Republic of Korea, the United States, and other countries are highly specialized and have been for some time (Henderson 2010). When heavy manufacturing was a major part of the economy, the typology of cities included cities producing steel, tex- tiles, automobiles, ships, aircraft, pulp and paper, and petrochemicals. Henderson (1974), using data from the United States, estimates that between 50 and 60 percent of the urban labor force lies in the nontraded sector (wholesale, retail, personal services, construction, utilities), with the remaining share engaged in the traded sector.4 In current times, small cities have become specialized in consumer service activities such as retirement, health, and insurance services (Henderson 2010). Very little research has yet been carried out on the scale economies arising from service activities. Specialization of Cities Cities have specialized in certain activities since the fourteenth century, and early writers in urban hierarchy have classified cities based on their engage- ment in primary industries, secondary industries, and tertiary industries (MacKenzie 1933). Later, Ogburn (1937) classified seven types of cities based on their industrial and economic activity. These included (a) trading cities, (b) factory cities, (c) transportation cities, (d) mining cities, (e) plea- sure cities, (f) health resort cities, and (g) college cities. Duncan and Reiss (1956) classified cities based on (a) regional location, (b) economic activity, (c) economic specialty, and (d) population size and growth rate. The authors noted the interdependence of cities in the urban hierarchy. Table 4.1 shows the spatial allocation of manufacturing and business services in U.S. cities for 1910 and 1995. In 1910, manufacturing was part of the urban landscape. As urbanization and development continued, manufacturing activity was replaced by business services. Henderson (2010) cites three principal reasons cities become inefficient for stan- dardized manufacturing: 1. Firms and industries as a whole have accomplished much learning and adoption of foreign technologies and no longer benefit so much from the learning environment of the largest cities. Spatial Concentration and Specialization 53 Table 4.1 Manufacturing and Business Services in the United States, by Size of City, 1910 and 1995 % of employment Share of Share of Share of employment in employment in employment in Metro area population manufacturing business services consumer services 1910 Four largest MSAsa 35.1 6.2 18.5 Medium-size MSAsb 35.3 5.1 18.2 Small MSAsc 30.9 4.6 20.1 Nonmetro area 25.1 4.4 24.3 National average 30.2 5.0 21.2 1995 Over 2.5 million 14.3 21.3 23.3 1 million to 2.5 million 15.2 19.3 22.8 500,000 to 1 million 15.8 17.7 23.6 250,000 to 500,000 19.1 15.5 23.2 Under 250,000 18.8 13.3 24.6 Nonmetro area 26.9 9.1 22.6 National average 17.2 17.7 23.2 Source: Holko 1999. Note: MSA = metropolitan statistical area. a. Refers to the largest MSAs (private sector employment over 600,000: New York, Chicago, Boston, Philadelphia). b. Private sector employment between 100,000 and 600,000 c. Private sector employment under 100,000. 2. Cities become very expensive locations, with high rents and labor costs. Infrastructure and skilled labor are in greater relative abundance in other locations. 3. The business service sector is expanding, demanding locations inside large cities and outbidding manufacturing for central city lands in them (Henderson 2010). Henderson (2010) posits that territorial specialization is associated with a certain population size. Business and professional services suggest a large city by population size. Similarly, high-tech industries that undergo substantial technological progress locate in larger cities, where they also benefit from local agglomeration—for example, the aircraft industry in Los Angeles or the research and development (R&D) seg- ment of the electronics industry in Tokyo (Henderson 2010). Cities like London and New York are global financial service cities, with a very small manufacturing sector (table 4.2 gives the example of New York). Strong institutions in the economic and legal environment contribute to 54 Geography of Growth Table 4.2 Share of New York County (Manhattan) in Total Private Employment in the United States, 1997 Sector % of private employment All industries 1.8 Headquarters 3.0 Financial headquarters 11.7 Financial services 12.0 Security brokers 25.0 Business services 7.5 Advertising 15.0 Source: Henderson 2010. the development and sustainability of these global cities, which are also cultural cities and attract a creative class of worker. In between the large cities and small and medium-size cities are metro areas that are more diverse. Theoretical modeling and empirical work suggest that the share of service activity increases in line with the size of the metro area (Henderson 2010). Cities in developed countries that have made the transition from manufacturing to services and have become specialized in various service activities represent a new typology of cities. Among this typology are knowledge cities, creative cities, global cities, and green (eco) cities. The following sections examine this typology. Knowledge Cities The comparative advantage of developed economies lies in their knowl- edge base. The generation and application of the knowledge base define an area’s competitiveness and growth at the local, regional, and national lev- els.5 Knowledge may take the form of investment in R&D, a qualified and skilled labor force, high-quality entrepreneurship, or all three. Knowledge enhances productivity through innovation in products, services, and pro- cesses (Lever 2002). Numerous writers attest to the association between the urban knowledge base and economic growth and development (Knight 1995; Kresl and Singh 1999; Lambooy 2000; Lever 2002). As the knowledge base of the city increases, the nature of development changes. Knowledge-based services (financial services, insurance, and communica- tions) outgrew knowledge-based manufacturing industries (high-technol- ogy and medium-technology industries) for the Organisation for Economic Co-operation and Development (OECD) countries as a whole between 1985 and 1998 (Lever 2002). Spatial Concentration and Specialization 55 Critical to the explanation of economic growth and development is the distinction between tacit and codified knowledge. Codified knowl- edge is available to all businesses and at low or zero cost. It does not impart any competitive advantage. An example is the Internet. Tacit knowledge is available to only limited contacts and is often passed through face-to-face communication. It does confer competitive advan- tage. Large cities, where inhabitants are in contact with one another, benefit from tacit knowledge. Lever (2002) examines a multidimensional index of the scale of the knowledge base for European cities. This study presents a wider defini- tion of the knowledge base and is a departure from other studies that define it as the number of R&D establishments per million inhabitants or workers (Lever 2002). Three dimensions of the knowledge base are con- sidered: tacit knowledge, codified knowledge, and knowledge infrastruc- ture. Knowledge infrastructure is captured by telecommunications infrastructure (table 4.3). Using the variables listed in table 4.3, Lever (2002) develops a general index comprising seven measures: presence of corporate producer service companies in the knowledge sectors (finance, law, marketing, research), connectivity of the local airport, the hosting of commercial conferences and exhibitions, the rate of new enterprise formation, two variables for the size of the local universities (number of students and number of pub- lished academic research papers in science, computing, medicine, and technology), and the quality of local telecommunications infrastructure (Lever 2002). The index is then applied to 19 European cities that scored on at least four of the seven measures,6 and the mean rankings are calcu- lated. The results are shown in table 4.4. London and Paris emerge as “world cities� and are ranked in the top three positions in all but one of the seven variables (university size). National capitals fare well as knowledge-based cities, and many other cit- ies in Germany also fare well on the knowledge base score. This may reflect the decentralized system of government there. Furthermore, the knowledge base in London and Paris is so heavily concentrated in the finance, law, and administrative sectors that no other cities qualify for inclusion. The mean score increases as the rankings fall, and Lever (2002) suggests that this may reflect the peripheral location of these cities from the center of the European Union. London and Paris fare less well across the measures of economic success: annual percentage employment change, annual percentage change in gross value added per worker, and the shift-share residual, which standardizes for industrial structure at the Table 4.3 Dimensions of Knowledge Base: Measures and Results 56 Dimensions of knowledge base and measures Data source Results Tacit knowledge (Rank of ) leading world cities in four Globalization and World Cities Research Alpha cities (London, Milan, Paris) service sectors (accountancy, advertising, Group (Taylor and Walker 2001) Beta cities (Brussels, Madrid, Moscow, Zurich) banking and finance, and law) Gamma cities (Amsterdam, Barcelona, Berlin, Budapest, Copenhagen, Dusseldorf, Geneva, Hamburg, Munich, Prague, Rome, Stockholm, Warsaw) Airport connectivity (number of Buursink (1994) Amsterdam, Brussels, Copenhagen, Dusseldorf, Frankfurt, connections and change in number over Geneva, London, Munich, Paris, Vienna, Zurich a period, 1991–93) Fairs and trade exhibitions (composite Regression study that defined the economic Barcelona, Birmingham, Bologna, Cologne, Dusseldorf, index based on local population size, advantage to holding fairs and trade Frankfurt, Hanover, London, Madrid, Milan, Munich, Paris rental levels, local per capita income, exhibitions (Rubalcaba-Bermejo and infrastructure, transport, and weather) Cuadrado-Roura 1995) New enterprise formation Registered new businesses (not cited) Not cited Codified knowledge Number of students in local universities Local universities (not cited) Not cited Volume of academic and scientific papers Total number of published papers; number Berlin, Cambridge, Copenhagen, Edinburgh-Glasgow, in refereed journals of papers per 1,000 inhabitantsa London, Madrid, Manchester-Liverpool, Moscow, (Matthiessen and Schwartz 1999) Oxford-Reading, Paris, Randstadt, Stockholm Knowledge infrastructure Quality of telecommunications provision Rankings based on technical definitions of Amsterdam, Berlin, Brussels, Frankfurt, London, Madrid, the pricing of services, the choice of Milan, Paris, Stockholm, Zurich physical infrastructure available, and the availability of the most advanced and sophisticated connections (Finnie 1998) Source: Compiled from Lever 2002. a. Number of papers per 1,000 inhabitants yielded the following rank: Cambridge, Oxford-Reading, Geneva, Basel, Bristol-Cardiff, Zurich, Stockholm, Helsinki, Copenhagen, Randstadt, Munich, Edinburgh-Glasgow. Spatial Concentration and Specialization 57 Table 4.4 The Knowledge Base and Economic Performance in Selected Cities Annual % Change in gross Knowledge employment value added Shift-share base mean change, per worker (%), residual, City score 1985–96 1985–96 1978–96 London 2.8 −0.3 2.0 −25.7 Paris 3.7 −0.1 3.3 −11.8 Frankfurt 6.0 0.2 3.1 6.5 Amsterdam 6.8 2.5 2.0 21.0 Stockholm 7.0 1.0 3.5 8.7 Milan 7.0 0.1 3.0 4.7 Cologne 7.2 0.3 2.7 4.8 Bologna 7.5 0.0 2.9 0.6 Zurich 8.0 1.7 3.3 3.0 Brussels 8.3 −0.8 2.8 −22.1 Madrid 10.0 1.5 4.1 5.5 Munich 10.0 −0.1 3.5 10.6 Copenhagen 10.0 0.2 1.6 −7.1 Dusseldorf 10.3 0.2 2.5 0.6 Barcelona 10.8 1.6 3.3 1.6 Geneva 12.0 1.7 3.0 1.0 Berlin 12.3 1.7 3.5 1.2 Rome 13.3 0.4 2.9 10.5 Vienna 14.5 0.5 2.3 −23.3 Source: Lever 2002. start of the period. This may reflect the fact that these cities do not com- pete with smaller cities once their reliance on financial services, law, and administration is taken into account, and the urban agglomeration disec- onomies of high rents, high living costs, and congestion outweigh the knowledge base of these cities. Gabe et al. (2010) argue for a broader interpretation of knowledge that would combine educational achievement and skills from various occupations. Classifying knowledge in this manner identifies clusters of U.S. and Canadian metropolitan areas by similar knowledge traits. Their study focuses on six specific groupings: making regions, which are char- acterized by manufacturing activity; thinking regions, characterized by knowledge about the arts, humanities, information technology(IT), and commerce; comforting regions, with a high knowledge about mental health; building regions, with a high knowledge about construction and transportation; innovating regions, with a very high knowledge about information technology, arts, commerce, and engineering; and working 58 Geography of Growth regions, characterized by low knowledge in information technology and commerce. Using data on occupations in an area provides a broader measure of human capital than using just college attainment, because this approach captures the skills and knowledge acquired in the work- place. The authors find that this broader knowledge variable identifies clusters and is a better predictor of regional economic development than college attainment. Furthermore, in a fixed-effects regression, engineer- ing, enterprising, and building regions have higher levels of productivity and earnings per capita. Teaching, understanding, working, and comfort- ing regions have lower levels of economic development. Education Level and City Growth A significant part of the work on knowledge cities focuses on the relation- ship between the education level of a city’s inhabitants and its growth. Education is an important ingredient in local agglomeration economies, and cities, by their nature, speed the accumulation of human capital.7 Cities with an educated population grow faster than cities where inhabitants have less education. Glaeser and Saiz (2004) find this state- ment to be true for more than a century of data in the United Kingdom and the United States. More recently, in the two decades prior to 2000, the population of metro areas in which more than 25 percent of adults held college degrees grew 45 percent. By contrast, the population in metro areas in which less than 10 percent held a college degree grew just 13 percent. In a similar vein, Shapiro (2006) finds that, during the period 1940 to 1990, a 10 percent increase in a metropolitan area’s concentration of human capital is associated with an increase in that area’s employment of 0.8 percent. Shapiro (2006, 324) refers to a “sub- stantial body of literature that confirms this correlation between human capital and local area employment (or population) growth.�8 Furthermore, Glaeser, Ponzetto, and Tobio (2011) examine the relationship between education and city growth and find that, as the share of the adult popu- lation increased 5 percent in 1970, predicted growth between 1970 and 2000 increased about 8 percent. Table 4.5 examines the regression results from equation 4.1: log (Y2000/Y1970) = B * Schooling1970 + Other Controls, (4.1) where Y denotes one of three outcome variables: population, median income, and self-reported housing values. Other controls refers to the initial values of population, median income, and housing values and three region dummies (the Midwest is omitted). The equation permits the Table 4.5 Metropolitan Area Regressions Log change in population, Log change in median income Log change in median housing 1970–2000 (2000 US$), 1970–2000 value (2000 US$), 1970–2000 Indicator (1) (2) (3) (4) (5) (6) Log population, 1970 −0.007 −0.007 0.003 0.003 0.056 0.057 (0.019) (0.019) (0.006) (0.006) (0.013)** (0.013)** Log median income in −0.769 −0.841 −0.391 −0.403 −0.297 −0.328 2000 and 1970 (0.191)** (0.191)** (0.061)** (0.062)** (0.133)* (0.135)* Log median housing 0.273 0.272 0.173 0.174 −0.008 0.004 value in 2000 and 1970 (0.117)* (0.115)* (0.037)** (0.038)** (0.081) (0.082) South dummy 0.146 −0.133 −0.010 0.015 0.028 0.012 (0.054)** (0.122) (0.018) (0.040) (0.038) (0.087) East dummy −0.054 −0.077 −0.044 −0.039 0.054 0.041 (0.057) (0.158) (0.016)** (0.052) (0.040) (0.112) West dummy 0.384 0.632 0.797 −0.145 0.299 0.052 (0.051)** (0.135)** (0.142)** (0.044) (0.035)** (0.096) College completion 1.528 0.797 0.802 among population (0.445)** (0.142)** (0.310)* 25 and older 1970 (continued next page) 59 60 Table 4.5 (continued) Log change in population, Log change in median income Log change in median housing 1970–2000 (2000 US$), 1970–2000 value (2000 US$), 1970–2000 Indicator (1) (2) (3) (4) (5) (6) South dummy * % BA 3.840 0.673 0.405 in 1970 (0.772)** (0.252)** (0.548) East dummy * % BA 1.498 0.839 0.424 in 1970 (1.310) (0.428) (0.929) West dummy * % BA −0.573 1.364 2.257 in 1970 (0.812) (0.265)** (0.576)** Midwest dummy * % 1.314 0.583 0.363 BA in 1970 (0.597)* (0.195)** (0.424) Constant 5.264 6.074 2.275 2.407 2.81 3.040 (1.576)** (1.595)** (0.504)** (0.521)** (1.100)* (1.132)** Number of observations 257 257 257 257 257 257 R2 0.427 0.466 0.379 0.396 0.339 0.362 Source: Glaeser, Ponzetto, and Tobio 2011. Note: Standard errors in parentheses. * = significant at 5 percent, ** = significant at 1 percent. Spatial Concentration and Specialization 61 impact of education to be estimated separately by region (South, East, West, and Midwest) and is thus a departure from the approach of other studies of education and city growth. B interacts with four region dum- mies and thereby allows the impact of schooling on population, income, and housing value growth to differ by region. The second regression in table 4.5 allows the impact of education in 1970 to differ by region. In the South, which shows the strongest effect, “a 5 percent increase in the share of the adult population with a college degree in 1970 is associated with 19 percent faster population growth� (Glaeser, Ponzetto, and Tobio 2011, 28). The results for the Midwest are also significant, and a 5.0 percent increase in the share of adults with a college degree in 1970 is associated with a 6.5 percent predicted increase in population. The Northeast shows the second largest coefficient of the group and is similar to the coefficient for the national average, but it is insignificant. The coefficient for the West is negative and insignificant. Regressions 3 and 4 in table 4.5 examine the effect of median growth in income. Glaeser, Ponzetto, and Tobio (2011) note that mean income reverts, except perhaps in areas with high housing values, which may indicate a migration of wealthier people to areas with more amenities. Coefficients on the regional dummies, apart from the West, where income increased the least, are statistically insignificant. There is a strong association between median income and initial education levels: “As the share of population with college degrees in 1970 increased by 5 percent, median income rose by 4 percent more since then� (Glaesser, Ponzetto, and Tobio 2011, 28). Moreover, this may reflect a return to skills and the tendency of highly educated people to move to areas already rich in human capital (see also Moretti 2003). Education has a positive impact on median income growth at the regional level (regression 4). The biggest impact is in the West (0.7 log points increase for a 5 percent increase in the share of those with a college degree), and the least impact is in the Midwest (less than half that found in the West). Regressions 5 and 6 examine the impact of education on the appre- ciation of median housing values. The West saw much greater apprecia- tion in housing values compared to the other regions. In total, housing values increased about 4 percent more when the share of the population in 1970 with a college degree increased by 5 percentage points (Glaeser, Ponzetto, and Tobio 2011). Finally, turning to regression 6, the results indicate a much larger appreciation in the West: prices increased by more than 10 percent for a 5 percentage point increase in the 1970 share of 62 Geography of Growth population with a college degree. The results for the other regions are far lower and statistically insignificant. Further studies examine the relationship between a city’s education level and other economic variables, such as migration, wages, and sectoral employment. These studies are representative of the empirical work on knowledge cities that deal with omitted-variable bias.9 Table 4.6 lists these variables and the studies that have used them. The results from the empirical studies in table 4.6 show the positive impact of knowledge, measured by college education, on growth of the city or metropolitan statistical area (MSA). Controlling for other growth- inducing variables raises this positive impact at the level of the MSA. Glaeser and Saiz (2004) find little impact at the level of the city and attribute this to the high level of service employment or better weather at this spatial level. However, college education is a more powerful pre- dictor of growth at the MSA level, becoming even stronger when other control variables are included. Moretti (2003) refers to the long-run trend of increasing education in the United States. The features of this are a wide dispersion of human capital among cities (between 1990 and 2000, the fraction of college graduates rose from about 10 percent in the least educated cities to above 40 percent in the most educated cities [Moretti 2003]) and an increasing stock of college graduates (cities with a larger stock of human capital in 1990 experienced larger increases over the next decade). For the most part, empirical studies of the knowledge city have focused on college education. Glaeser and Saiz (2004) also include the high school dropout rate as an alternative measure. This alternative mea- sure is a stronger (negative) correlate of education at the level of the city compared to the MSA—the correlation between the share of high school dropouts and population growth is −28 percent for cities and −18 per- cent for metropolitan areas. The correlations suggest that the impact of higher education may be more important at the MSA level, whereas the impact of low education is more important at the city level. Controlling for high school dropout rates and unemployment rates10 at the city level significantly reduces the impact of higher education on city growth. Avoiding low human capital11 matters more for smaller units of geogra- phy. Shapiro (2006) finds no evidence to indicate a growth effect for high school graduates. The empirical studies also attempt to identify the connection between skills and growth. For the most part, productivity accounts for this connection, but some authors (Glaeser and Saiz 2004; Shapiro 2006) Table 4.6 Underpinnings of Knowledge Cities Source and dependent variable Independent variable Observations Results Shapiro (2006) Growth in Initial employment; log % prime- 495 metropolitan areas, A 10% increase in share of college-educated residents is employment age white males with a college 1940–90 associated with an 0.8% increase in employment. degree Growth in wages Initial wages; log % prime-age white 495 metropolitan areas, A 10% increase in share of college-educated residents is males with a college degree 1940–90 associated with a 0.2% increase in wages. Growth in rental price Initial rental price; log % prime-age 495 metropolitan areas, A 10% increase in share of college-educated residents is white males with a college degree 1940–90 associated with a 0.7% increase in rental price. Growth in house value Initial house value; log % prime-age 495 metropolitan areas, A 10% increase in share of college-educated residents is white males with college degree 1940–90 associated with a 0.7% increase in house value. Moretti (2003) Change in percent of Initial level of college; population; 237 metropolitan areas, Overall fraction of college graduates grew faster in cities college educated family income; black*; Hispanic; 1990–2000 that were larger and richer in 1990; the percentage of immigrants*; agriculture; Hispanics is negatively correlated with changes in college manufacturing*; high tech; share; the percentage of agricultural jobs is negatively Northeast; Midwest; South; West correlated with changes in college share; the percentage of high-tech jobs is a strong predictor of change in college share; change in college share was 3.7% (northeastern cities), 3.6% (midwestern), 3.2% (western), and 2.8% (southern). Winters (2008) ln in-migration Share with a bachelor’s degree 323 PMSAs/MSAs, A 10% increase in share with a bachelor’s degree increases 1995–2000 inmigration by 5%. ln out-migration Share with a bachelor’s degree 323 PMSAs/MSAs, A 10% increase in share with a bachelor’s degree increases 1995–2000 outmigration by 3%. 63 (continued next page) Table 4.6 (continued) 64 Source and dependent variable Independent variable Observations Results ln net migration Share with a bachelor’s degree 323 PMSAs/MSAs, A 10% increase in share with a bachelor’s degree increases 1995–2000 net migration by 2%. ln in-migration, ln out- Share with bachelor’s degree; 323 PMSAs/MSAs, Adding controls raises the coefficients on share with a migration, ln net population; median family income; 1995–2000 bachelor’s degree; population and median family income migration manufacturing share; January have a negative effect on inmigration, outmigration, and temperature; July temperature; net migration; increases in the average January daily low precipitation; Midwest; South; temperature increase both the inmigration rate and the West net migration rate; increases in average July daily temperature increases net migration. Glaeser and Saiz (2004) Difference in ln Share of population with a college 723 cities For the MSA regressions, a 1% increase in share of adult population between degree 318 MSAs, 1970, 1980, population with a degree increases the decadal growth census years 1990, 2000 rate by approximately 0.5%. For the city-level regressions, a 1% increase in the share of adult population with a degree increases the decadal growth rate by approximately 0.2%. Difference in ln Share of population with a college 723 citiesa Controlling for the listed independent variables shows little population between degree; initial level of population; 318 MSA, effect on future city growth but does increase the impact census years Ln average heating days; Ln 1970, 1980, 1990, 2000 of the education variable on the future growth rate of the average precipitation; share of MSA. workers in manufacturing; share of workers in professional services; share of workers in trade; unemployment rate; share of high school dropouts Source: Glaeser and Salt 2004; Moretti 2003; Shapiro 2006; Winters 2008. Note: PMSA = primary metropolitan statistical area; MSA = standard metropolitan statistical area. * = results are insignificant and not reported in the results column. a. Cities with a population of more than 30,000 in 1970. Spatial Concentration and Specialization 65 also suggest consumption or amenity factors. Glaeser and Saiz (2004) suggest that movement in wages and house prices sheds light on the productivity and consumption story. For example, increases in nominal wages and house prices stem from production-led growth, while decreases in real wages stem from consumption-led growth. However, Shapiro (2006, 330) finds that controlling for wages and rents implies that one-third of the employment growth effect stems from “rapid improvement in the quality of life.� The quality-of-life explanation oper- ates through consumer amenities, such as bars and restaurants, rather than from improvements in crime, schools, or pollution (Shapiro 2006). The focus on quality of life and consumption-led growth has led to research into the concept of the creative city, discussed below. In fact, the creative city is a refinement of the knowledge city. The Creative City The concept of the creative city is a late twentieth-century construct. It gives a spatial context to creativity—the creative pursuits of individuals and industries—and suggests economic development potential. The creative city has been viewed as a home for the creative class (Florida 2002), as an engine of structural change, as a catalyst for economic revi- talization, as a facilitator of public and private partnership, and as an urban success story. The creative city typology can be applied to both large and small cities, dependent on a number of factors, which are dis- cussed further below. The Creative Class Florida (2002, 18) asserts that the creative class, “a fast-growing, highly educated, and well-paid segment of the workforce on whose efforts corpo- rate profits and economic growth increasingly depend,� is critical for eco- nomic growth. Roughly 38.3 million Americans or 30 percent of the U.S. workforce occupy the creative class and hold significant economic power. The average salary of a creative class worker in 1990 was US$48,752 com- pared to almost US$28,000 for a working-class worker and US$22,000 for a service class worker (Florida 2002). The economic effects of creativity depend on Florida’s so-called three Ts: talent, tolerance, and technology: • Talent, or creative share of the workforce, based largely on demographic, educational, and occupational characteristics 66 Geography of Growth • Tolerance, or diversity, based on indexes related to sexual orientation and bohemianism • Technology, or innovation, measured by patent activity and the high- technology share of the economic base. Florida (2002) develops a creativity index, which is a mix of four equally weighted factors: the creative class share of the workforce, high- tech industry (using the Milken Institute’s tech pole index), innovation (using patents per capita), and diversity (using the gay index). This index forms a baseline view of an area’s (city or region) position in the creative economy, which Florida asserts is suggestive of a “region’s longer run eco- nomic potential� (Florida 2002, 22). Florida computes the index for large cities (a ranking of 49 metro areas reporting populations over 1 million in the 2000 census), medium-size cities (a ranking of 32 metro areas report- ing populations 500,000 to 1 million in the 2000 census), and small cities (a ranking of 63 metro areas reporting populations 250,000 to 500,000 in the 2000 census). Table 4.7 shows these results for the top and bottom three cities in each size category. Each dimension—talent, tolerance, technology—is necessary to attract the creative class of worker and generate economic growth. The creative class is involved in wide-ranging occupations from arts and entertainment to high-technology, finance, and high-end manufactur- ing occupations. Among these occupational groups, Florida (2002) identi- fies three types of creative individuals. First is a core group, the “super-creative core� who exhibit an entrepreneurial spirit in “producing new forms or designs that are readily transferable.� Markusen (2006b) also identifies core cultural workers and the high rate of self-employment among this group—45 percent compared with 8 percent of the work- force as a whole. Acs and Megyesi (2007) also identify a strong entrepre- neurial element in the creative city. The second group includes the creative professionals who work in knowledge-intensive industries and possess a high level of formal education. Markusen (2006b) notes the high level of investment in human capital by artists and cultural workers. Zucker (1994) notes that artists are extraordinary citizens who have high rates of political and community participation. The third group repre- sents workers who transcend the old distinctions between white-collar and blue-collar work. As an example, Florida (2002, 19) asserts that, as today’s technicians, secretaries take on “increased responsibility to inter- pret their work and make decisions.� The result is an increase in creativity and a swelling of the creative class. Spatial Concentration and Specialization 67 Table 4.7 Creativity Rankings in the United States, by City Size Creativity Creative Creative High-tech Innovation Diversity City rank and size index workers(%) rank rank rank rank Top three cities Large cities San Francisco, CA 1,057 34.8 5 1 2 1 Austin, TX 1,028 36.4 4 11 3 16 San Diego, CA 1,015 32.1 15 12 7 3 Medium cities Albuquerque, NM 965 32.2 2 1 7 1 Albany, NY 932 33.7 1 12 2 4 Tucson, AZ 853 28.4 17 2 6 5 Small cities Madison, WI 925 32.8 6 16 4 9 Des Moines, IA 862 32.1 8 2 16 20 Santa Barbara, CA 856 28.3 19 8 8 7 Bottom three cities Large cities Memphis, TN 530 24.8 47 48 42 41 Norfolk, VA 555 28.4 36 35 49 47 Las Vegas, NV 561 18.5 49 42 47 5 Medium cities Youngstown, OH 253 23.8 32 32 24 32 Scranton, PA 400 24.7 28 23 23 31 McAllen, TX 451 27.8 18 31 32 9 Small cities Shreveport, LA 233 22.1 55 32 59 57 Ocala, FL 263 16.4 63 61 52 24 Visalia, CA 289 22.9 52 63 60 11 Source: Florida 2002. Given the occupational profile of the creative class worker, it is not surprising that regions with high growth, centers of learning, and exper- tise attract the creative class. The creative class accounts for more than one-third of the workforce in the Washington, DC, area; the Raleigh- Durham area; Boston; and Austin. Florida (2002) identifies a similar proportion of creative class worker in the college towns of East Lansing, MI, and Madison, WI. Comunian and Faggian (2011) investigate the relationship between creative cities and creative universities in the United Kingdom. Not all regions benefit from what Florida asserts is a “new geography of class.� In fact, Florida (2002) notes that inequality may increase in a creative city, where well-paid, highly educated people push out an 68 Geography of Growth older population who can no longer afford to live in the areas they once inhabited. Why do some geographic areas fail to generate a creative core? In answering this question, Florida first considers why some places become destinations for the creative class. The creative class of workers values visible diversity—different food, music, people, varied nightlife, indige- nous street-level culture, and outdoor recreation—as well as authenticity of place that combines urban grit alongside renovated buildings, a co- mingling of young and old, of people as well as place. Examples of such places that have successfully combined high-tech industry, outdoor ame- nities, lifestyle amenities, creativity, and innovation (the three Ts) are the greater Boston area (Route 128 suburban complex, Harvard University, Massachusetts Institute of Technology, and several charming inner-city Boston neighborhoods); the Seattle area (suburban Bellevue and Redmond, beautiful mountains and countryside, revitalized urban neighborhoods); the San Francisco Bay area (posh inner-city neighborhoods and ultra-hip districts like SoMa—South of Market—lifestyle enclaves like Marin County, as well as the Silicon Valley); and Austin (traditional high-tech developments, lifestyle centers for cycling and outdoor activities, and a revitalizing university-downtown community centered on vibrant Sixth Street, the warehouse district, and the music scene) (Florida 2002). Failure to adapt to the “demands of the creative age� has much to do with areas trapped by their past successes (Florida 2002, 24). Olson (1982) suggests that areas that fail to transition are experiencing an “insti- tutional sclerosis.� This translates into being trapped in the “culture and attitudes of the bygone organizational age, unable or unwilling to adapt to current trends� (Florida 2002, 24). Glaeser (2011) notes that Boston has reinvented itself at least three times since the 1970s, whereas cities like Detroit and Cleveland have failed to transition to what Florida terms the “creative age.� Reinvention and Structural Change Acs and Megyesi (2007) argue that diverse areas have lower entry barri- ers, making it easier for human capital with various backgrounds to enter an area and stay there. They also associate entrepreneurship with creativ- ity and note that it is more apt to flourish in areas rich in creativity. In the same vein, Cohendet, Grandadam, and Simon (2010) refer to the cre- ative city as a place for ideas to flourish and take shape, ultimately result- ing in economic growth and wealth. Liu-Wei and Yin-Ko (2010) link the creative city to an urban environment capable of generating creativity, Spatial Concentration and Specialization 69 innovation, and, thus, income growth. Markusen (2006b, 1) suggests that the creative city “heralded a new revitalization strategy for older indus- trial cities� and that urban and economic development planners of com- munities of all sizes have increasingly turned to arts and culture as development tools. As an example, Acs and Megyesi (2007) combine data on the core group and creative professionals for Baltimore and other industrial areas (table 4.8). The benchmarked areas show a huge increase in population over the decade 1990 to 2000. Romein and Trip (2010) suggest that structural change arising from several forces—economic (globalization and an economy built on ser- vices), geopolitical (vanishing national borders and the rise of regions as engines of growth), technological (improved information and communi- cation technology and transport), and sociocultural (consumption, amenities)—herald the creative city. It is not just occupations that are labeled creative; entire branches of industries are also termed cultural, such as the arts, performances, heritage-based products, and creatively designed products. Creative people are therefore the most crucial resource for the economic performance of a creative city. On the basis of questionnaires and regressions, Florida (2002, 223) asserts, “Regional eco- nomic growth is driven by the locational choices of creative people—the holders of creative capital—who prefer places that are diverse, tolerant, and open to new ideas.� Florida’s work has been criticized for not being new and for ignoring “the productive dimensions of the cultural indus- tries� (Pratt 2008, 2).12 Peck (2005) lambastes Florida’s promotion of the creative class, castigating it as an elitist place-marketing ploy. While Florida highlights one critical part of the creative city and contributes Table 4.8 Creative Class Occupations, Ranked by Percentage Change Target statistical area % change, 1990–2000 Chicago, IL (PMSA) 169 Cleveland, OH (PMSA) 151 Pittsburgh, PA (PMSA) 139 Baltimore, MD (MSA) 126 Philadelphia, PA–NJ (PMSA) 123 Milwaukee, WI (PMSA) 123 St. Louis, MO–IL (MSA) 117 Detroit, MI (PMSA) 108 Source: Acs and Megyesi 2007. Note: PMSA = primary metropolitan statistical area; MSA = metropolitan statistical area. 70 Geography of Growth enormously to this part, his approach does not necessarily exclude the productive base of the economy. Romein and Trip (2010) differentiate between “innovation production milieus� and “urban consumption milieus,� while noting that it is the close association of both that ensures the success of the creative city. The for- mer focuses on innovative ideas and processes from inception to market realization across clusters of firms, not all of which are creative, but all of which benefit from the close proximity that an urban environment pro- vides. The urban consumption milieu focuses on the qualities of place and life in a city that makes individuals want to move and stay there. In a sense, capital (investments and jobs) follows creative labor. In noting the tendency of artists to gravitate to inner-city areas, Markusen (2006a) sug- gests a revitalizing role for areas that may have lost population. Romein and Trip (2010) identify key elements of success for creative cities—social climate; representation; labor market and employment; buzz and atmo- sphere; built environment; living and residential environment; amenities; clusters and incubators; and policy, government, and governance. These elements show how difficult it is to disentangle the production and con- sumption bases of the creative city. Pratt (2008) suggests that policy makers would achieve more successful regeneration outcomes if they would view the cultural industries as an object that links production and consumption, manufacturing and services. This is a more useful approach in interpreting and understanding the significant role of cultural produc- tion in contemporary cities and how it relates to growth. One of the reasons why smaller towns or cities may be more successful at fostering creativity and generating economic growth may stem from their consumption base. The focus on occupations emphasizes the human capital aspect of economic growth, an aspect that has generated huge cur- rency among growth theorists, development planners, and policy makers. The cultural sector is of particular relevance here, given that artists as core cultural workers make considerable investments in human capital, move easily across commercial, nonprofit, and community sectors, and have exceptionally high rates of self-employment (table 4.9). Markusen (2006a) applies to the cultural sector a consumption-base theory of economic growth—the portion of local economic activity that is sold to local resi- dents and acts as a growth catalyst. Residents are assumed to spend on local cultural products that benefit the resident creative class and the local economy. Creative class workers are then assumed to spend their incomes locally, generating a positive growth multiplier for the local economy. Glaeser and Saiz (2004) find that amenities or consumption spending is Spatial Concentration and Specialization 71 Table 4.9 Creative Workers: Consumers and Producers Total Self- % self- Secondary Occupational title employment employed employed Primary job job Writers and authors 138,900 94,377 68 80,509 13,868 Visual artists 307,254 155,159 50 129,109 26,050 Artists and related workers 148,682 80,022 54 70,731 9,291 Arts directors 50,664 27,139 54 23,988 3,151 Fine artists, painters, sculptors, illustrators 23,192 12,866 55 11,372 1,494 Multimedia artists and animators 74,826 40,017 53 35,371 4,646 Photographers 130,442 65,432 52 54,024 14,408 Camera operators, TV, video, motion picture 28,130 6,705 24 4,354 2,351 Performing artists 176,463 42,724 24 38,174 4,550 Actors 63,033 10,992 17 9754 1,238 Producers and directors 76,125 24,995 33 21,683 3,312 Dancers and choreographers 37,305 6,737 18 6,737 0 Dancers 19,992 3,854 19 3,854 0 Choreographers 17,313 2,883 17 2,883 0 Musicians, singers, composers 215,425 83,121 39 56,770 26,351 Music directors and composers 54,271 21,354 39 14,584 6,770 Musicians and singers 161,154 61,767 38 42,186 19,581 Designers 531,921 168,806 32 132,827 35,979 Commercial and industrial designers 51,823 16,088 31 12,659 3,429 Fashion designers 14,844 4,353 29 3,425 928 Floral designers 103,993 33,832 33 26,621 7,211 Graphic designers 211,871 67,422 32 53,052 14,370 Interior designers 60,050 19,325 32 15,206 4,119 Merchandise displayers, window trimmers 77,221 23,881 31 18,791 5,090 Set and exhibit designers 12,119 3,905 32 3,073 832 Architects 136,378 29,678 22 23,809 5,869 Architects, except landscape and naval 113,243 24,253 21 19,457 4,796 Landscape architects 23,135 5,425 23 4,352 1,073 Total, all artistic occupations 1,506,421 573,865 38 461,198 112,667 Total, all occupations 144,013,600 11,451,600 8 9,926,000 1,525,600 Source: Markusen 2006. 72 Geography of Growth more likely to catalyze growth in cities than in metropolitan areas, which rely more on increases in productivity for their economic growth. Harnessing the Potential of Creative Cities The United Nations (UN 2004) outlines several ways in which the poten- tial of the creative city might be unleashed to benefit developing and developed economies: • Urban regeneration through culture. The concept of the creative city has been tested in response to the economic decline of industrial cities in Australia, Europe, and the United States over the past two decades. These experiences have shown that industries in fields such as televi- sion, cinema, multimedia, music, books, and festivals can flourish in cities that provide efficient transport, communications, and public protection infrastructure combined with coordinated public policies that encourage innovation and small businesses in the creative fields. • Public-private partnership as a key to effective policy. Planners take into account the role of creativity during economic policy planning in order to integrate their tangible and intangible cultural assets into the educa- tion systems, natural environment, and geographic location. Cities across the developed world are establishing municipal services to sus- tain the local creative economy, facilitating cooperation between the private and public sectors as well as civil society; some have even gone so far as to develop creativity indexes based on the three Ts of technol- ogy, talent, and tolerance. • Creativity as an unexploited opportunity. Discussion of creativity remains at the academic level or policy level. Planners and the general public are unaware of or underestimate the value of creativity for the community; political or artistic figures do not champion the role of culture; administrative resources, skills, and capacities to manage such projects are in short supply; or clear and usable indicators do not exist for measuring success. • The need for UNESCO’s Creative Cities Programme. Designed to pro- mote the social and economic development of cities in both the devel- oped and the developing world, the program will emphasize the role of creativity and the arts and create a platform for information exchange between cities. Spatial Concentration and Specialization 73 • Impact far beyond the economy. Creative cities programs have already been tried and tested on a limited scale and have proved to be innova- tive in finding new ways to promote social and economic development by stimulating the creation of new enterprises and cultural diversity for both struggling as well as prosperous city communities. The Global City The “global city�13 is a term that was popularized by Sassen in 1991 with her book of that title (Sassen 1991). According to Sassen (2010), the global city makes new norms. In order for this to happen, the city must be complex and diverse. These factors are often a function of size, but not all large cities or megacities are global cities. For example, Tokyo is a global city, but Mumbai or São Paulo, both megacities, are not necessarily global. Sassen attests that many of the global cities of today are old-world cities that have reinvented themselves, citing Istanbul, London, and New York as examples. In contrast, Miami is a global city, combining complexity and diversity and making new norms, but it is not a megacity. This was not the case prior to the 1990s. Since then, several factors have coalesced to make Miami a global city: the infrastructure of international trade developed by Miami’s Cuban population; real estate development spurred by wealthy individuals from Latin America; the opening up of Latin America and setting up of regional headquarters in Miami by firms from all over the world; a mix of cultures in a small, concentrated space; and a burgeoning creative class (Sassen 2010). Sassen (2009) identifies key structural trends in the economy that are contributing to the rise of global cities worldwide. These trends are predicated on a growing demand for intermediate services—for example, insurance, accounting, legal, financial, consulting, software programming, and even traditional sectors. These services tend to be located in an urban- ized environment where tacit and codified knowledge are maximized. As firms become less local in their operations, expanding into national, regional, and global markets, management operations become more com- plex, and the firm is likely to outsource its corporate functions that pre- viously had been managed in-house. Advances in technology have fostered outsourcing of many routinized sectors, but control remains at the center, and, with it, centralized headquarter functions have grown, facilitated by the development and growth of the intermediate sector. Cities house the headquarters.14 74 Geography of Growth The specialized firms that emerge to fulfill these intermediate functions are themselves subject to agglomeration economies. These arise from the “mix of firms, talents, and expertise from a broad range of specialized fields� (Sassen 2009, 56). The agglomeration economies rely on the exchange of information, and the urban environment is key in providing the face-to-face communication and exchange of knowledge opportunities. Sassen (2009, 59) notes, “Cities can generate kinds of knowledge, both formal and informal, that go beyond the sum of recognized knowledge actors (e.g. professionals and professional firms in the case of the economy),� which she terms “urban knowledge capital.� It has the same features as the tacit and codified knowledge identified by Lever (2002). A key aspect of cities is their centrality, which relies on density. Density has typically been associated with a downtown or central business dis- trict. However, while centrality remains critical in today’s global economy, the geography of this has extended to include other spatial forms, such as the city-region, for example, or indeed the global city-region.15 The Global Cities Indicators Program (GCIP), created in 2006 by the World Bank with funding from the government of Japan, helps member cities to monitor their performance.16 The GCIP provides a framework for the collection of city indicators that are comparable and consistent over time and place. Each member city is responsible for updating its data on the web portal.17 This effort enables “cities to measure, report, and improve their performance and quality of life, facilitate capacity building, and share best practices� (Bhada and Hoornweg 2009, 1). The Global City Indicators Facility (GCIF) at the University of Toronto18 took over the GCIP in 2008 and oversees the development of indicators, while also assisting cities to join the program. The GCIF is structured around themes that are organized into two broad categories covering city services and quality of life (tables 4.10 and 4.11). The indicators listed are core indicators; other supporting indicators are also used (see http://www.cityindictors.org /themes.aspz#Education). The two categories—city services and quality of life—are organized around 20 themes, consisting of core and supporting indicators. Cities are expected to report on core indicators annually and are encouraged to report on supporting indicators. There are presently 27 core indicators and 38 supporting indicators. The GCIF has identified a set of 10 future indexes for the various themes. These are constructed as weighted combinations of the indica- tors and give a more complete view of city performance or quality of life (box 4.1). Spatial Concentration and Specialization 75 Table 4.10 Global City Indicators: City Services Theme Indicator Education Percentage of children completing primary and secondary educa- tion; student-teacher ratio Fire and Number of firefighters per 100,000 population; number of fire-related emergency deaths per 100,000 population response Health Under-five mortality per 1,000 live births; number of in-patient hospi- tal beds per 100,000 population; number of physicians per 100,000 population; average life expectancy Recreation Square meters of public indoor recreation facility space per capita; square meters of public outdoor recreation facility space per capita Safety Number of homicides per 100,000 population; number of police officers per 100,000 Solid waste Percentage of city population with regular solid waste collection; Percentage of city’s solid waste that is recycled Transportation Number of kilometers of high-capacity public transit system per 100,000 population; number of kilometers of light passenger transit system per 100,000 population; number of personal automobiles per capita; annual number of public transit trips per capita Wastewater Percentage of city population served by wastewater collection Water Percentage of city population with potable water supply service; domestic water consumption per capita; Percentage of city popula- tion with sustainable access to improved water source Energy Percentage of city population with authorized electrical service; total residential electricity use per capita Finance Debt service ratio (debt service expenditures as a Percentage of a municipality’s own source revenue) Governance Percentage of women employed in the city government workforce Urban planning Jobs-housing ratio Source: http://www.cityindicators.org/themes.aspx#Education. Table 4.11 Global City Indicators: Quality of Life Theme Indicator Civic engagement Voter participation (percentage of eligible voters) Culture Percentage of jobs in the cultural sector Economy City product per capital-city unemployment rate Environment PM-10 concentration greenhouse emissions measured in tons per capita Shelter Percentage of city population living in slums Social equity Percentage of city population living in poverty Subjective well-being Subjective well-being index Technology and innovation Number of Internet connections per 100,000 population Source: http://www.cityindicators.org/themes.aspx#Education. 76 Geography of Growth Box 4.1 Indexes Used in the Global City Indicators Program • Competitiveness • Social capital • Creativity • Subjective well-being • Greenhouse gas • Total energy use • Governance • Urban accessibility • Recreation and culture • Water quality. Source: Bhada and Hoornweg 2009. Each city is responsible for supplying and updating its data on the web portal. The GCIF is a host for globally standardized data, providing free web-based information and assisting cities by identifying and sharing expertise on specific areas of performance so that they may strengthen their policy and management (McCarney 2010). The information can be used to generate reports by peer groups (land area, region, climate type, gross operating budget, and population or gross domestic product [GDP] per capita) or by themes. More than 30 cities have joined the GCIF since its inception, and figures for 2010 indicate 74 member cities. The city membership by population category is shown in figure 4.1. The GCIF is an important tool for urban planners and policy makers, providing, as it does, a series of themes that facilitate measuring city per- formance, capturing trends over time and place, and monitoring the global role being played by cities (table 4.12). Green Cities/Eco Cities “Cities represent a challenge and an opportunity for climate change pol- icy� (Corfee-Morlot et al. 2009, 3). Responsible for most of the world’s economic activity, population, innovation, output, and employment, cit- ies arguably produce most of the global greenhouse gas emissions. Furthermore, cities, especially those located along coastal regions, are Spatial Concentration and Specialization 77 Figure 4.1 GCIF Membership in 2010, by Population Category 18 16 14 number of cities 12 10 8 6 4 2 0 up to 100,000 250,000 500,000 1 million over 4 100,000 to to to to million people 250,000 500,000 1 million 4 million people people people people people Source: McCarney 2010. Table 4.12 Role of Standardized Indicators for Cities Purpose Rationale Measuring city performance Role of cities has expanded to include addressing climate change, partnering with private sector and civic organiza- tions, and attracting foreign investment. Indicators are intended to determine municipal capacity for delivering services, managing growth, providing enhanced account- ability, as well as determining management and financial capacity. National governments are increasingly looking at fiscal discipline at the local government level. Capturing trends over time There is increasing need to know the quality of life, eco- and across cities nomic and demographic trends, and environmental mea- sures adopted in cities. Indicators enable development organizations to monitor aid effectiveness. Indicators can determine benchmarks and targets for cities based on the experience of other cities and enable cities to share best practice. Playing a global role In a more global world, cities are increasingly competing for investments, international events, and corporate and insti- tutional headquarters. Cities are playing an increasingly active role in climate change negotiations. Cities are trying to “brand� themselves and become individual members of a wider urban concept. Source: Bhada and Hoornweg 2009. 78 Geography of Growth vulnerable to the effects of climate change such as urban heat island effects. At the same time, “cities are much better for the environment� (Glaeser 2011, 201). Living in high-rises and walking to work are better for the environment than residing in leafy suburbs in large houses and driving to work (Jacobs 1961, 1969; Owen 2009). Glaeser (2011) dis- cusses the work he carried out with Matthew Kahn on a carbon inventory of new housing throughout America (see Glaeser and Kahn 2010). Some of the results are shown in table 4.13. Summary facts from table 4.13 and Glaeser (2011) suggest the follow- ing conclusions: 1. Big cities mean less driving: on average, when population doubles, carbon dioxide emissions per household due to driving decline by almost a ton per year. 2. City-dwellers use less gas than suburbanites. 3. Electrical appliances account for two-thirds of residential energy use; urbanites use less electricity than suburbanites. 4. The main factor explaining the difference between cities is summer heat. 5. Bigger, denser homes use less electricity. On average, a single-family detached home consumes 88 percent more electricity than the aver- age apartment in a five-or-more-unit building. Table 4.13 Summary of Results on Carbon Emissions per Home, 2006 Cause of carbon Highest carbon emissions Lowest carbon emissions emissions per home per home Driving (A) Southern cities; Greenville, SC; New York Nashville, TN; Oklahoma City, OK; Atlanta, GA Electricity (B) Houston, TX; New Orleans, LA; Coastal California; Northeast; Memphis, TN; Dallas, TX; San Francisco, CA; San Juan, Phoenix, AZ CA; New York, NY; Boston, MA; Las Vegas, NV Natural gas (C) California; Detroit, MI; Grand Rapids, Florida; Miami, FL MI; Buffalo, NY; Chicago, IL; Minneapolis, MN A + B + C + public Houston, TX; Birmingham, AL; San Diego, CA; San Francisco, transit Nashville, TN; Memphis, TN; CA; Los Angeles, CA; San Jose, Oklahoma City, OK CA; Sacramento, CA Source: Glaeser 2011. Spatial Concentration and Specialization 79 6. More centralized metropolitan areas use less electricity than more sprawling places. 7. Natural gas, America’s primary source of warmth, is responsible for almost 20 percent of residential carbon emissions. 8. Adding household emissions (driving, electricity, natural gas) and public transit together shows that cities are greener than suburbs. 9. However, the differences between metro areas are even larger than the differences between individual cities and their suburbs. 10. Therefore, coastal California is the greenest part of the country, and the Deep South is the brownest. Cities have not always been held in such high esteem. Ruskin, an art critic in nineteenth-century London and “an early advocate of town plan- ning,� was a proponent of the small town surrounded by a greenbelt. Greenbelts were a feature of English town planning—for example, London’s greenbelt from 1947 covers 2,000 square miles—and elsewhere (Toronto, Pacific Northwest). One of the major figures in urban planning in the United States, Ebenezer Howard, championed the concept of the “garden city� that would be surrounded by a greenbelt to prevent the town from expanding beyond 32,000 inhabitants. Inhabitants of these cities would “live in nice houses and gardens at the center, walk to work in factories at the rim, and be fed by farms in an outer greenbelt� (Kunzig 2011, 132). The first garden city was Letchworth, England. A further aspect of “bringing the countryside into the city� was to build parks in cities (Glaeser 2011, 203).19 These efforts to merge country and city— greenbelts, garden cities, and parks—were overshadowed by the develop- ment of suburbia in the late nineteenth century. Not all developers were so inclined, but the 1920s, which benefited from architects like Raymond Hood and Hugh Ferriss, turned out to be the “high-water mark for verti- cal America� (Glaeser 2011, 205).20 Rising income and cheap transporta- tion have exacerbated this trend. Furthermore, government subsidies for highways and homeownership have contributed to sprawl as well as indi- viduals’ preferences for large homes on large lots. Urbanization continues. People are moving in droves to the cities, par- ticularly in the developing economies; in China and India, in particular, the sheer numbers pose the biggest challenge for urban development. If, as Glaeser (2011) remarks, carbon emissions in China and India rise to U.S. levels per capita, the world’s carbon consumption will increase 139 percent, even if their population stays the same. Most of the carbon emissions in China come from industry. Chinese households are thrifty 80 Geography of Growth energy users—the typical Beijing household emits 3.997 tons of carbon dioxide per year compared to 43 tons in the typical Washington, DC, household (Zheng et al. 2009). Both negative and positive aspects of the urbanization patterns of China and India are evident. On the positive side, both countries have cities that are extremely dense. Mumbai, with more than 50,000 people per square mile, is almost twice as dense as Bangalore, Kolkata, and New York City, each with more than 20,000 people per square mile. In China, Shenzhen has more than 15,000 people per square mile. These levels of density are compatible with public transport and are challenging for car usage. However, Shanghai and Beijing, at 20 million and 17 million inhabitants, respectively, are roughly one-tenth as dense as New York City and less than half as dense as Los Angeles. Car ownership rates in both China and India are increasing exponentially, and India’s Tata Group has produced a car for US$2,500 (Glaeser 2011). Glaeser (2011, 222) concludes, “If the future is going to be greener, then it must be more urban. Dense cities offer a means of living that involves less driving and smaller homes to heat and cool.� The manner in which cities develop offers many possibilities for the environment. Managing this change requires input from many different levels of government. The 2009 OECD report “Cities, Climate Change, and Multilevel Governance� provides an in-depth examination of this topic and advocates a multilevel governance framework to ensure that cities develop in an eco-sustainable manner (Corfee-Morlot et al. 2009). The multilevel governance framework explores the relationships between the various levels of government—local, regional, and national—in clos- ing or narrowing any policy gaps for climate change. Policy gaps are examined along two dimensions: vertical and horizontal. The vertical dimension of multilevel governance calls for national governments to work closely with local and regional governments and vice versa. While local and regional governments may be the agents of change, change often cannot proceed without modifying the legal and institutional frameworks of the country. The horizontal dimension recognizes the chances for transmitting information and best practices between cities, regions, and nations (Corfee-Morlot et al. 2009). The report concludes with some general observations of good multilevel governance practice: 1. National policies can powerfully enable local action on climate change adaptation and mitigation. 2. There is significantly greater potential for experimentation at local scales, which in turn can be a testing ground for national governments. Spatial Concentration and Specialization 81 3. Close collaboration between local and national authorities to build capacity to address climate change will improve the chances that local authorities will exploit potential for cost-effective mitigation and adap- tation to climate change. 4. Some effective cross-sectoral regional or urban development strategies appear to be driven by the climate change imperative, where climate change mitigation and adaptation are seen to be a potential source of regional economic development (Corfee-Morlot et al. 2009). The multilevel governance framework represents the ideal in manag- ing climate change. The reality is far from this: “Climate policy at the city-scale remains fragmented, and the basic tools to facilitate good deci- sion making are still lacking� (Corfee-Morlot et al. 2009, 87). The report suggests that the following tools could assist cities to be more effective in their climate change efforts: • Harmonized inventory of greenhouse gas emissions and reporting pro- tocols to allow cities to monitor and compare progress in mitigating emissions, to assess cost-effectiveness of additional mitigation options, and eventually to become active participants in international carbon markets. • Regional impact science and other policy-relevant research programs to support the interface between expert information and local knowledge and promote local understanding of climate change risk and policy options—from assessment to management—for better mitigation and adaptation decision making. • Urban climate policy networks to build on regular channels of com- munication among national planners and regional and local government officials as well as among local stakeholders and decision makers about targets, goals, strategies, and measures (Corfee-Morlot et al. 2009). The report concludes by noting the progress made on a multilevel governance framework in various countries and cities. Conclusion The chapter has examined several constructs by which cities have been identified in recent years—the knowledge city, the creative city, the global city, and the green or eco city. These are typically developed-country constructs. This typology arises from the continuing structural change at 82 Geography of Growth the level of the economy, which sees more and more employment in the service sectors as routinized manufacturing moves to the edge of the cities or even farther afield. Globalization further emphasizes this trend. At the same time, urban living has become more attractive; lower crime rates in inner-city areas have attracted inhabitants, and the rise of the creative class has contributed to the economic development and sustain- ability of the city. Density and diversity are in vogue. Furthermore, urban living is good for the environment. Notes 1. “Theory and empirical evidence suggest scale and knowledge externalities may interact, so that scale benefits are enhanced by knowledge accumula- tion—information spillovers are more beneficial the more educated the population� (Henderson 2010, 520). 2. “Moving from a city of 250,000 to one of 2.5 million is associated empirically with an 80 percent increase in commuting times and housing rental prices� (Henderson 2010, 521). 3. In addition, natural advantage is often a prime reason for the concentration of industry in a geographic location. 4. Roughly half of the 243 urban areas examined were specialized in subsectors of manufacturing, while the rest were specialized in nontraded sectors of education, banking and commerce, medical, and government services. 5. Lever (2002) cites OECD (1999) for demonstrating that investing in research and development, nationally, could be linked to competitive performance and economic growth and that this was the basis for economic development in the developed world. A similar point was made in the U.K. government white paper “Our Competitive Future: Building the Knowledge-Driven Economy� (U.K. Department of Trade and Industry 1998). At the local level, the Scottish executive advocated policy initiatives on economic sectors or clusters that are based on research, knowledge, information, and creativity (Lever 2002). 6. Four cities were excluded from the original 23 European cities owing to dif- ficulties in measuring economic performance during the process of marketi- zation in the 1990s, for example, Budapest, Moscow, Prague, and Warsaw. 7. “Human capital spillovers occur at the city level because skilled workers pro- duce more product varieties and thereby increase labor demand� (Glaeser, Ponzetto, and Tobio 2011, 1). 8. Shapiro (2006), citing Glaeser, Scheinkman, and Shleiffer (1995); Glaeser and Shapiro (2003); Simon (1998, 2004); Simon and Nardinelli (2002). 9. Variables that are correlated with both education and city growth are missing from the regressions. Spatial Concentration and Specialization 83 10. Glaeser and Saiz view dropout and unemployment rates as measures of human capital, but at the lower end of human capital distribution: “Differences in the unemployment rate across cities (less so across metropolitan areas) are generally time invariant and reflect characteristics of the labor force and the industry structure in the city� (Glaeser and Saiz 2004, 57). 11. “In other words, a local neighborhood, in contrast to a region, succeeds by avoiding large numbers of low-educated workers� (Glaeser and Saiz 2004, 58). 12. “The idea of the creative class is far from new; in fact, it is a revival of the high-tech ‘boosterism’ and place marketing� (Pratt 2008, 2). 13. Some writers criticize the concept of the global city as being too exclusive (McCann 2004) and too Western (Robinson 2002). 14. “The number of headquarters is what specifies a global city� (Sassen 2009, 56). 15. The global city-region was identified by Geddes (1915) and Hall (1966) and was the focus of an international conference in 1999. 16. The pilot cities were Belo Horizonte, Porto Alegre, and São Paulo in Brazil; Montreal, Toronto, and Vancouver in Canada; Bogotá and Cali in Colombia; and King County in Washington State, United States. 17. See http://www.cityindicators.org. 18. The Global City Indicators Facility offices opened in Toronto in October 2008, with support from the World Bank’s Development Grant Facility, the University of Toronto, the government of Canada, and participating cities. The GCIF was officially launched in Nanjing, China, on November 3, 2008, at the United Nations World Urban Forum. 19. “In the United States, Frederick Law Olmsted specialized in bringing bucolic vistas to the heart of a city�—for example, Central Park in New York, Jackson Park in Chicago, Belle Isle in Detroit—as well as “green spaces in Buffalo, Louisville, Milwaukee, Montreal, and Washington, DC� (Glaeser 2011, 204). 20. “Between 1930 and 1933, five new buildings opened that soared above 849 feet, the height of the tallest skyscraper in Western Europe today. America would not build another tower that tall for another thirty-six years� (Glaeser 2011, 205). References Acs, Zoltan, and Monika I. Megyesi. 2007. “Creativity and Industrial Cities: A Case Study of Baltimore.� Discussion Paper on Entrepreneurship, Growth, and Public Policy, Max Planck Institute of Economics, Jena, Germany. Bhada, Perinaz, and Dan Hoornweg. 2009. “The Global City Indicators Program: A More Credible Voice for Cities.� Directions in Urban Development, World 84 Geography of Growth Bank, Washington, DC, June. http://siteresources.worldbank.org/INTURBAN DEVELOPMENT/Resources/336387-1226422021646/Directions4.pdf. Buursink, J. 1994. “Euroservices and Euroairports: The Position of East European Cities as Eurocities.� Paper presented to the conference “Cities on the Eve of the 21st Century,� Institut Fédératif de Recherches sur les Economies et les Sociétés Industrielles (IFRESI), Lille, France. March. Cohendet, Pierre, David Grandadam, and Laurent Simon. 2010. “The Anatomy of the Creative City.� Industry and Innovation 17 (1, February): 91–111. Comunian, Roberta, and Alessandra Faggian. 2011. “Creative Universities and Creative Cities: Exploring the Geography of Bohemian Graduates in the UK.� Working Paper, School of Arts, University of Kent. An updated version appears in Handbook of Cities and Creativity, ed. Charlotta Mellander, Åke E. Andersson, and David Anderson. Cheltenham, U.K.: Edward Elgar, 2011. Corfee-Morlot, Jan, Lamai Kamal-Chaoul, Michael G. Donovan, Ian Cochran, Alexis Robert, and Pierre-Jonathan Teasdale. 2009. “Cities, Climate Change, and Multilevel Governance.� OECD Climate Change Working Paper 14, Organisation for Economic Co-operation and Development, Paris. Duncan, Otis D., and Albert J. Reiss Jr. 1956. Social Characteristics of Urban and Rural Communities, 1950. New York: John Wiley and Sons. Finnie, G. 1998. “Wired Cities.� Communications Week International, May 18, 19–22. Florida, Richard. 2002. “The Rise of the Creative Class.� Washington Monthly 34 (May): 15–26. Gabe, Todd, Jaison R. Abel, Adrienne Ross, and Kevin Stolarick. 2010. “Knowledge in Cities.� Staff Report 470, Federal Reserve Bank of New York, September. Geddes, Patrick. 1915. Cities in Evolution. London: Williams and Norgate. Glaeser, Edward L. 2011. Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter, Greener, Healthier, and Happier. New York: Penguin Press. Glaeser, Edward L., and Matthew E. Kahn. 2010. “The Greenness of Cities: Carbon Dioxide Emissions and Urban Development.� Journal of Urban Economics 67 (3): 404–18. Glaeser, Edward L., Giacomo A. M. Ponzetto, and Kristina Tobio. 2011. “Cities, Skills, and Regional Change.� NBER Working Paper 16934, National Bureau of Economic Research, Cambridge, MA. Glaeser, Edward L., and Albert Saiz. 2004. “The Rise of the Skilled City.� Brookings-Wharton Papers on Urban Affairs 5: 47–106. Glaeser, Edward L., José A. Scheinkman, and Andrei Shleiffer. 1995. “Economic Growth in a Cross Section of Cities.� Journal of Monetary Economics 36 (1): 117–43. Spatial Concentration and Specialization 85 Glaeser, Edward L., and J. A. Shapiro. 2003. “Urban Growth in the 1990s: Is City Living Back?� Journal of Regional Science 43 (1): 139–65. Hall, Peter. 1966. The World Cities. New York: McGraw Hill. Harris, Nigel. 1991. “Urbanization and Economic Development: Territorial Specialization and Policy.� In City, Class, and Trade: Social and Economic Change in the Third World, ed. Nigel Harris, 12–64. New York: St. Martin’s Press. Henderson, J. Vernon. 1974. “The Sizes and Types of Cities.� American Economic Review 64 (4): 640–56. ———. 2010. “Cities and Development.� Journal of Regional Science 50 (1): 515–40. Holko, J. 1999. “Can I Get Some Service Here? Information Technology, Service Industries, and the Future of Cities.� Working Paper, Public Policy Institute of California. http://ssrn.com/abstract=985712 or doi:10.2139/ssrn.985712. Jacobs, Jane. 1961. The Death and Life of Great American Cities. New York: Random House. ———. 1969. The Economy of Cities. New York: Random House. Knight, Richard. 1995. “Knowledge-Based Development: Policy and Planning Implications for Cities.� Urban Studies 32 (2): 225–60. Kresl, Peter Karl, and Balwant Singh. 1999. “Competitiveness and the Urban Economy: Twenty-four Large U.S. Metropolitan Areas.� Urban Studies 36 (5-6): 1017–27. Kunzig, Robert 2011. “The City Solution to Earth’s Problems.� National Geographic (December): 124–47. Lambooy, J. A. 2000. “Regional Growth, Knowledge, and Innovation.� In The Knowledge-Based Economy, ed. A. Kulinski and W. M. Orlowski, 100–13. Warsaw: KBN. Lever, W. F. 2002. “Correlating the Knowledge Base of Cities with Economic Growth.� Urban Studies 39 (5-6): 859–70. Liu-Wei, Li, and Pei Yin-Ko. 2010. “A Local City to a Creative City: An Examination of Taichung, Taiwan.� International Journal of Sustainable Development 13 (1-2): 111–21. MacKenzie, R. D. 1933. The Metropolitan Community. New York: McGraw Hill Book Company. Markusen, Ann. 2006a. “A Consumption Base Theory of Development: An Application to the Rural Cultural Economy.� Paper presented at the confer- ence “Opportunities and Challenges Facing the Rural Creative Economy,� Northeastern Agricultural and Resource Economics Association (NAREA), 86 Geography of Growth Mystic, CT, June 13–14. http://nercrd.psu.edu/CreativeEconomy /MarkusenPaper.pdf. ———. 2006b. “Cultural Planning and the Creative City.� Paper presented at the annual meetings of the American Collegiate Schools of Planning, Ft. Worth, TX, November 12. Matthiessen, G. W., and A. W. Schwartz. 1999. “Scientific Centres in Europe.� Urban Studies 36 (3): 453–77. McCann, E. J. 2004. “Urban Political Economy beyond the ‘Global City.’� Urban Studies 41 (12): 2315–33. McCarney, P. L. 2010. “Global City Indicators Facility: City Indicators and Comparative Measurement.� World Urban Forum, UN-HABITAT, Rio de Janeiro, March 24. Moretti, Enrico. 2003. “Human Capital Externalities in Cities.� NBER Working Paper 9641, National Bureau of Economic Research, Cambridge, MA. OECD (Organisation for Economic Co-operation and Development). 1999. Science, Technology, and Industry Scoreboard: Benchmarking Knowledge-based Economies. Paris: OECD. Ogburn, W. F. 1937. Social Characteristics of Cities. Chicago, IL: International Association of City Managers. Olson, Mancur. 1982. The Rise and Decline of Nations: The Political Economy of Economic Growth, Stagflation, and Social Rigidities. New Haven, CT: Yale University Press. Owen, David. 2009. Green Metropolis: Why Living Smaller, Living Closer, and Driving Less Are the Keys to Sustainability. New York: Riverhead Books. Peck, Jamie. 2005. “Struggling with the Creative Class.� International Journal of Urban and Regional Research 29 (4): 740–70. Pratt, A. C. 2008. “Creative Cities: The Cultural Industries and the Creative Class.� LSE Research, London. Robinson, Jenny. 2002. “Global and World Cities: A View from off the Map.� International Journal of Urban and Regional Research 26 (3): 531–54. Romein, Arie, and Jan J. Trip. 2010. “Key Elements of Creative City Development: An Assessment of Local Policies in Amsterdam and Rotterdam.� Delft University of Technology, OTB Research Institute for Housing, Urban, and Mobility Studies, Delft. Rubalcaba-Bermejo, Luis, and Juan R. Cuadrado-Roura. 1995. “Urban Hierarchies and Territorial Competition in Europe: Exploring the Role of Fairs and Exhibitions.� Urban Studies 32 (2): 379–401. Sassen, Saskia. 1991. The Global City: New York, London, Tokyo. Princeton, NJ: Princeton University Press. Spatial Concentration and Specialization 87 ———. 2009. “Cities Today: A New Frontier for Major Developments.� Annals, AAPSS 626 (November): 53–71. ———. 2010. “Swoons over Miami: Saskia Sassen Responds to FP’s Global Cities Index.� Foreign Policy, August 27. http://www.foreignpolicy.com/articles/2010 /08/27/miami_swoon?hidecomments=yes. Shapiro, Jesse M. 2003. “Smart Cities: Explaining the Relationship between City Growth and Human Capital.� Harvard University, Cambridge, MA. ———. 2006. “Smart Cities: Quality of Life, Productivity, and the Growth Effects of Human Capital.� Review of Economics and Statistics 88 (2): 324–35. Simon, Curtis J. 1998. “Human Capital and Metropolitan Employment Growth.� Journal of Urban Economics 43 (March): 223–43. ———. 2004. “Industrial Reallocation across U.S. Cities, 1977–97.� Journal of Urban Economics 56 (1): 119–43. Simon, Curtis J., and Clark Nardinelli. 2002. “Human Capital and the Rise of American Cities, 1900–1990.� Regional Science and Urban Economics 32 (1): 59–96. Taylor, P. J., and D. R. F. Walker. 2001. “World Cities: A First Multivariate Analysis of Their Service Complexes.� Urban Studies 38 (1): 23–47. U.K. Department of Trade and Industry. 1998. “Our Competitive Future: Building the Knowledge-Driven Economy.� White Paper, Department of Trade and Industry, London. UN (United Nations). 2004. Creative Cities: Promoting Social and Economic Development through Cultural Industries. New York: United Nations. Winters, J. V. 2008. “Why Are Smart Cities Growing? Who Moves and Who Stays?� Working Paper 2008-10-1, Andrew Young School of Policy Studies, Georgia State University. http://aysps.gsu.edu/usery/Papers.html. Zheng, Siqi, Rui Wang, Edward L. Glaeser, and Matthew E. Kahn. 2009. “The Greenness of China: Household Carbon Dioxide Emissions and Urban Development.� NBER Working Paper 115621, National Bureau of Economic Research, Cambridge, MA. Zucker, Laura. 1994. “The Artist in Los Angeles County.� In The Arts: A Competitive Advantage for California, ed. KPMG Peat Marwick, 25–34. Washington, DC: Policy Economics Group, KPMG Peat Marwick LLP. CHAPTER 5 The Attributes and Role of “Smart Cities� Growth in developing countries since the early 1980s has been strongly buttressed by globalization and by a wave of innovations released by new general-purpose technologies that have transformed the electronic, electrical engineering, telecommunications, and biopharmaceutical indus- tries and created the Internet. Some countries, mainly in East Asia but also in Latin America, absorbed the new technologies and, through heavy investment in production capacity, infrastructure, and skills, emerged as successful manufacturers and exporters of industrial products. The tempo of global change quickened in the 1990s. It slowed toward the end of that decade, because of the East Asian crisis, but recovery was swift, with world trade and capital flows expanding at record-setting rates between 2005 and 2007.1 However, the financial crisis starting in 2008, the worldwide recession through much of 2009, the severe contraction of trade, and the urgent need for external and internal adjustments by coun- tries with large current account imbalances have given rise to concerns over the medium-term growth prospects of the world economy and the future efficacy of the principal external drivers of growth in recent years: import demand from the leading Organisation for Economic Co-operation and Development (OECD) countries and the offshoring of tradable activities from the United States and Western Europe to economies where production costs are lower. 89 90 Geography of Growth These concerns are motivating a reexamination in industrializing economies of development strategies reliant on processing and assembly- type industries that generate relatively little domestic value added. The viability of investment and export-led growth is being questioned, and countries with trade surpluses are looking for ways to increase the share of domestic consumption in final demand and lessen the reliance on investment as the primary driver of growth. This effort has focused increasing attention on measures to raise the contribution of total factor productivity (TFP) so as to compensate partially or wholly for a decline in investment. If TFP is to displace other sources of growth, policy makers are searching for a combination of factors that will lead to steadily increasing productivity of industry and services (Mokyr 1999). Productivity is a function of the efficient allocation and use of resources, technological capabilities, and innovation across the full spec- trum of economic activities. To maximize gains in productivity, industrial- izing countries will need to address four priorities: • Products and services that will be in growing demand and subject to technological change • A competitive business environment and a financial system that, in concert, lower the barriers to the entry and exit of firms • Incentives for research and development (R&D) with the intention of building world-class innovation capabilities in areas with the greatest long-term commercial potential • The quality of the scientific and technical workforce and the steady accumulation of intangible factors in business and institutions so as to raise efficiency, promote entrepreneurship, increase the returns from research, and encourage profitable innovation. With industrial development of a modern economy almost wholly concentrated in cities, productivity gains accruing from technological progress and from innovation will be spearheaded by urban centers. The experience of advanced countries suggests that a country has only a few such centers of innovation or “smart cities.� Hence, the viability of a “productivity-led� strategy in an industrializing context will rest on the effectiveness of policies—national and local—to groom one or a small number of smart cities that not only are technologically dynamic and innovative but also realize the industrial scale needed to contribute sub- stantially to the overall growth rate of the national economy. This chapter profiles smart cities and discusses policies that can contribute to their flowering and growth. Smart cities are not called into The Attributes and Role of “Smart Cities� 91 existence by the wave of a policy maker’s wand. In recent times, they have morphed from cities that have a strong base of industries with large research content. Therefore, the chapter first underscores the significance of productivity as a source of growth and identifies those industries and products with robust growth prospects, which are the focus of rapid tech- nological change and could provide the underpinnings of a smart city. It draws on the experience of China, which, among the middle-income coun- tries, is most aggressively pursuing the objective of developing smart cities. The chapter then examines the attributes of smart cities that are responsi- ble for their success. It concludes with policy suggestions for how aspiring countries could develop the potential of a few nascent smart cities. Growth and Technology-Intensive Subsectors The centrality of capital for growth in the world as a whole since 1980 is highlighted by Jorgenson and Vu Khuong (2009), who show that capital was the source of 54 percent of growth in 1989–95 and 41 percent during 2000–06, exceeding the contribution of other factors. However, the com- pelling development is the increasing importance of TFP, which accounted for 36 percent of growth in the most recent period compared with less than a fifth in the first. If this trend persists, TFP will become the princi- pal driver of growth, as it already is in the advanced countries (Comin and Hobijn 2010),2 and its persistence will depend less on the intersectoral transfer of resources and more on technology advances, the diffusion of technology, and the narrowing of technological gaps among production units within a subsector. Innovation, or the successful exploitation of new ideas and technology, is the cornerstone of this process.3 A striking example of the salience of total factor productivity is apparent from the partitioning of the sources of growth in China, the second largest and fastest-growing economy in the world (Bosworth and Collins 2007). Capital and TFP contributed 3.2 and 3.8 percent, respec- tively, to China’s gross domestic product (GDP) growth between 1978 and 2004.4 During the period 1993 to 2004, their shares were 4.2 and 4.0 percent, respectively (table 5.1). Capital and TFP contributed 2.2 and 4.4 percent of industrial growth during 1978–2004 compared with 3.2 and 6.2 percent during 1993–2004 (table 5.2). Cross-country empirical evidence from other countries suggests that TFP has risen much faster in the technology-intensive electrical and nonelectrical machinery subsectors (Jorgenson, Ho, and Stiroh 2007). This has enlarged the output share of those industries and raised the average TFP for manufacturing as a whole. 92 Geography of Growth Table 5.1 Sources of GDP Growth in China, 1978–2004 annual % rate of change Source 1978–2004 1993–2004 Output 9.3 9.7 Employment 2.0 1.2 Output per worker 7.3 8.5 Physical capital 3.2 4.2 Land 0.0 0.0 Education 0.2 0.2 Factor productivity 3.8 4.0 Source: Bosworth and Collins 2007. Table 5.2 Sources of GDP Growth in the Industrial and Services Sectors in China, 1978–2004 annual % rate of change Industry Services Source of growth 1978–2004 1993–2004 1978–2004 1993–2004 Output 10.0 11.0 10.7 9.8 Employment 3.1 1.2 5.8 4.7 Output per worker 7.0 9.8 4.9 5.1 Physical capital 2.2 3.2 2.7 3.9 Education 0.2 0.2 0.2 0.2 Factor productivity 4.4 6.2 1.9 0.9 Source: Bosworth and Collins 2007. Over the same two periods, capital contributed 2.2 and 3.2 percent to industrial growth in China, respectively, and TFP contributed 4.4 and 6.2 percent. In 1978–2004, services derived 2.7 percent of growth from capital and 1.9 percent from TFP. The contribution of TFP to services (a sector where technological advances have been slower) fell to just 0.9 percent a year between 1993 and 2004. With China investing more than 46 percent of GDP in 2009–10 and capital spending subject to decreasing returns, as is evident from rising incremental capital output ratios (Yu 2009), the scope for squeezing out additional growth through even larger injections of capital has been largely exhausted. Investment as a share of GDP must decline, and if growth rates in the 7–8 percent range are to be maintained, the share of TFP would need to rise even higher. This also applies to other middle- income countries, where the investment rates are trending downward and threaten to depress growth rates that are already well below the peaks reached in the 1990s. Undoubtedly, reducing intersectoral and The Attributes and Role of “Smart Cities� 93 intrasectoral gaps in productivity will boost TFP in all the industrializing economies, but raising and maintaining the contribution of TFP to levels in excess of the 1.5 percent average in most countries will require accel- erating technological change and innovation, which in turn will be paced by the evolving composition of industry. This raises an important question regarding subsectors that are growing most strongly, are likely to undergo rapid technological change, and are likely to register the largest productivity gains. Answers must necessarily be hedged because an examination of past trends casts a narrow beam of light into the near future only. International and Chinese experience sug- gests that manufacturing is the leading source of technological innovation. It has more links to other activities, including services, and the highest direct and indirect job multipliers (see Yusuf and Nabeshima 2010). This is borne out by recent trends in production, trade, patenting activity, and exports. These are not a sufficient basis for targeting industry, but they do indicate the nature of industrial opportunities for countries seeking to restore their growth rates through the midwifery of smart cities. Export Composition and Growth From the data on the fastest-growing global exports during 1997–2007 and the most rapidly expanding exports for the Asia region (tables 5.3 and 5.4), three manufactured products stand out: optical devices, tele- communications and transport equipment, and white goods. In 1985, more than 60 percent of China’s exports were resource- and agro-based products and primary products. Electronics and other high-technology products accounted for a little more than 5 percent of the total. Five years later, the share of the former group had been cut almost by half, and by Table 5.3 Fastest-Growing Manufactured Exports Worldwide, 1997–2007 Type of product Average growth rate (%) Optical instruments and apparatus 77.0 Platinum and other metals 74.0 Glycosides; glands or other organs 50.7 Other nitrogen-function compounds 49.0 Other articles of precious metal 48.4 Nickel and nickel alloys, unwrought 46.4 Nickel and nickel alloys, worked 40.3 Cyclic hydrocarbons 40.0 Orthopedic appliances 39.2 Medicaments (including veterinary) 39.2 Source: UN Comtrade data. 94 Geography of Growth Table 5.4 Fastest-Growing Manufactured Exports from Asia, 1997–2007 Type of product Average growth rate (%) Dish washing machines, household 1,703.0 Other articles of precious metals 198.7 Radiotelegraphic and radiotelephonic 147.8 Cellulose acetates 135.5 Silver, unwrought, unworked or semimanufactured 135.1 Aircraft 126.1 Optical instruments and apparatus 122.1 Reaction engines 111.4 Nickel and nickel alloys, unwrought 109.5 Drawn or blown glass, unworked 104.6 Source: UN Comtrade data. 2006, it was down to 12 percent. The big gainers were exports of elec- tronic and telecommunications products and office equipment, the shares of which grew from 5.4 percent in 1985 to more than one-third in 2006. A very similar transformation can be seen in the export mix of other Southeast Asian countries. Further information on China’s exports can be gleaned from tables 5.5 and 5.6. Transport equipment, electrical equipment, chemicals, and machinery emerge as the leading industries that are also contributing the most exports. Pattern of Imports Imports of manufactures by industrializing countries, many of which are from technologically more advanced countries, provide further clues by highlighting the demand for products and services that cannot be met competitively from local sources and pointing to opportunities for upgrad- ing and diversification. Again, China can illuminate the situation because it is a large importer of intermediate and capital goods that support its assembly industries. China’s imports for 2002–08 are presented in table 5.7. The data are highly aggregated, but they indicate that growth rates for electronics, computers, telecommunications, and optoelectronics are slow- ing and their shares in total imports are falling (table 5.8). The demand for life sciences and biotechnology products remains robust, and their share of total imports, although still small, is on the rise. Scale favors electronic components and telecommunications, while growth is in the life sciences. R&D and Patenting The R&D intensity of individual industries and trends in patenting during 2005–09, by identifying the most technologically dynamic subsectors, The Attributes and Role of “Smart Cities� 95 Table 5.5 Fastest-Growing Manufacturing Industries in China, 1996–2003 Industry Average growth rate (%) Transport equipment 505.3 Iron and steel 496.4 Industrial chemicals 476.8 Machinery, except electrical 474.0 Food products 464.8 Machinery, electric 352.8 Professional and scientific equipment 17.6 Petroleum refineries 16.0 Furniture, except metal 14.4 Nonferrous metals 14.1 Source: UN Comtrade data. Table 5.6 Top 10 Exports from China, 2006 Description Trade value (US$ millions) Complete digital data-processing machines 43,384 Peripheral units, including control and adapting units 37,594 Television, radio broadcasting, transmitters, other 35,776 Parts, nes of and accessories for machines of headings 7512 and 752 32,786 Parts, nes of and accessories for apparatus falling in heading 76 31,474 Electronic microcircuits 21,306 Other sound recording and reproducer, nes; video recorders 21,266 Footwear 21,015 Children’s toys, indoor games, and so on 18,011 Outerwear, knitted or crocheted, not elastic or rubber- ized; other, clothing accessories, nonelastic, knitted or crocheted 14,892 Source: UN Comtrade data. Note: nes = not elsewhere specified. offer additional guidance on industrial prospects. For this purpose, patents registered with the U.S. Patent and Trademark Office (USPTO) and the World Intellectual Property Organization (WIPO) can provide a global perspective and also reveal the trends in patenting by Chinese residents. Because it is costly to register with the USPTO and the WIPO and the evaluation process is both standardized and exacting, the patents approved by these bodies tend to be, on average, of higher quality than patents registered elsewhere. 96 Geography of Growth Table 5.7 Imports to China, 2002, 2005, 2008 US$ (100 millions) Indicator 2002 2005 2008 Total merchandise 2,951.7 6,599.5 11,325.6 Total industrial products 2,459.0 5,122.4 7,701.7 Machinery and electronics 1,555.9 3,503.8 5,386.6 Percentage of merchandise 52.7 53.1 47.6 Percentage of industrial products 63.3 68.4 69.9 High-tech products 828.4 1,977.1 3,418.2 Percentage of merchandise 28.1 30.0 30.2 Percentage of industrial products 33.7 38.6 44.4 Source: Ministry of Science and Technology (http://www.sts.org.cn/sjkl/gjscy/index.htm); State Statistics Bureau; General Administration of Customs. Table 5.8 Imports of High-Tech Products as a Percentage of Total Imports in China, 2002, 2005, 2008 Imports 2002 2005 2008 Computers and telecommunications 9.50 9.10 7.03 Life science technologies 1.00 0.70 0.70 Electronics 11.50 15.30 14.20 Computer-integrated manufacturing 3.10 2.50 2.20 Aerospace 1.60 1.30 1.20 Optoelectronics 0.50 0.50 4.30 Biotechnology 0.04 0.02 0.03 Materials 0.70 0.40 0.50 Other technologies 0.20 0.03 0.04 Source: Ministry of Science and Technology (http://www.sts.org.cn/sjkl/gjscy/index.htm); State Statistics Bureau; General Administration of Customs. As presented in table 5.9, the top five categories approved by the USPTO are drug, bio-affecting and body treating compositions (3.1 per- cent), semiconductor device manufacturing process (2.9 percent), active solid-state devices (2.7 percent), multiplex communications (2.4 per- cent), and telecommunications (2.0 percent). Residents of China who registered with the USPTO received the largest number of patents for electronic and electrical devices, followed by communications devices, software, pharmaceutical compounds, and optical devices. The two leading categories of patents approved by the WIPO are elec- tronic and electrical devices and chemical compounds, including pharma- ceutical and biotech products (table 5.10). These are followed by mechanical engineering patents and patents for instruments, including optical devices. Electronic and electrical industries dominated patenting The Attributes and Role of “Smart Cities� 97 Table 5.9 Top USPTO Patents Worldwide, 2005–09 Class Class title % of total patents 424 Drug, bio-affecting and body-treating compositions 3.1 (includes class 514) 438 Semiconductor device manufacturing: process 2.9 257 Active solid-state devices (for example, 2.7 transistors, solid-state diodes) 370 Multiplex communications 2.4 455 Telecommunications 2.0 435 Chemistry: molecular biology and microbiology 1.7 532 Organic compounds (includes classes 532–70) 1.6 375 Pulse or digital communications 1.3 359 Optical: systems and elements 1.3 385 Optical waveguides 1.1 123 Internal-combustion engines 1.0 356 Optics: measuring and testing 0.9 280 Land vehicles 0.7 530 Chemistry: natural resins or derivatives; peptides 0.5 or proteins; lignins or reaction products thereof 296 Land vehicles: bodies and tops 0.5 180 Motor vehicles 0.4 426 Food or edible material: processes, compositions, 0.2 and products 99 Foods and beverages: apparatus 0.1 452 Butchering 0.1 Source: USPTO. in the United States from 1960 to 2005 and also contributed the most to gains in productivity (tables 5.11 and 5.12). The R&D data and patent statistics provide a window on the distribu- tion of technological activity and point to those industries that are likely to be a focus of innovations as patented knowledge is commercialized. When the data on patents are combined with the data on trade, electronic and optical devices are in the lead with respect to global demand and technological prospects. Chemical products and transport and engineering products fall into second and third places. This is the crude ranking of industrial activities that emerges from trends in a few select indicators. This ranking is in line with casual empiricism and the information pre- sented in the business literature: over the next five or more years, elec- tronic, communication, and optical industries will remain the leading subsectors in the world. Chemical and biological products, drawing on the vast amount of research in the life sciences, will also be of significance. 98 Geography of Growth Table 5.10 Share of WIPO Patents, by Sector, 2007–09 % of all patents % of China’s China’s patents Sector and field of technology issued patents as % of all patents Total 100.00 100.00 3.15 I Electrical engineering 29.48 53.14 5.67 1 Electrical machinery, apparatus, energy 5.20 5.38 3.25 2 Audio-visual technology 3.16 2.46 2.45 3 Telecommunications 4.61 11.33 7.73 4 Digital communication 4.69 25.76 17.28 5 Basic communication processes 0.87 0.78 2.84 6 Computer technology 6.37 5.11 2.53 7 IT methods for management 1.27 0.70 1.72 8 Semiconductors 3.31 1.62 1.54 II Instruments 16.23 7.86 1.52 9 Optics 2.96 1.59 1.69 13 Medical technology 5.90 2.72 1.45 III Chemistry 29.61 18.49 1.97 15 Biotechnology 3.61 1.98 1.73 16 Pharmaceuticals 37.67 4.55 2.34 18 Food chemistry 1.11 0.72 2.04 19 Basic materials chemistry 3.42 1.68 1.54 20 Materials, metallurgy 2.00 1.37 2.16 21 Surface technology, coating 2.04 1.08 1.67 22 Microstructural and nanotechnology 0.25 0.04 0.45 23 Chemical engineering 2.76 2.08 2.38 24 Environmental technology 1.51 1.20 2.49 IV Mechanical engineering 18.31 12.93 2.22 32 Transport 3.46 2.21 2.01 Source: China State Intellectual Property Office. Note: Under the WIPO approach, one application may have several classes of intellectual property and may belong to different fields of technology. In this case, every technology field is counted. As a result, the sum of the total number of all technology fields could be larger than the total number of applications in the year. Table 5.11 Top Five Patenting Industries in the United States, 2006 Rank Industry 1 Electronic components and accessories and com- munications equipment 2 Office computing and accounting machines 3 Professional and scientific instruments 4 Electrical transmission and distribution equipment 5 Industrial organic chemistry Source: World Intellectual Property Organization. The Attributes and Role of “Smart Cities� 99 Table 5.12 Top Five Industries Contributing to TFP Growth in the United States, 1960–2005 Rank Industry 1 Computers and office equipment 2 Electronic components 3 Telephone and telegraph 4 Food 5 Rubber and miscellaneous plastics Source: World Intellectual Property Organization. This suggests that the aspiring smart city in an industrializing country should be building on an emerging or established comparative advantage in such manufacturing activities. This is not to say that other industries and tradable services should be excluded from consideration, only that virtually all of the dynamic smart cities in the world have achieved their standing because of the electronic, information technology (IT), tele- communications, transport, and biotech industries. For some, these are now providing the stepping-stones to the development of “green� indus- tries, new materials, and advances in nanotechnology. Having established a strong presence in several of the most dynamic industries, Chinese firms, for example, are eager to move up the value chain from the assem- bly and testing of standardized products to the design and manufacture of differentiated parts and components and innovative products that generate higher profit margins.5 What Makes Cities Smart From the perspective of tomorrow’s smart cities, IT, electronics, biotech, chemical, and yet undiscovered general-purpose technologies6 will serve as the springboards for tackling a new generation of problems with novel solutions and laying the groundwork for new industries. As W. Brian Arthur (2009, 164, 169) observes, “Innovation arises when people are faced with … well-specified problems. … Novel technologies arise from a combination of existing technologies.� He rightly notes that a general-purpose technology “does not just offer a set of limited functions, it provides a vocabulary of elements that can be put together—programmed—in endlessly novel ways for endlessly novel purposes� (Arthur 2009, 88). Thus, the makings of the next techno- logical revolution are already in place and primed for a new round of innovation. What is needed is the orientation of research efforts toward key longer-term problems backed by the credible commitment 100 Geography of Growth of resources to the deepening of scientific knowledge and to the nur- turing of technologies that weave together findings from relevant fields. National policy can provide the incentive framework for tech- nology development and urbanization, but, because of trade and com- petition, the forces of comparative advantage are exacerbating the differences among the regions of a country. For this reason, municipal policies and local innovation systems will determine the emergence of smart cities (Acs 2000). When East Asian and Latin American economies were attempting to accelerate industrialization and exports, it was important to build produc- tion capacity in processing industries as widely as possible, and in those circumstances, breadth and scale mattered most. By investing in produc- tive assets and borrowing technology from abroad, manufacturing indus- tries could be quickly built up using tried incentives. This explains the rise of industrial cities in Asia, Eastern Europe, and Latin America. A few could become smart cities. In the Right Place Cities with long-term technology or innovation potential are likely at a minimum to be distinguished by (a) a strategic location in a prosperous and growing urban region, (b) a robust recent history of urban develop- ment and industrialization, and (c) adequate land area to accommodate future growth. Climate, environmental conditions, accessibility, and potential amenities have traditionally favored coastal cities, which tend to have first-mover advantages and rich hinterlands and are the focus of migration. For these reasons, China’s “open cities� of the 1980s were all coastal cities. But some smart cities in Europe and the United States are located in the interior, and some are emerging in China as well—cities such as Changsha, Wuhan, Xian, and Zhengzhou. Virtually all of the European and North American cities owe their standing to a strategic location, an industry with one or several leading firms, or the presence of established major teaching and research institutions. Harnessing Intelligence Before industrial cities can become smart cities, enhancing the depth and quality of human capital is critical. Smart cities require institutional mechanisms and research infrastructure for generating ideas and ways of debating, testing, and perfecting these ideas. Smart cities can achieve rapid and sustinable growth of industry by bringing together and fully harnessing four forms of intelligence: the human intelligence inherent in The Attributes and Role of “Smart Cities� 101 local knowledge networks, the collective intelligence of institutions that support innovation through a variety of channels, the production intelli- gence of the industrial base, and the artificial intelligence that can be derived from the effective use of digital networks and online services (Komninos 2008). Smart cities are open to ideas and thrive on the het- erogeneity of knowledge workers drawn from all over the country—and the world. Moreover, such cities are closely integrated with other global centers of research and technology development, and their teaching and research institutions must compete with the best for talent and validation of their own ideas. Last but not least, because smart cities are at the lead- ing edge of the knowledge economy, their design, physical assets, attributes, and governance need to reflect their edge over others. Industrial cities can become smart cities, and a strong manufacturing base is an important asset, as in Munich, Seattle, Seoul, Stuttgart, Tokyo, and Toulouse. But industry is not a necessary condition: Cambridge (United Kingdom), Helsinki, Kyoto, and San Francisco are not industrial cities; they are smart cities that have acquired significant high-tech or IT pro- duction capabilities. Being Big or in an Urban Region Research on agglomeration economies has pointed to the productivity gains that accrue to large cities from scale, diversity, and density of activ- ities and from the apparent superlinearity (albeit modest) of innovations in relation to the size of the city (Carlino, Chatterjee and Hunt 2007; Carlino and Hunt 2009; Gill and Goh 2010; Glaeser and Gottlieb 2009; Rosenthal and Strange 2004; World Bank 2009).7 Diversified urban economies recover more quickly from shocks. Moreover, evidence sug- gests that, because of their greater innovativeness, large cities can serve as nurseries for high-tech, new start-ups (Carlino, Chatterjee, and Hunt 2007; Duranton and Puga 2001). However, it is desirable not to overstate the advantages of size, especially for smart cities. A meta-study of the empirical research on agglomeration economies finds that the productiv- ity gains are in the 3 percent range (Melo, Graham, and Noland 2009). Furthermore, many cities noted for innovation are medium-size cities such as Austin, Boston, Raleigh, San Francisco, and Seattle in the United States and Cambridge (United Kingdom), Eindhoven, Helsinki, Munich, Stockholm, and Toulouse in Europe.8 Even the entire population of Silicon Valley does not exceed 2.6 million. Productivity gains from local- ization economies can be realized by mid-size cities that specialize in manufacturing activities (Henderson 2010). If they are located in an 102 Geography of Growth urban region, they can realize economies of agglomeration, specialization, and scale (a polycentric urban region).9 A typical urban region in an industrializing country is likely to be composed of a large core city ringed by smaller satellite or edge cities. The core city with a broad economic base and clusters of business services serves as the hub of the region and the major source of knowledge gen- eration and spillovers. The more specialized neighboring cities, with lower land and housing costs, host clusters of industrial firms and other activi- ties. The urban core can provide the technological leadership and many of the supporting business services, but innovation is frequently a region- wide activity, and smart cities can be the smaller ones, with specialized clusters and other attributes, which are examined below. Human Capital and High Technology Almost by definition, export-oriented and sustainable cities are (and will be) ones that produce and attract large numbers of skilled and technical workers, raise the quality of human capital,10 and nurture a local innova- tion system.11 The quality of human and knowledge assets is what makes a city smart and entrepreneurial. International research suggests that the presence of “star scientists� can initiate virtuous spirals in the fields where innovation is keyed to scientific advances. The biotech cluster in San Diego arose because several star scientists chose to locate there because the city was enticing and the university offered singular opportunities. First, smart cities have a high ratio of science and technology (S&T) workers in the labor force. Table 5.13 provides a classification of S&T workers for whom data are available in the United States. A similar clas- sification is available for China and other countries. Second, smart cities host several universities, and tertiary-level enrollment is high. Third, the industrial composition of the city favors industries that employ large num- bers of S&T workers and have high rates of patenting (see table 5.14). Fourth, smart cities usually attract one or a few major firms with a focus on dynamic industries that invest heavily in R&D and rely on innovation to maintain competitiveness. Table 5.15 lists U.S. cities ranked by the per- centage of high-tech jobs in the total workforce. Table 5.16 does the same for IT jobs. Smart cities use information and communication technology (ICT) to support industry, the education and research infrastructure, and gover- nance. In the future, ICT will be critical to sustainability because it will enable urban centers to contain energy consumption (for example, in Singapore), provide better services, and build more resilient infrastructure. The Attributes and Role of “Smart Cities� 103 Table 5.13 Science and Technology Occupations in the United States Occupation employment statistics code Occupational title 13017 Engineering, math, natural sciences managers 22102 Aeronautical and astronautical engineers 22105 Metallurgists or metallurgical, ceramic, and materials engineers 22108 Mining engineers 22111 Petroleum engineers 22114 Chemical engineers 22117 Nuclear engineers 22121 Civil engineers 22123 Agricultural engineers 22126 Electrical and electronic engineers 22127 Computer engineers 22128 Industrial engineers, except safety 22132 Safety engineers, except mining 22135 Mechanical engineers 22138 Marine engineers 24102 Physicists and astronomers 24105 Chemists, except biochemists 24108 Atmospheric and space scientists 24111 Geologists, geophysicists, and oceanographers 24199 All other physical scientists 24302 Foresters and conservation scientists 24305 Agricultural and food scientists 24308 Biological scientists 24311 Medical scientists 25102 Systems analysts 25103 Database administrators 25105 Computer programmers 25111 Programmers, numerical tool, and process control 25310 Mathematical scientists 25312 Statisticians Source: Markusen et al. 2004. Smart cities have also taken the lead in providing affordable housing and space for industry and tradable services, thereby ensuring that new indus- tries and talented people remain and contribute to local development. By retaining industry and not segregating the population by income, the city avoids sharp cleavages in income distribution. A city that is top ranked with respect to high-tech and IT scores is Seattle, the home of Boeing and Microsoft. Table 5.17 shows the compo- sition of employment in Seattle by subsector, underscoring the impor- tance of activities notable for their technology intensity, such as aircraft 104 Geography of Growth Table 5.14 S&T Jobs in Selected High-Tech Industries in the United States, 1997 Standard industrial 1997 U.S. S&T occupations as classification Description employment % of industry total 376 Guided missiles and space 76,808 42.7 vehicles and parts 737 Computer programming, data 1,425,663 40.7 processing 381 Search, detection, navigation, 185,888 34.1 guidance equipment 871 Engineering, architectural, and 938,469 30.5 surveying services 357 Computer and office equipment 277,495 30.1 873 Research, development, and 491,699 26.7 testing services 366 Communications equipment 294,531 20.8 372 Aircraft and parts 415,022 17.0 482 Telegraph and other message 815,427 16.4 communications 131 Crude petroleum and natural gas 100,308 15.9 Source: Markusen et al. 2004. Note: S & T = science and technology. Table 5.15 High-Tech Jobs in Selected Cities in the United States, 1997 Metropolitan statistical area Share of workforce (%) Number of jobs (thousands) San Jose, CA 41.3 289.1 Seattle, WA 21.1 174.9 Boston, MA 20.9 281.5 Washington, DC 20.3 321.6 Austin, TX 19.7 75.7 Orange County, CA 18.4 152.4 Raleigh-Durham, NC 16.8 69.0 San Diego, CA 16.4 112.7 Dallas, TX 16.4 197.9 Salt Lake City, UT 16.2 60.6 Source: Markusen et al. 2004. and measuring instruments, and for IT intensity, such as insurance, com- puter programming, and architectural services. Industrialized and Export Oriented From the experience of the OECD countries and the evidence cited ear- lier in this chapter, it appears that an export-oriented manufacturing base The Attributes and Role of “Smart Cities� 105 Table 5.16 IT Jobs in Selected Cities in the United States, 1997 Metropolitan statistical area Share of workforce (%) Number of jobs (thousands) San Jose, CA 21.2 148.7 Washington, DC 17.5 277.1 Boston, MA 16.2 218.5 Orange County, CA 13.9 114.9 Denver, CO 13.5 88.0 Raleigh-Durham, NC 13.5 55.2 Minneapolis–St. Paul, MN 12.5 133.5 Dallas, TX 12.5 150.8 Austin, TX 11.5 44.0 San Diego, CA 11.2 77.1 Source: Markusen et al. 2004. Table 5.17 Key High-Tech Sectors in Seattle Standard industrial classification Description Employment 367 Electronic components and 4,787 accessories 372 Aircraft and parts 74,500 381 Search, detection, navigation, 15,593 guidance equipment 382 Laboratory apparatus and 5,166 analytical, optical instruments 384 Surgical, medical, and dental 5,606 instruments 631 Life insurance 4,235 737 Computer programming and 23,174 data-processing services 871 Engineering, architectural, and 14,906 surveying services 874 Management and public 7,406 relations services Total high-tech industry 174,902 employment High-tech specialization index 2.23 Source: Markusen et al. 2004. is a precondition for the rise of a smart city. As noted, some kinds of manu- facturing industries are among the leading innovators and have registered the highest gains in productivity. Even in the U.S. economy, with its heavy emphasis on services, manufactures account for 62 percent of exports (in 2008), with the 10 leading metropolitan areas responsible for a large 106 Geography of Growth share of the total. Istrate, Rothwell, and Katz (2010, 7) note, “The intro- duction of innovative products often precedes exports … and metro areas are the home to most inventors of patents and a disproportionate share of R&D, science, and even venture capital investments .... There is evi- dence [also] that export-oriented industries produce more patents if they are located near other firms in the same industry.� Industries that nurture a dense network of suppliers facilitate innovation by reducing the cost of bringing ideas to fruition.12 Manufacturing industries are also more likely to attract foreign direct investment (FDI) and to benefit from spillovers.13 The significance of manufacturing as the basis for technology-intensive and innovative activities is even more apparent in the industrializing countries, such as Brazil, China, Malaysia, Poland, and Thailand. Smart cities in these countries will arise from the ranks of the leading centers of industry, such as Bangkok, Penang, Shanghai, and Shenzhen. Walkable Cities that are livable, energy efficient, and well furnished with social capital are designed to be walkable. This means, in practice, that they are compact, zoned for mixed use, safe, pedestrian friendly (with green spaces, sidewalks, shaded lanes, street lights, and underpasses integrated into the walking experience), and have readily accessible public transport. Designing a city to be walkable also minimizes the likelihood of urban sprawl, which results in a healthier population and reduces energy and infrastructure costs (Frumkin, Frank, and Jackson 2004). Compactness and density facilitate face-to-face encounters, which, as Venables (2010, 2) observes, “allow high-frequency exchange of ideas and complex dis- course … the building of trust .… [Moreover,] larger and thicker labor markets can improve the quality of the match between firms with par- ticular skill needs and workers with particular skill attributes, can increase competition in the matching process, and can increase the frequency of meetings.� Wuhan, for example, has the topography and the potential to morph into a city as attractive for the Chinese (and, eventually, interna- tional) creative class as Austin or San Diego or Singapore, but the poten- tial of its many watercourses has yet to be exploited, and little attention has been paid to redesigning the city to reverse its drabness and sprawl. Penang equally has the makings of an innovative city, but it lacks a com- prehensive action plan (see Yusuf 2008). Sustainable Sustainability has taken on much greater significance in the face of impending climate change, but not only because global temperatures The Attributes and Role of “Smart Cities� 107 are rising.14 For a city to thrive and to grow, the availability of adequate supplies of water and energy is a must, as is the effectiveness of infra- structure for disposing sewage and waste and of regulations for manag- ing environmental pollution. Sustainable cities are notable for the quality of governance, and they maintain sound finances with the help of fiscal planning, local tax instruments, intergovernmental transfers, budgetary rules, and accounting procedures. Sustainability in the con- text of an urban region demands systematic coordination among munic- ipalities to ensure the effective planning of infrastructure and also coordination of taxation and zoning.15 In the future, sustainable cities will need to be much more energy frugal and “green� and to strengthen their ability to sustain shocks—financial and weather or climate related. A sustainable urban center in the average lower-middle-income country will need to plan on accommodating a large increase in the population and adopting measures to avoid the spread of urban poverty and of slums (see Linn 2010). In middle-income countries, cities will also have to prepare for the aging of populations. Even the most dynamic Chinese and Indian cities do not meet most of the criteria of sustainability.16 Existing metropolitan regions will need to be significantly reshaped, and emerging cities will need to be much better designed. Because China, India, and other developing countries in Asia and Africa are only partly urbanized, there is scope for improvement, but past mistakes will be costly to undo. Connected Urban connectedness is important at two levels. Successful metropolitan cities are open, trade oriented, and innovative. In a globalized environ- ment, this depends on the quality of the transport and the ICT infrastruc- ture that links the city to the rest of the country and the world and facilitates the flow of goods, services, and capital as well as the circulation of people and ideas. Connectedness at the local level using ICT can deliver efficient solutions for the development of energy, transportation, housing, and buildings. It can promote commercial, social, and academic networking and the creation of social and research capital (it induces face-to-face encounters), which is good for productivity and for livability. Villa and Mitchell (2009, 11) observe, “Knowledge workers are opting for more collaborative and flexible forms of work that allow them to contrib- ute when they want, from virtually anywhere, and with almost anyone. At the same time, the speed demands and complexity of knowledge work have increased significantly, driving the need to collaborate and engage a broader workgroup to obtain needed results.� 108 Geography of Growth Catalyzing Innovation Cities become innovative because existing industries or institutions help to nucleate new activities and start a chain reaction. The process can be initiated by any number of catalysts: the transformation of a local univer- sity, the creation of a new institution, the arrival of a major firm, a small cluster of dynamic start-ups, or some other catalytic event that energizes a combination of intellectual and productive activities. There are virtually no instances in the past two decades of innovative cities being success- fully made to order anywhere in the world. The attempts to engineer science cities such as Tsukuba in Japan and Daejeon in the Republic of Korea as well as other technopoles in Europe have rarely lived up to expectations. Most often, existing urban centers became innovative places because a critical mass of human capital, productive assets, and infrastruc- ture and service providers was catalyzed by a firm, a leader, a university chancellor, or some other event. China is currently taking a top-down approach to creating smart cities. Only time will tell whether the out- comes will match expectations. Toward an Urban Innovation Strategy For smart cities, openness and connectivity are more important than scale. They contribute to the productivity of research and the generation as well as the testing of ideas. However, a minimum level of urbanization economies arising from industrial diversity can confer important benefits by providing a mix of technologies and production expertise out of which innovations can arise and which provide the soil for new entrants to take root. Connectivity via state-of-the art telecommunications and transport infrastructure is a source of virtual agglomeration for an intel- ligent city that confers the advantages of a large urban center without the attendant disadvantages of congestion and pollution. In this respect, the smaller smart cities of Europe and the United States enjoy the advantages of livability without sacrificing the productivity gains accruing from agglomeration. The inland cities of China, such as Changsha, Wuhan, Xi’an, Zhengzhou, and others, have a broad and diverse industrial base; all have the scale to become smart cities, but they do not fit the description of “open� cities connected to other nodes of innovation in China and the rest of the world. These cities remain inward looking and protective of local industry. Wuhan is a production center for optoelectronics, but it is not comparable, for example, with Warsaw (Indiana) in the United States, a The Attributes and Role of “Smart Cities� 109 small city that has become the global leader of the medical equipment and devices industry and the home of leading companies such as Biomet, Medtronic, Symmetry, and Zimmer. Neither can Wuhan’s leading univer- sities, which are among the best in China, compare with the expertise accumulated by Purdue University (in Indiana) and Indiana University in medical technologies. Likewise, Penang in Malaysia hosts a large cluster of electronics firms and is a major source of Malaysia’s exports, but in spite of the efforts of the national and local authorities, Penang is far from becoming a smart city. To exploit the innovation potential inherent in virtual agglomera- tion, smart cities need to network actively with other centers through- out the region and the world and build areas of expertise, as Wuhan can in optoelectronics. This calls for embracing a culture of openness and activism on the part of major local firms and universities to trans- late such a culture into commercial and scientific linkages that span the globe. Wuhan will be recognized as an innovation hotspot for opto- electronics when a few local firms enter the ranks of the world’s lead- ing companies in this field and local universities are viewed as doing path-breaking research in optoelectronics. The remarkable feature of China’s leading inland cities is that each one has moved aggressively to build tertiary institutions and research facilities, trains thousands of engineers and scientists, and is home to one or two universities, which are among the top ranked in China. Chengdu, Xi’an, Zhengzhou, and the others have managed to groom a few firms that could become industrial anchors for local clusters, much like ARM and Cambridge Consultants served as the anchors for the electronics cluster in Cambridge (United Kingdom).17 Several inland cities such as Chengdu, Chongqing, Dalian, and Shenyang have also been successful in persuading multinational corporations to set up production facilities, which augment manufacturing capabilities and create the preconditions for a concentra- tion of the value chain.18 Moreover, the leading inland cities are investing in the transport infrastructure to improve connectivity, and all have estab- lished industrial parks to provide space and services for industry to grow. These, plus a full suite of incentives, satisfy most of the preconditions for the emergence of innovative industrial clusters. What might be missing is focus. The inland cities want to develop several of the industries desig- nated as high tech. For example, electronics, automobiles, biotech, renew- able energy, and advanced materials are on the shopping list of all cities vying to become the smart cities of tomorrow. All of these cities are attempting to upgrade local industries so as to move “up the value chain� 110 Geography of Growth and to link this with a localization of the innovation value chain. All are aiming to increase local value added so as to maximize well-paid jobs and expand the urban revenue base. Although this sets the stage for intense competition, it also could lead to a waste of resources, as cities bid for a limited pool of talent, offer generous incentives to attract domestic and international companies, and protect local producers in an effort to deepen technological capabilities. The end result could be a suboptimal dispersion of scientific talent and of research and production facilities. Instead of a few world-class centers with substantial innovation capabilities and a focus on one or a few tech- nologies, there is the risk that the inland cities would fail to acquire the critical mass of expertise in any area and fail to build innovative clusters. The competition among cities can lead to a massive expenditure on R&D infrastructure and on production capacity, most of it redundant, as each city attempts to raise local value added and reel in more of the innovation value chain. This may have worked when Chinese cities were beginning to produce manufactures for an expanding global market and investing in production capacity was a safe bet. Developing innovative capacity in various smart cities requires a different approach, and capacity building is only one part of the strategy. The innovativeness of cities is related directly to the quality of human talent. China’s coastal cities have been quicker off the mark because they have been more successful in nurturing quality, retaining the most tal- ented knowledge workers, and attracting the cream of the knowledge workers from other parts of the country. The coastal cities are also more open and accessible to outsiders and have integrated with global knowl- edge networks. For smaller inland cities to become innovative smart cit- ies, they will need to specialize and pull in some of the best brains in their fields of specialization from across the country. Any serious attempt to become an innovative city built on the quality of talent, which after all is the life blood of innovation, will have to combine urban design and renewal with a focus on developing a few core areas of world-class expertise. It may be misleading to think that the only industries appropriate for smart cities are the so-called high-tech ones with the largest number of patents in recent years. These deservedly attract the most attention and resources; however, many traditional industries can generate handsome returns through innovations that leverage findings in the life sciences and ICT. The dairy industries in Denmark and New Zealand, two of the lead- ing exporters, have enhanced competitiveness and profitability with the The Attributes and Role of “Smart Cities� 111 help of innovations that improve herd management, optimize the feed of animals, and monitor the condition of individual head of cattle. Efforts to reduce water consumption by the meat-packing and beverage industries and to control pollution are prompting a host of innovations that contrib- ute to the bottom line of firms. The textile industry is improving the variety of its offerings and the attributes of materials as a result of advances in nanotechnology. The huge construction materials industry is primed for technological change, as the efforts to minimize greenhouse gases gather momentum. Likewise, manufacturers of machinery and equipment, at the heart of the industrial economy, are also faced with the challenge of designing machines and techniques so as to use different kinds of materials, reduce waste, and lessen energy consumption. The point is that successful smart cities in industrializing countries do not all have to join the rush toward the electronics, biotech, transport, and renewable energy sectors. There are plenty of other low-hanging fruit, and there are numerous innovations to be made in seemingly mundane indus- tries, some of which will require an adroit combination of technologies— the food-processing industry being one. This industry, which is a natural for cities in northeastern China, such as Changchun, is ripe for innova- tions to cut back sharply on waste, pollution, and energy and water use and to introduce foods that are more nutritious and safeguard health. Aspiring smart cities hosting medium-tech industries can consider whether the future focus of innovative activities could be on some of these industries rather than the fashionable high-tech ones. Their com- parative advantage in innovation might lie in food processing and not in the auto industry. And food processing may call for the development of research in the life sciences in a few specific areas, such as packaging. In other words, a realistic assessment of innovation potential must start from a clear understanding of existing competitive advantage and promising future niches for which competition will not be too fierce. In electronics and auto parts, competition will be deadly, and inland cities might well consider whether they want to invest scarce human resources and capital in becoming, at best, the second-ranked innovative cities in a high-tech industry as opposed to the leading innovative city in a medium-tech or even a formerly low-tech industry, which they are able to revolutionize through innovation. Such innovation is more likely to be inclusive than innovation in advanced materials, for example. Although human talent is the main contributor to the intelligence of cities, the firms that conduct most of the downstream research have a large role to play. The innovativeness of the business sector is a function 112 Geography of Growth of many factors, some of which, such as management and the investment climate, are listed above. With respect to cities in several industrializing countries, two points need to be emphasized. First, state-owned and state-controlled enterprises continue to account for a significant share of production in key industries. Second, although the innovation systems created by the cities are encouraging new entrants, it is not apparent from the low rate of exit that truly innovative firms are being groomed or that struggling firms are being allowed to fail in sufficient numbers. State- owned enterprises tend to be among the least innovative firms and low on the scale of productivity. The larger their share of gross value of indus- trial output (GVIO) and R&D spending, the more protective municipal governments will be of local industry and the less easy it will be for inland cities to enhance innovation capabilities. Furthermore, attempting to build high-tech industries by supporting the entry of firms producing standardized products using well-established technologies is not a prom- ising strategy. Policy Measures That Facilitate Technological Upgrading and Innovation Some policy options with regard to the two-track strategy are outlined below. More efficient business and technical services and government procurement can facilitate the success of the two-track approach. Building “Smart Cities� The central government can promote urban innovation capabilities through several measures. First, the government can enhance the incentives to innovate country- wide by taking steps to increase the integration of the national economy and discourage local protectionism. This would intensify the degree of competition among domestic firms and the competitive pressures from imports, increasing both entry and exit of firms and encouraging firms to compete on the basis of technology. Pricing energy and other nonrenew- able resources appropriately, setting national standards (including envi- ronmental standards and standards encouraging energy efficiency) for products, and enforcing these standards would also generate pressures to upgrade technologies, which some Western countries have done to good effect. The ability of smaller firms to meet these standards would be facilitated by strengthening the industrial extension system and providing smaller firms with access to laboratory, testing, and certification facilities. The Attributes and Role of “Smart Cities� 113 The German Fraunhofer Institutes and the Industrial Development Corporation of Norway are good models for industrializing countries to adapt. In Japan, the TAMA (Technology Advanced Metropolitan Area) Association provides its member firms, most of which are of small and medium sizes, with laboratory facilities and testing equipment plus other services. Second, the central government can take the initiative in building countrywide research networks that enhance the sharing, absorption, and development of technology. Research consortia in Japan, Korea, and the United States have assisted in disseminating the latest technologies and pushing the technology frontier in selected areas. Recognizing the cost and complexity of research in frontier fields, even the largest firms are finding it desirable to specialize and to form partnerships with other firms or with universities when developing sophisticated new products or tech- nologies. In addition to consortia, the technological and innovative capa- bilities of nascent smart cities would benefit if both domestic and foreign firms could be persuaded to locate some of their R&D centers, not just their production facilities, in the cities. Third, international experience suggests that smart cities house leading research universities that compete with each other and with other uni- versities throughout the country. Smart cities are home to at least two to three of a country’s top-ranked schools, and these institutions can mobi- lize the funding to sustain cross-disciplinary postgraduate and postdoc- toral programs and set up specialized, well-staffed research institutes so as to achieve a level of performance comparable to that of institutions in more advanced countries. Many high-tech multinational corporations are investing in R&D facilities outside of their home countries. Smart cities can derive spill- overs from facilitating such investment in R&D infrastructure and in the creation of intangible assets. Cities also gain from significant spillover effects arising from the knowledge and experience imparted to the local workforce, the reputational gains for cities that will come to be seen as science hubs, and the contribution that such research can make to indus- trial upgrading locally. Fourth, the most important contribution universities can make to innovation is by generating ideas and serving as a breeding ground for entrepreneurs who are the vehicles for transforming ideas into commer- cial products and services. Central and municipal governments are in a position to enlarge the share of basic research and to ensure the continu- ity of funding, both of which could build innovation capacity in the smart 114 Geography of Growth cities. The National Institutes of Health in the United States played a central role in the boom in the life sciences because it was and is a source of large and stable funding, much of it for basic research done in universi- ties. This funding financed countless research programs, trained thousands of PhDs, supported postdocs, and created the depth of expertise that enabled the United States to become the leader in the field of biotech. To maximize the spillovers from the government-sponsored research and contests to develop particular types of technologies, one possibility is to make the findings of this research widely available. In the 1950s and 1960s, the research on electronics financed by the U.S. government was shared generously, and this enabled many companies to come up to speed and become innovators themselves. Good research is inseparable from a stringent and disciplined process of refereeing and evaluation of research findings. The research community needs to take the initiative in this area, but the government could provide the parameters. The universities can also take the lead in thickening the scientific culture of their cities by promoting public lectures and exhibitions and contributing to the teach- ing of science in local schools. Fifth, there is the perennial issue of risk capital for innovative firms. Although some public risk capital is available in the industrializing coun- tries, private venture capital for smaller private firms that are trying to scale up is still scarce. One partial solution is to increase lending by banks to high-tech private firms—and not mainly to government-linked compa- nies. Such lending by local banks to local firms and the creation of bank- led relational networks are a mode of financing that seems to work in the United Kingdom and the United States and complements the resources of entrepreneurs, “angel� investors, and venture capitalists. Too little bank financing goes to private firms, especially the riskier high-tech ones. Sixth, high-tech industry depends on a vast range of technical skills to staff factories, render IT support, repair complex equipment, and provide myriad other services. Smaller firms and start-ups frequently have diffi- culty finding such skills and can rarely afford to provide much training in-house. Hence public-private initiatives to secure and replenish the base of technical skills essential for a smart city can circumvent market failures and promote desirable forms of industrial activity, aside from minimizing both frictional and structural unemployment. Labor market institutions can be strengthened and made nondiscriminatory by setting up multilevel professional advisory agencies and increasing the provision of vocational training for which there would be a demand from expanding and new enterprises. The Attributes and Role of “Smart Cities� 115 Industrial cities have attracted a range of business service providers such as engineering research centers and productivity centers, but many of them lack market orientation and suffer from funding and skills short- ages. It is important to make them more functional and more responsive to private sector needs through a public-private partnership approach. However, there are some good examples in China that could be repli- cated. Figure 5.1 illustrates the example of Shanghai’s R&D public ser- vice platform, which offers a wide range of business and extension services. These services cover the innovation development process, including the sharing of scientific information, technology testing and transfer services, and support for entrepreneurship and management. Seventh, although universities across the industrializing world churn out huge numbers of graduates each year, the quality of the training pro- vided is frequently weak. In the meantime, employers experience a serious Figure 5.1 Shanghai R&D Public Service Platform testing base cooperation management science and decision- technology making literature support resources entrepreneur- security ship business innovation services equipment technology sharing transfer scientific figure industry testing sharing professional technology Source: Shanghai Municipality Science and Technology Commission 2006. 116 Geography of Growth shortage of highly skilled technicians, engineers, and executives. This com- bination of low-skill glut and high-skill shortage poses a difficulty for the skill transfer that companies need to improve the quality of their output or move to a more value added link in the chain. Transfer of managerial experience is one of the key ways in which FDI contributes to the Chinese economy. In effect, the education system in China does not sufficiently encourage creativity and initiative, meaning that new graduates often lack the skills most needed as the economy strives for technological maturity. To move forward, both the private sector itself and the government need to invest more in improving human resources management in pri- vate small and medium enterprises (SMEs). The following measures can be considered through a public-private partnership approach (for more details, see Zhang et al. 2009): • Using the legal instruments of confidentiality agreements and competi- tion restrictions to protect technical secrets from being taken by R&D personnel when they leave the firm • Improving labor market conditions by using relevant services. For exam- ple, local governments could create skill development centers to (a) pro- vide SMEs with management and technical training, especially related to innovation; (b) provide information on the demand for and supply of various skills and the premium on various job categories through close relationships with schools, training institutions, and the labor market; (c) collect and disseminate success stories about the management of skilled employees and the promotion of an innovation culture • Strengthening policies supporting training and vocational education by reviewing the ceiling on tax-deductible training expenditures (2.5 percent of wage bill) of enterprises and redefining the role of the government in vocational education. In addition to human resources management, improvements can also be made in facilitating the collaboration of SMEs with knowledge institu- tions and enhancing innovation services. Eighth, scope exists for making better use of demand-side instruments such as government procurement and standard setting. Combined with adequate efforts to guard against the potential risks of rent seeking and protectionism, this would go a long way to encourage the demand for innovation services. However, the procurement policy can be a double- edge sword. The key to success lies in open competition. Some potential risks in this area need to be carefully addressed: (a) the risk of turning the The Attributes and Role of “Smart Cities� 117 government procurement instrument into one that protects national and local products from international and national competition, (b) the chal- lenge of following the procedures laid out to identify the “indigenous innovation products� for the government catalogue, and (c) the risk of making government merely a passive taker of what domestic suppliers offer, rather than a demanding buyer of technologically sophisticated products (Zhang et al. 2009). The demand for innovation could be increased through government standard setting. Standard setting allows governments and other entities to generate demand for advances in, for example, the performance, safety, energy efficiency, and environmental impact of products. To generate more demand for innovation, certain measures could be taken: (a) focus- ing exclusively on product improvement and resisting the temptation to use standard setting to protect or help domestic or local industry; (b) tak- ing European Union or U.S. standards as a technical starting point, while looking for ways to improve product performance; (c) involving industry leaders more in standard setting (this needs to be done in a productive way); and (d) changing the role of government from sole standard setter to time-sensitive driver of industrial consensus (Zhang et al. 2009). Identifying and Promoting Smart Cities Industrializing countries need to embark on a new urban development strategy to realize their growth expectations. Such a strategy will have to be centered on transforming the leading urban centers into smart cities that are not only industrially dynamic but also fruitful sources of innovation. Various cities would benefit from such policies; however, because each city differs with respect to resource base and comparative advantages, additional polices differentiated according to the circumstances would be needed to accelerate the transition to a smart city and create the founda- tions for sustainable innovation. Such differentiated policies that factor in the capabilities of a city, actual and potential, can be constructed using longitudinal data on trade, investment, industrial composition, and the labor market, combined with the gathering of qualitative information from the principal players in a local innovation system (see the annex to this chapter). Picking tomorrow’s smart cities is both easy and difficult. The easy part is identifying cities that are already demonstrating their innovativeness and need to smarten up their act. The hard part is identify- ing the future performers from a long list. 118 Geography of Growth Annex: Technology Capability and Innovation Criteria Cities: Profile and technological capacity • Population of city • Population growth • GDP per capita • Overall GDP growth rate since 2000 and growth by sectors (in com- parison with main competitors in China) • Number of high-tech companies • Percent of workers in high-tech fields • Percent of workforce with advanced degrees • Number and skill composition of in-migrants (since 2000); where do they go? • Number and skill composition of out-migrants • FDI Science and technology input-output indicators • Number of firms filing research joint ventures • Number of research institutes • Number of full-time R&D personnel • Total public funds invested in R&D (and distribution of spending) • R&D funds per capita of R&D personnel • Patents registered by residents at their national offices • Receipts of royalty and license fees • Number of scientists and engineers in workforce • Number of scientific publications (in major journals, past five years) University sector • Number of tertiary institutions and enrollment or number of graduates; percentage in science and engineering disciplines • Percentage of high-achieving graduates who stay in municipality (top 5 percent of class); where do those who leave go? • Enrollment in doctoral programs • Enrollment in postdoctoral programs • Spending on research as a fraction of university budgets; as a fraction of total spending on R&D in municipality • Number of spinoffs • Number of contracts with enterprises; what kinds of contracts? The Attributes and Role of “Smart Cities� 119 • Number of patent applications; number of patents granted? • Strongest university departments (by what criteria? national ranking) • Leading research institutes (by what criteria? national ranking) Manufacturing sector • Largest three manufacturing sectors (percent of GVIO) • Fastest-growing three manufacturing sectors (and share of GVIO) • Top five exports to rest of the country and the world in 2000 and latest year; top five fastest-growing exports • Largest five firms by turnover • Top three firms in the three fastest-growing manufacturing subsectors (how does the largest firm in this set compare in size with the leading firm in this subsector?) • Number of new entrants in these three subsectors in the past five years • Number of exits from these subsectors in the past five years • Number of high-impact firms in these subsectors; that is, firms that doubled their output value in five years • R&D spending by subsectors as a percentage of turnover • Expenditure on technology licensing in the past five years • Number of patents applied for; number granted • FDI in these subsectors over the past five years; national investment in these subsectors; national firms with subsidiaries in city • Labor force composition of fastest-growing subsectors; increase in employment in three fastest-growing subsectors • Number of industrial and technology parks and incubators; numbers of firms in parks, change in number since 2005 Firms in selected subsectors (based on firm surveys) • Proportion of enterprises that conduct R&D activities to total number of enterprises • Proportion of R&D expense to total sales income • Proportion of R&D personnel to total staff • Proportion of enterprises that apply for patents • Proportion of enterprises that possess patents • Proportion of enterprises creating new products in the past three years • Percentage of sales derived from new products 120 Geography of Growth • Proportion of enterprises making any technique improvement in the past three years • Proportion of enterprises having cooperation with high-level education and research institutes • Proportion of product export-oriented enterprises • Enterprises whose proportion of sales of new products or products using new techniques is more than 25 percent of total sales • Proportion of employees with tertiary- or graduate-level qualifications • Proportion of staff who received training abroad • Firm size • Plant vintage • Foreign ownership or part of multinational group • Existence of formal R&D department Policies and incentives • Principal industrial policy objectives • Key policy incentives for achieving these objectives employed since 2000–01 • Results of policy incentives; which were most effective? • Main problem areas and policy challenges Notes 1. The stability of the renminbi during this difficult period assisted the recovery of East Asian economies. World merchandise trade rose by almost 7 percent a year, and annual inflows of foreign direct investment increased from US$959 billion in 2005 to US$1.8 trillion in 2007 (UNCTAD 2008; WTO 2008). 2. Helpman (2004, 33) observes, “More than 60 percent of the variation in income per worker is explained by differences in TFP. The role of TFP is even greater in explaining the cross-country differences in the growth rate of income per worker rather than the differences in the level of income per worker. In the former case, differences in TFP account for 90 percent of the variation.� 3. The literature on the use of knowledge for economic growth originated with the writings of Fritz Machlup in the early 1960s (Machlup 1973). Machlup’s The Production and Distribution of Knowledge in the United States launched the idea of the information revolution and the knowledge society. See Lin (2007) for a detailed account of the genesis and elaboration of the knowledge economy concept. The Attributes and Role of “Smart Cities� 121 4. The sources of growth in China are estimated, among others, by Badunenko, Henderson, and Zelenyuk (2008); Urel and Zebregs (2009), and Wang and Yao (2003); all of whom find that capital played the leading role. Time-series analysis arrives at similar results. A more recent estimate by Kuijs (2010) pegs the contribution of TFP during 1995–2009 at 2.7 percent and the contribution of capital at 5.5 percent. 5. Although China’s exports of manufactures overlap with those of the United States, there are wide differences in quality and technological sophistication (Edwards and Lawrence 2010). 6. With general-purpose technologies, countries can expect growth through innovations to accelerate quickly after a period of gestation. See Helpman (2004). 7. In the United States, 10 large metro regions are responsible for a third of all patents. 8. Two-thirds of the patents in the United States are not assigned to parties from a large metropolitan area. 9. A background empirical study based on micro-level data in 2007 shows that industrial agglomeration has played a significant role in determining the pro- ductivity of industrial enterprises in China. The productivity effects of indus- trial agglomeration, however, differ across regions, scales, and sectors. The coastal region has especially benefited from agglomeration, and there is scope for interior regions to replicate the coastal region’s experiences (He and Wang 2010). 10. This involves improving the quality of education and of health services. 11. Universities can play a large part in drawing students from other parts of the country and overseas to a city. After they graduate, some stay, adding to the talent pool and helping to create a critical mass of entrepreneurship and skills. See Berry and Glaeser (2005). 12. Helsley and Strange (2002) make the point that the concentration of an industry facilitates not just the generation of ideas but also their realization. 13. A large literature links industrial productivity to exports and to FDI (with qualifications). A recent study by Istrate, Rothwell, and Katz (2010) points to the contribution of large metropolitan areas to U.S. exports, in particular, exports of manufactures. 14. Kahn (2010) discusses how climate change is likely to drive adjustment, inno- vation, and a redistribution of the population among cities through the price mechanism. 15. The absence of coordination can lead to the decline of cities and the out- migration of industry. See, for instance, Pugh O’Mara (2002) on the plight of Philadelphia, where the lack of coordination among the 238 local municipalities 122 Geography of Growth in the Greater Philadelphia area has contributed to the industrial decline of the city. 16. Sanyal, Nagrath, and Singla (2010) describe and discuss the limited progress to date by Indian cities. 17. ARM (Advanced RISC Machines) was established in 1990 as a joint venture between Acorn Computers, Apple, and VLSI Technologies. It is the leading producer of microprocessors for mobile telecommunications. 18. However, most of the more than 600 R&D centers established by multina- tional corporations are in the coastal cities, chiefly Beijing and Shanghai. References Acs, Zoltan. 2000. Regional Innovation, Knowledge, and Global Change. London: Pinter Arthur, W. Brian. 2009. The Nature of Technology: What It Is and How It Evolves. New York: Free Press. Badunenko, Oleg, Daniel J. Henderson, and Valentin Zelenyuk. 2008.“Technological Change and Transition: Relative Contributions to Worldwide Growth during the 1990s.� Oxford Bulletin of Economics and Statistics 70 (4): 461–92. Berry, Christopher R., and Edward L. Glaeser. 2005. “The Divergence of Human Capital Levels across Cities.� Discussion Paper 2091, Harvard Institute of Economic Research, Cambridge, MA, September. http://ssrn.com/abstract= 794551 or doi:10.2139/ssrn.794551. Bosworth, Barry, and Susan M. Collins. 2007. “Accounting for Growth: Comparing China and India.� NBER Working Paper 12943, National Bureau of Economic Research, Cambridge, MA. Carlino, Gerald A., Satyajit Chatterjee, and Robert M. Hunt. 2007. “Urban Density and the Rate of Invention.� Journal of Urban Economics 61 (3): 389–419. Carlino, Gerald A., and Robert M. Hunt. 2009. “What Explains the Quantity and Quality of Local Inventive Activity?� Working Paper 09-12, Federal Reserve Bank of Philadelphia, Philadelphia, PA. Comin, Diego A., and Bart Hobijn. 2010. “Technology Diffusion and Postwar Growth.� NBER Working Paper 16378, National Bureau of Economic Research, Cambridge, MA, September. Duranton, Gilles, and Diego Puga. 2001. “Nursery Cities: Urban Diversity, Process Innovation, and the Life-Cycle of Products.� American Economic Review 9 (5): 1454–77. Edwards, Lawrence, and Robert Z. Lawrence. 2010. “Do Developed and Developing Countries Compete Head to Head in High Tech?� NBER Working Paper 16105, National Bureau of Economic Research, Cambridge, MA. The Attributes and Role of “Smart Cities� 123 Frumkin, Howard, Lawrence D. Frank, and Richard Jackson. 2004. Urban Sprawl and Public Health: Designing, Planning, and Building for Healthy Communities. Washington, DC: Island Press. Gill, Indermit S., and Chor-Ching Goh. 2010. “Scale Economies and Cities.� World Bank Research Observer 25 (2): 235–62. Glaeser, Edward L., and Joshua D. Gottlieb. 2009. “The Wealth of Cities: Agglomeration Economies and Spatial Equilibrium in the United States.� Journal of Economic Literature 47 (4): 983–1028. He, Canfei, and Junsong Wang. 2010. “Spatial Restructuring of Chinese Manufacturing and Productivity Effects of Industrial Agglomeration.� World Bank, Washington, DC. Helpman, Elhanan. 2004. The Mystery of Economic Growth. Cambridge, MA: Belknap Press of Harvard University Press. Helsley, Robert W., and William C. Strange. 2002. “Innovation and Input Sharing.� Journal of Urban Economics 51 (1): 25–45. Henderson, J. Vernon. 2010. “Cities and Development.� Journal of Regional Science 50 (1): 515–40. Istrate, Emilia, Jonathan Rothwell, and Bruce Katz. 2010. “Export Nation: How U.S. Metros Lead National Export Growth and Boost Competitiveness.� Metropolitan Policy Program, Brookings Institution, Washington, DC, July. http://www.brookings.edu/~/media/Files/rc/reports/2010/0726 _exports/0726_exports_istrate_rothwell_katz.pdf. Jorgenson, Dale W., Mun S. Ho, and Kevin J. Stiroh. 2007. “The Sources of Growth of U.S. Industries.� In Productivity in Asia: Economic Growth and Competitiveness, ed. Dale W. Jorgenson, Masahiro Kuroda, and Kazuyuki Motohashi, Northampton, MA: Edward Elgar. Jorgenson, Dale W., and M. Vu Khuong. 2009. “Growth Accounting within the International Comparison Program.� ICP Bulletin 6 (1): 3–28. Kahn, Matthew. 2010. Climatopolis: How Our Cities Will Thrive in the Hotter Future. New York: Basic Books. Komninos, Nicos. 2008. Smart Cities and Globalization of Innovation Networks. New York: Routledge. Kuijs, Louis. 2010. “China through 2020: A Macroeconomic Scenario.� Working Paper 9, World Bank, Beijing. Lin, B. C. 2007. “A New Vision of the Knowledge Economy.� Journal of Economic Surveys 21 (3): 553–84. Linn, J. F. 2010. “Urban Poverty in Developing Countries: A Scoping Study for Future Research.� Wolfensohn Center for Development Working Paper 19, Brookings Institution, Washington, DC. Machlup, Fritz. 1973. The Production and Distribution of Knowledge in the United States. Princeton, NJ: Princeton University Press. 124 Geography of Growth Markusen, Ann, Karen Chapple, Daisaku Yamamoto, Gregory Schorock, and Pingkang Yu. 2004. “Gauging Metropolitan ‘High-Tech’ and ‘I-Tech’ Activity.� Economic Development Quarterly 18 (1): 10–24. Melo, Patricia C., Daniel J. Graham, and Robert B. Noland. 2009. “A Meta- Analysis of Estimates of Urban Agglomeration Economies.� Regional Science and Urban Economics 39 (3): 332–42. Mokyr, Joel. 1999. “Editor’s Introduction: The New Economic and the Industrial Revolution.� In The British Industrial Revolution: An Economic Perspective, 2d ed., ed. Joel Mokyr, 1–127. Oxford: Westview Press. Pugh O’Mara, Margaret. 2002. “Learning from History: How State and Local Policy Choices Have Shaped Philadelphia’s Growth.� Greater Philadelphia Regional Review (March). Rosenthal, Stuart S., and William C. Strange. 2004. “Evidence on the Nature and Sources of Agglomeration Economies.� In Handbook of Urban and Regional Economics, Vol. 4, ed. J. Vernon Henderson and J. F. Thisse, 2119–71. Amsterdam: Elsevier. Sanyal, Sanjeev, Sumati Nagrath, and Gorika Singla. 2010. “The Alternative Urban Futures Report: Urbanisation and Sustainability in India; An Interdependent Agenda.� Report prepared by Mirabilis Advisory, Gurgaon, India, for the World Wildlife Foundation India. Shanghai Municipality Science and Technology Commission. 2006. “The Innovation System of Shanghai.� Presentation made to an Organisation for Economic Co-operation and Development delegation, Shanghai, October 9. UN Comtrade (United Nations Commodity Trade). Statistics Database. http:// comtrade.un.org/db/. UNCTAD (United Nations Conference on Trade and Development). 2008. “Foreign Direct Investment May Have Peaked in 2007, Annual Report Reveals.� Press Release, UNCTAD, Geneva. Urel, Bulent, and Harm Zebregs. 2009. “The Dynamics of Provincial Growth in China: A Nonparametric Approach.� IMF Staff Papers 56 (2): 239–62. Venables, Anthony J. 2010. “Productivity in Cities: Self-Selection and Sorting.� Discussion Paper Series 507, Department of Economics, University of Oxford, October. Villa, Nicola, and Shane Mitchell. 2009. “Connecting Cities: Achieving Sustainability through Innovation.� Paper presented at the Fifth Urban Research Symposium 2009, “Connected Urban Development,� Cisco Systems, Internet Business Solutions Group. Wang, Yan, and Yudong Yao. 2003. “Sources of China’s Economic Growth, 1952– 99: Incorporating Human Capital Accumulation.� China Economic Review 14 (1): 32–52. The Attributes and Role of “Smart Cities� 125 World Bank. 2009. World Development Report 2009: Reshaping Economic Geography. Washington, DC: World Bank. WTO (World Trade Organization). 2008. “Developing, Transition Economies Cushion Trade Slowdown.� Press Release, WTO, Geneva. Yu, Yongding. 2009. “China’s Policy Response to the Global Financial Crisis.� Richard Snape Lecture, Productivity Commission, Melbourne. Yusuf, Shahid. 2008. “Can Clusters Be Made to Order?� In Growing Industrial Clusters in Asia: Serendipity and Science, ed. Shahid Yusuf, Kaoru Nabeshima, and Shoichi Yamashita. Washington, DC: World Bank. Yusuf, Shahid, and Kaoru Nabeshima. 2010. Two Dragon Heads: Contrasting Development Paths for Beijing and Shanghai. Washington, DC: World Bank. Zhang, Chunlin, Douglas Zhihua Zeng, William Peter Mako, and James Seward. 2009. Promoting Enterprise-Led Innovation in China. Directions in Development. Washington, DC: World Bank. CHAPTER 6 Globalization, Urban Regions, and Cluster Development Globalization has resulted in not just a closer integration of economies but also a tightening of the links among major cities. Advances in information and communication technology (ICT) and transport tech- nologies, together with the modernization of urban infrastructure, have further facilitated interaction among cities at many different levels and contributed to the emergence of global urban regions. Cities, like Bangkok, Seoul, and Shanghai, lie at the core of urban regions and benefit from agglomeration economies that arise from specialization and scale of production and from industrial diversity that promotes spillovers and the emergence of new activities. Research suggests that each doubling of city size can raise productivity by between 3 and 14 percent. Urban regions are characterized by a con- centration of services, high-tech and creative activities, and nascent industries in the core city, with large-scale manufacturing coalescing in nearby medium-size cities and more specialized cities. This arrange- ment optimizes the gains from urbanization economies in the core city and localization economies in the hierarchy of medium and small- size cities in the urban region. 127 128 Geography of Growth A Holistic Approach to Development Globalization has created new channels for comparing experiences and sharing lessons. At the same time, it has sharpened the competition for final goods and mobile human capital. This competition is multidimen- sional, and it is forcing cities within urban regions to take a holistic approach to development and to compete on many different fronts—the business climate and the urban infrastructure being just two areas, with others, such as livability and urban amenities, acquiring more significance. To attract resources and sustain the momentum of development, cities need to demonstrate their ability to enhance growth potential by cultivat- ing several vibrant and preferably interlaced leading subsectors. Growth potential depends also on the age structure of the population, whether it is expanding or not, and the quality of the workforce. Quality, more than volume, of human capital appears to be the most significant determinant of growth. Recent research also suggests that, in view of the importance of entrepreneurship, innovation, adaptation, and invention for techno- logical convergence among countries, the absolute quality of talent and skills might have a strong bearing on economic performance. Growth potential also depends on how close firms are to technologi- cal frontiers. Proximity to the frontier increases returns to research and development (R&D) and the scope for raising productivity through enhanced technological capabilities. In this context, cities signal their potential by their reputation for technological dynamism and openness to ideas. The Role of Clusters The growth imparted by leading sectors can be magnified by the forma- tion of specialized clusters of networked firms that compete, cooperate, and deepen markets for labor, give rise to intangible capital, generate technological spillovers, and promote start-up activity. A symbiotic relationship between manufacturing firms and service pro- viders, as is emerging in the Bangkok, Thailand; Hong Kong SAR/Shenzhen/ Guangzhou/Dongguan, China; and Seoul, Republic of Korea urban regions, for example, can lead to an unbundling of activities and greater specializa- tion, to the advantage of both parties. A significant share—close to 37 percent—of the employment generated by the export of manufactures by U.S. companies is in upstream and downstream services. In fact, manufac- turing gives rise to employment multipliers of up to five and six that are far larger than the multipliers associated with services. Globalization, Urban Regions, and Cluster Development 129 Clusters generally form around nuclei. Urban centers with a strong development orientation and leadership, such as Beijing and Shenzhen, are attractors, and those with a preexisting industrial base can be a source of skills and intangible assets. These assets, which include scientific and non- scientific R&D, software, worker training, brand equity, product design, and organizational capability, have accounted for 27 percent of the growth in the United States since 1995. Major research-oriented firms or multina- tional corporations can provide a nucleus as well, and there are plenty of examples from Cambridge (United Kingdom), Medicon Valley, San Diego, Silicon Valley, and elsewhere of firms such as CCL and Acorn, Hewlett Packard, Hybritech, and Novo Nordisk spawning scores of daughter enter- prises and helping to scale up the activities of a cluster. Multinational corporations and local firms are also giving rise to spinoffs and new start-up firms in Beijing, Seoul, Shenzhen-Guangzhou, and Taipei-Hsinchu. To thrive and grow, clusters require anchors. The size and affluence of the urban market (as in Seoul, Shanghai, and Tokyo) is one of the most important, but there are other anchors of consequence as well. Research universities have an increasing role if they can supply high-quality skills, contribute to network formation—local and global—and enrich the local knowledge economy by way of tacit knowledge, workshops, patenting, publications, trouble shooting, and the dialogue on technology. Vocational training institutions, physical and social infrastructure, affordable housing, and recreational facilities are among some of the other anchors. How a city goes about developing these anchors determines its overall competi- tiveness in the global economy. Competitive clusters must be capable of upgrading, diversifying, and incubating new industries. Silicon Valley, for example, has served as a breed- ing ground for several kinds of clusters, and both Beijing and Shanghai are attempting to develop multiple high-tech activities. A dynamic cluster has several attributes. It has an entrepreneurial culture that leverages the resources of universities and firms; it benefits from the local presence of “angel� investors and venture capitalists who support and mentor local activities; it combines the advantages of specialization in key fields with an openness to new ideas; it has the capacity to learn from mistakes and to unlearn; and it has a “buzz� in national and global circles. History shows that many clusters formed accidentally because of a decision to locate an important facility (such as the National Aeronautics and Space Administration center in Houston, Texas), a university, or a firm that emerged as a major player in the industry (for example, Dell in Austin, Texas, and, perhaps, Huawei in Shenzhen). History further shows 130 Geography of Growth that these chance events might have foundered were it not for supporting initiatives taken by urban leaders and national governments. Policies to Support Clusters The supporting policies can take many forms. Strategic foresight exercises can assist governments in mapping out a long-term cluster development strategy and providing the stable long-term financing for R&D that research-intensive activities frequently require (for example, the backing by the National Institutes of Health has been critical to the success of biopharmaceutical research in the United States and the training of a legion of researchers). Complementing these are policies to ensure the supply of quality skills. For cities, policies that attract industry—domestic and foreign—need to be supplemented by policies that secure the city finances and ensure that services and housing meet the expectations of industries that are aware of and compare opportunities in other cities throughout the world. But providing services and infrastructure is not enough: cities must also market themselves aggressively by organizing events and seeking out business nationally and internationally. Such marketing is the most reli- able way of infusing capital and ideas into existing clusters and sowing the seeds of new clusters. Governments seek to stimulate the geographic dimension of innova- tion and industrial clustering by promoting technopoles, high-tech parks, bio-parks, and industry parks and by encouraging venture capital and financial services to be located in the vicinity of such parks. This is in line with the knowledge spillovers that Marshall ([1890] 1920) first envis- aged.1 Firms and areas in close proximity to vibrant areas will benefit and learn from each other, leading to increasing returns and further agglom- eration. Knowledge is transferred through interpersonal contacts and interfirm mobility of workers. Knowledge and innovation tend to spill over locally first and to diffuse geographically over time. The pace and extent of this reaction-diffusion varies geographically and depends on the stage of industry life cycle and the importance of tacit knowledge. Closer proximity of firms, which helps to lower transaction and transport costs, also contributes to this process of diffusion. Sustaining Clusters Learning and knowledge transfer occur through networking—social and economic—such as user-producer relationships, user associations, mobility Globalization, Urban Regions, and Cluster Development 131 of workers, spinoffs of new firms from larger, old firms, local libraries and information centers in the area, and so on. The ability of local firms to tap into such tacit knowledge depends on the existence of social links and open lines of communication in the area. In general, it is the creative atmosphere in the locality that contributes the most. New clusters are formed when a pool of skilled labor and university-trained human capital is available. In this context, there is an important role of diaspora, where returning human capital comes back, bringing some savings for risk taking and investment in new firms. For well-established clusters to continue to thrive, it is necessary to have an external link or source of knowledge and learning. Therefore, links to regional innovation systems or some new source of knowledge are necessary for the local area to upgrade itself. Localization and globalization coexist in high-tech clusters. Over longer distances, col- laboration between the Silicon Valley and Chinese provincial govern- ments and universities would benefit China. The Indian Institutes of Technologies benefited from their cooperation with U.S. firms and clusters. For newly formed clusters, the local areas need more intense places of interaction in the local area, and external links are less impor- tant for a while. How do multinational corporations that locate within a local cluster respond to such localization? Multinationals are no lon- ger the key vehicle for ICT development and diffusion. Instead, manag- ers and skilled human capital are the main actors in this process. Therefore, free mobility of these carriers of tacit knowledge is needed to sustain a cluster. Starting a cluster is entirely different from sustaining a cluster. Starting a cluster requires managerial skills, technical specialists, and access to technology and market opportunities. Empirical evidence indicates that governments should help to support specific new firms (say, specific tech firms) through industrial policy rather than invest large amounts in start- ing an industrial cluster. As entrepreneurs in emerging markets take the lead in research and innovation, there is increased recognition of their impact on global busi- ness structures. Previously, companies set up manufacturing units in Asia mainly for their cheap labor, while keeping the managerial functions in the West. However, in keeping with rising incomes and consumption in the emerging markets, the supply chains are now being altered. To under- stand these changes, there is a need to understand the reasons why some regions and countries are becoming the new innovation hubs, while other regions and countries are being left out. Initial research in developing 132 Geography of Growth countries indicates that the following factors are important for innova- tion clusters: • Urbanization. The rapid concentration of population in metropolitan cities reveals an increasing scope of opportunities that are attracting the young and educated to these hubs. • Educational improvement. A more educated working population attracts jobs requiring higher skills, including research and development skills. Thus, better education at the higher secondary and university levels will have a great influence on the level of innovation. • Business growth. A growing business sector indicates greater market interactions and, therefore, greater potential for innovation. • Macroeconomic factors. A stable and growing economy is necessary for innovative practices. Thus, indicators like gross domestic product and value added by a sector are important yardsticks for measuring the health of the economy in general. • Infrastructure improvement. A lack of infrastructural support would be an impediment to any business, including an entrepreneurial undertak- ing. Therefore, basic facilities like electricity, roads, transportation, and Internet access are necessary factors for innovative businesses, espe- cially start-ups. A review of 24 cities for which data are available indicates a positive correlation between agglomeration economies, population, and other indicators. For example, the percentage of urban population is positively correlated with the other indicators (figure 6.1). Conclusion Innovation clusters are important for economic development, locally, nationally, and internationally. Yet, too few exist. Mills, Reynolds, and Reamer (2008, 6) lament the “thin and uneven (presence of clusters) in levels of geographic and industry coverage, level and consistency of effort, and organizational capacity.� Furthermore, traditional clusters—for exam- ple, the automobile cluster in the Midwest—are under tremendous eco- nomic stress,2 which includes the individual worker, the supply chain, and the host community. Two key elements for successful innovation clusters are place and access to finance. Feldman (2009) notes that competitive advantage may be inherent to a certain locale. Seed and early-stage financing Globalization, Urban Regions, and Cluster Development 133 Figure 6.1 Ranking of Metropolitan Cities % of population in urban agglomeration 30 25 20 15 10 5 0 0 5 10 15 20 rank of city based on five factors rank of ranks linear (rank of ranks) Source: Authors based on data from the World Bank and Brookings Institution. capital is critical in order for innovative firms to be able to realize their ideas from inception through to market and commercialization. Policy makers can help. For example, the Obama administration has put forth an initiative describing a national innovation policy for the United States. Sallet, Paisley, and Masterman (2009) examine the innovation environment in the United States in which this national policy would operate. 1. In regions around the country, clusters of universities and high-tech companies partner with local and regional governments to boost tech- based economic growth and create jobs. The two best examples are Silicon Valley for computer technology and the Boston corridor for biotech development. 2. Job creation and business creation, the main economic benefits coming from innovative clusters, mostly spring from so-called “high-impact� companies (high-tech start-ups and established companies alike) that sell goods and services outside their clusters to both national and inter- national markets, drawing revenue back into the cluster (Acs, Parsons, 134 Geography of Growth and Tracy 2008). These “traded� services boost regional economic growth and national economic competitiveness. As measured by patent rates, productivity rates, and other innovation metrics, an innovation cluster creates new companies and new jobs in a helter-skelter but overall positive direction. 3. The federal government provides large sums of funding for basic scien- tific research and boasts a variety of programs to help companies and state and local governments to prepare executives and workers for employment at young, innovative companies seeking to commercialize this research. 4. The national innovation policy under consideration by the federal gov- ernment would link clusters with R&D firms and academic institutions, companies, and local and regional policy makers. The United States devotes 1 percent of the nation’s basic R&D budget to programs that support regional clusters, while Europe and China invest more. Notes 1. “Marshall ([1890] 1920) suggests that locations thick with similar activity generate valuable agglomeration economies for firms—namely, better access to skill labor (labor market pooling), specialized suppliers (shared inputs), and knowledge spillovers from competing firms� (Alcácer and Chung 2010, 1). 2. Automobile parts manufacturers told the Treasury Department early in 2009 that 130,000 jobs had been lost in 18 months (Economist 2009). References Acs, Zoltan J., William Parsons, and Spencer Tracy. 2008. “High-Impact Firms: Gazelles Revisited.� Study for the Office of Advocacy, U.S. Small Business Administration. http://www.sba.gov/advo/research/rs328tot.pdf. Alcácer, Juan, and Wilbur Chung. 2010. “Location Strategies for Agglomeration Economies.� Harvard Business School Working Paper 10-071, Harvard University, Cambridge, MA. Economist. 2009. “The American Car Industry.� The Economist, February 19. http://www.economist.com/business/displaystory.cfm?story_id=13145718. Feldman, Maryann. 2009. “Place Matters: Innovation Springs from Many Seeds, But Soil Is Equally Important.� Science Progress, Center for American Progress, Washington, DC, January. Marshall, Alfred. [1890] 1920. Principles of Economics: An Introductory Volume, 8th ed. London: Macmillan. Globalization, Urban Regions, and Cluster Development 135 Mills, Karen G., Elisabeth Reynolds, and Andrew Reamer. 2008. “Clusters and Competitiveness: A New Federal Role for Stimulating Regional Economies.� Policy Brief, Brookings Institution, Washington, DC. http://www.brookings .edu/reports/2008/04_competitiveness_mills.aspx. Sallet, Jonathan, Ed Paisley, and Justin R. Masterman. 2009. “Geography of Innovation: The Federal Government and the Growth of Regional Innovation Clusters.� Science Progress, Washington, DC, September 1. CHAPTER 7 Urban Development and Growth The world’s population crossed the 7 billion people mark in 2011, more than half of whom now make their home in a city.1 Each week, the ranks of urban residents are growing by 1 million, and on every single day some 20,000 new dwellings and 160 miles of road are added to the existing stock. China alone constructs 2 billion square miles of floor space each year, which is approximately half of the global total. By the middle of the century, demographers project a population of close to 9 billion, barring unexpected changes in fertility trends and unforeseen calamities;2 an estimated 70 percent of this vast number will rub shoulders in cities. More people and more cities are an inescapable part of the future, and if urban densities continue declining at about 2 percent a year, as they have through much of the twentieth century, the built-up area will expand at a far faster rate than the urban population. By one estimate, the urban population in developing countries could double by 2030, whereas the built-up area encompassed by cities would triple. Clearly, we and future generations are in for exciting times. The authors are greatly indebted to Lopamudra Chakraborti for assistance with the research for this chapter. 137 138 Geography of Growth The importance of cities predates the industrial revolution. Ancient civilizations arose in urban settings, starting with the earliest cities germi- nating in the marshy areas beyond Baghdad.3 Greek civilization would be a desert if it were emptied of Athens, Corinth, Sparta, Thebes, and other cities. The Roman Empire was a “world of cities,� with Rome and later Constantinople as its political, administrative, and cultural axes.4 Islamic civilization, the Renaissance, the glory of China under the Sung and Ming dynasties, the remarkable architectural achievements of the Mughals, and the later rise of capitalism are all inextricably linked to cities.5 Abstract from the urban context, drain out the technological, intellectual, political, economic, and artistic achievements that flowered in cities, and most of the richness of history simply melts away. The industrial revolution gave cities added prominence by enormously enlarging their economic signifi- cance. Agriculture and rural industry, long the economic heartland of nations, was displaced in a matter of decades by the concentration of eco- nomic power in cities, which were quick to exploit the potential of steam and the technologies that transformed the textile, metallurgic, machine building, chemical, and other industries starting in the mid-nineteenth century. European countries that embraced industrialization experienced rapid urbanization and the transfer of the economic center from the rural to the urban sector. Henceforth, national wealth was increasingly derived from manufacturing industry powered by fossil fuels. In nations where modern industry was slow to gain traction or did not take root at all, urban development was much feebler. With the widening use of steam power, cities became even more attractive because the factory-based manufactur- ing industry needed pools of labor (especially female workers),6 markets to absorb increased output (local, national, and international), and sup- porting infrastructure and services. Cities could provide all of these rela- tively efficiently and cheaply, and they simplified the logistics of input supplies, reduced the cost of intermediate goods, and facilitated the distri- bution of products to other markets. Until well into the 1960s, the growth and dynamism of cities in Western countries and Japan were paced by manufacturing activities.7 Thereafter, the role of industry as the leading sector was displaced by services, and the character of urbanization began inexorably to change. From the late 1950s onward, many more countries, many of which had recently gained independence, began pursuing development along Western lines by emphasizing industrialization. Assisted by tariff protec- tion and other government-provided incentives,8 manufacturing indus- tries, frequently state owned, were established in the primary cities, with Urban Development and Growth 139 the capital city, the seat of administrative authority, being the preferred location. This led to the emergence of new business and professional classes that quickly allied themselves with administrative and military elites controlling the state and responsible for making policy and distrib- uting rents. Thus began the concentration of wealth and power9 in the primary urban centers of developing nations, mimicking to some degree similar trends in the industrial countries, with an important difference: urbanization took off in the developing world without industry providing the main impetus in many cities. The research since the 1960s shows that urbanization is closely correlated with industrialization, but industry does not cause urbanization (Henderson 2010), as it arguably did between 1850 and 1960 in Western countries. What then explains the surge since 1950 that has carried the urbanization rate from less than 30 percent to more than half of the global population in 2010? Urbanization: From Canter to Gallop Five factors account for accelerating urbanization, and its structural characteristics and their persistence determine its dynamics, challenges, and policy implications, which will be discussed in the balance of this chapter. First, the demographic transition—a sharp decline in infant mortality and increasing life expectancy, followed by a much more gradual reduc- tion in fertility—has resulted in a ballooning of populations in developing nations. The natural increase has caused cities to grow and led to in situ urbanization—small towns and villages have mushroomed into cities in Brazil, China (Zhu et al. 2009), and Pakistan, for example. Brazil, in par- ticular, achieved European rates of urbanization by 2000.10 Greater rural population densities have pushed people to migrate, and the “urban advantage� (a point emphasized by UN-HABITAT 2010) and income gradients have exerted a parallel pull.11 With population pressures rising, cities are seen as beacons of opportunity that are disappearing in rural areas. And urbanization has been correlated with rising living standards, although inevitably the transfer of populations has led to rising rates of poverty (Ravallion, Chen, and Sangraula 2007). Those living on less than US$1 a day in urban areas rose from 19 to 24 percent between 1993 and 2002; over the same period, the urban share of the population as a whole rose from 38 to 42 percent. The urbanization of poverty was most rapid in Latin America, with the proportion of the poor living in urban areas rising from 50 to 60 percent between 1993 and 2002. By contrast, less 140 Geography of Growth than 10 percent of East Asia’s poor live in urban areas, largely because absolute poverty in China is overwhelmingly rural. Second, agricultural production is becoming less labor intensive, with machinery, chemicals, and energy serving as substitutes.12 Fewer hands are needed on farms, and if the highly productive agricultural systems in advanced economies are a mirror of what developing economies can expect, the share of the agricultural labor force in low- and middle- income countries will drop below 10 percent of the national total from the average of about 25 percent in 2007. Furthermore, dispersed small- scale rural industry, inefficient and polluting as it is,13 fights a losing battle against urban producers, which enjoy manifold advantages compounded by declining costs of surface transport and increasing efficiencies in distri- bution and marketing technologies. Third, technological advances and the evolving income elasticity of demand are responsible for structural changes that have enlarged the role of services. A stream of innovations has raised the productivity of manufacturing,14 contributing to growth, but also to a decline in relative prices of manufactures and employment in industry (table 7.1 shows the fall in the share of manufacturing between 1980 and 2008). Thus, the share of manufacturing in gross domestic product (GDP) is a shrinking proportion of the output in larger cities, although it remains high in smaller cities with industrial specializations. Meanwhile, rising demand for urban services and much slower gains in productivity have increased the share of urban services in GDP and employment. With the exception of China, services now dominate GDP everywhere,15 and in most cities in advanced countries, services provide the majority of jobs and generate more than half of the income. In fact, with industry pushed to the mar- gins of some urban economies, services are the economy. A fraction of services are tradable, but most urban services in developing countries are nontradable, and services constitute a small share of the exports of low- and middle-income countries, tourism being the largest.16 This has Table 7.1 Contribution of Manufacturing and Services to GDP, 1980–2008 Manufacturing value Services value added added (% of GDP) (% of GDP) 1980 2008 1980 2008 World 25 17 56 70 Middle income 26 21 41 54 Low income 12 13 42 49 Source: World Bank 2011a. Urban Development and Growth 141 long-term implications for the number and type of jobs that the urban economy is likely to create, for growth, and for exports to balance the city’s trade accounts. To be viable over the longer term, cities—much like countries—must have something to sell, with any shortfall being offset through capital transfers. Until a few decades ago, all growing cit- ies were industrial cities with export potential. This has ceased to be the rule with the rise of formal and informal services,17 which suspends the question over the future of urban centers that depend on transfers for their survival. Fourth, cities enable firms to specialize and realize scale advantages. These so-called localization economies are an important asset for mid- size industrial cities and a source of productivity gains from labor mar- kets, technological spillovers, and the benefits of clustering (proximity to other producers and suppliers of services). For larger urban centers, urbanization economies are more prominent. These are the economies arising from the multiplicity of industry and services that open the door to diversification and induce the entry of new firms. Together, these lead to significant productivity gains and higher average incomes (see figure 7.1 on the relationship between city size, industrial composition, city clustering, and incomes in China). Currid (2007, 460) notes,“Agglomeration may be even more important to maintaining the social mechanisms by which the cultural economy sustains itself [through nonmarket transac- tions].� A vast literature, mostly on cities in developed countries, has attempted to estimate the gains from agglomeration, whether from local- ization or urbanization or scale economies.18 Researchers differ on which gains matter more, but all agree that agglomeration pays, although how much productivity can be traced to size and diversity varies from 3 to 12 percent.19 A meta-analysis of elasticities drawn from 34 studies cau- tions that the gains from largeness should not be exaggerated (see Melo, Graham, and Noland 2009), but there is little or no evidence indicating that growth is disadvantageous for cities. Nevertheless, casual empiricism suggests that, as cities grow larger and become more complex, manage- ment and services provision become increasingly more challenging, and congestion, pollution, and crime diminish the quality of life, as, for instance, in Bangalore and many booming cities in China’s Pearl River Delta. Whether these collectively erode the productivity-enhancing advantages of size has been difficult to establish, and the debate on the merits of largeness continues.20 This point is examined further below. The fifth and final factor contributing to the vigor of urbanization is the role of cities in sparking ideas, stimulating social change by inculcating Figure 7.1 Strong Correlates of Urban Productivity (City GDP per Capita) in China, 2007 142 a. Average income by city size b. Average income by city manufacturing base average income (2007) by city manufacturing 100 80 average income (2007) by city 80 share (2000), in thousands size, yuan, in thousands 60 60 40 40 20 20 0 0 <800K 800K –1.3M 1.3M –2.2M 2.2M –8M >8M 1: low manufacturing 2 3 4 5: high manufacturing excludes ouside values excludes outside values average income (2007) by proximity to Beijing, Shanghai, c. Average income by urban clustering within 50 kilometers d. Average income by proximity to Beijing; Shanghai; and Hong Kong SAR, China average income (2007) by urban clustering and Hong Kong SAR, China, in thousands 100 100 within 50km (2000), in thousands 80 80 60 60 40 40 20 20 0 0 1: low urban cluster 2 3 4 5: high urban 1: close to Beijing; 2 3 4 5: remote excludes outside cluster Shanghai; Hong Kong SAR, from Beijing; Shanghai; values China Hong Kong SAR, China excludes outside values Source: Lall and Wang 2011. Note: Average city income is measured by GDP per capita. Urban Development and Growth 143 new values, and encouraging innovation in every sphere of life. Steven Johnson (2010, 16, 162) compares cities in all their variegated complexity to coral reefs “powerfully suited to the creation, diffusion, and adoption of good ideas. … They cultivate specialized skills and interests, and they create a liquid network where information can leak out of those subcul- tures and influence their neighbors in surprising ways. This is one reason for superlinear scaling in urban creativity.�21 Such innovation has buoyed productivity; equally, it has enhanced human capabilities and raised the quality of life. As cities in developing countries attempt to come to grips with increasing size, complexity, and the pressures arising from climate change, the innovative potential of cities will become ever more impor- tant and the basis not just of survival but also of prosperity. While continued urbanization appears to be a given, urban develop- ment is likely to evolve in different directions, with implications for growth and the quality of life. From the perspective of this book, the interesting issues pertain to the potential of the metropolitan model of urban development and how creatively metropolitan centers address the challenges coming from many directions. The Metropolitan Powerhouse Megacities with populations of 10 million and more have increased in number from 9 in 1985 to 23 in 2010 and account for almost half of the world’s wealth.22 Moreover, some of the megacities in East Asia account for a third or more of national GDP. A striking characteristic of the urban- izing tendencies in East Asia, Latin America, and the United States is the emergence of metropolitan regions composed of a cluster of cities that may or may not include a megacity. Bangkok, Jakarta, and Seoul are examples of metropolitan economies in which the core primary city has brought (or created) dormitory, secondary, and edge cities into its orbit. Guangzhou and Hong Kong SAR, China, encompass another vast metropolitan region that arose with great rapidity once China adopted the Open Door Policy in 1979, and industry began transferring from Hong Kong SAR, China, to the Pearl River Delta.23 The Washington, DC, metro region and others in the United States and in Europe are examples of networked city flotillas. For many reasons, urbanization might take the form of the metro region in the future, with isolated cities becoming endangered species.24 It is a commonplace that urban development flourishes in certain geo- graphic locations. In the United States, more than half of the population live within 50 kilometers of a coastline, and mild climates attract people.25 144 Geography of Growth A prosperous hinterland, space for a city to expand, and adequate sup- plies of water are other geographic considerations.26 Some of the world’s largest cities have been established at strategic points on riverine plains and close to river deltas, locations that facilitate the transport of goods. Most of the choice spots are taken, and because the availability of fresh- water and climatic considerations may bulk larger, urbanization will most likely concentrate around the most promising existing centers, although rising sea levels will imperil several low-lying coastal regions and cities and some of the 360 million people who currently live in these areas (figures 7.2 and 7.3 show the countries in East Asia with the most vulner- able populations; China is in the forefront, followed by Indonesia). The need to economize on energy use and on the cost of providing urban infrastructure makes the metropolitan model, with its compact design, a more viable proposition than a relatively isolated city (Glaeser 2011). The metropolis can also internalize urbanization and localization econo- mies by combining a portfolio of cities in a single urban domain. The core city with diverse services and the advanced emerging industries that Figure 7.2 Exposure of People to Cyclones and Earthquakes, 2000 and by 2050 Source: Jha and Brecht 2011 (adapted from National Disaster, UnNatural Disasters: The Economics of Effective Prevention, World Bank and United Nations, 2010). Urban Development and Growth 145 Figure 7.3 Coastal Population of Selected Countries That Are Highly Vulnerable to Sea Level Rise 90 80 70 Population (millions) 60 50 40 30 20 10 0 China Indonesia Vietnam Thailand Philippines Source: Jha and Brecht 2011, adapted from Prasad et al. 2009. draw oxygen from proximity to centers of research can be the primary source of urbanization or Jacobs economies, while smaller peripheral specialized cities can serve as sites for industrial activities requiring cheaper land for factories and cheaper accommodations for workers. By yoking these different kinds of cities together with an efficient multi- modal transport system, the metropolitan region can maximize the gains from agglomeration and market size economies. By expanding in the vertical plane, it can also squeeze a lot more people into a place with proven locational advantages and capitalize on an existing foundational infrastructure and possibly a brand name. A broad economic base and a large urban market make it easier for a metropolitan region to meet its financing needs and minimize fluctuations in revenue streams, while keeping tax rates at moderate and competitive levels. Revenue adequacy underwrites industrial capabilities and provides the means for a city to adapt to changing circumstances, calling for the displace- ment of older industries by newer ones and the renewal of infrastructure and buildings so as to incorporate the latest technologies (such as information and communication technologies and green technologies enabling buildings to use less water and energy) and accommodate changing lifestyles.27 “Green technologies� is a loose term currently embracing a range of technologies aimed at conserving nonrenewable resources, controlling and minimizing waste, and squeezing greater productivity out of available resources through new technological fixes. Green technologies come in many flavors, such as materials that reduce the weight and facilitate the recycling of auto bodies and parts; technologies that substitute nontoxic 146 Geography of Growth for toxic materials and cut down waste; sensor technologies that help to smooth traffic flows, provide early warning of impending infrastructure breakdowns or disasters, improve land productivity by monitoring soil moisture content, and detect toxic pollutants; software that helps to man- age smart grids and energy use throughout the economic system and software that simplifies equipment and cuts capital costs, for example, of diagnostic and measuring devices; and design innovations that lessen energy and water consumption in cities and buildings or allow for effi- cient disposal, salvage, recycling, and longevity of products (Tomlinson 2010). The list goes on and includes manufacturing technologies that are energy frugal and reduce waste, technologies for the construction and food-processing industries—both vast consumers of resources—and new technologies for the information technology industry itself, a major user of energy (for production and use of products) and rare metals and a generator of toxic waste (see Fiksel 2009). Several of these technolo- gies—in particular, those associated with transport, energy generation, food production and processing, and construction and maintenance of infrastructure—give rise to dense links with manufacturing and services activities. Replacing the global stock of automobiles and internal combus- tion engines with hybrid and electric vehicles with smaller direct carbon signatures has enormous implications for industry. So does the building of equipment for urban public transport. The three biggest sources of greenhouse gases (excluding humans, rice cultivation, and cattle) are power plants, transport equipment, and buildings. If the vast majority needs to be replaced to minimize climate change and “green� becomes the order of the day, manufacturing industry will have to take on a chal- lenge. And once green is “it,� every other activity will be affected, requir- ing redesign, software retooling, and change in the structure of industry. No metropolitan region ever optimizes on all of these fronts, and when there are many adjacent municipal jurisdictions, coordinating infrastructure development, embarking on revenue-raising arrangements, and sharing financial burdens can be severely challenging. By failing to arrive at coher- ent and mutually advantageous outcomes through negotiated give-and- take, multijurisdictional metropolitan entities are squandering the economic and financial benefits of agglomeration. Metropolitan Challenges Too many cities in advanced and developing countries are failing to exploit the urban advantage and are struggling to cope with the growth Urban Development and Growth 147 of populations and the associated crowding, pollution, traffic, shortages of housing and services, increasing poverty and inequality, spread of slums, and environmental degradation.28 Very few cities in developing countries are able to generate enough jobs for the growing workforce, and unem- ployment is endemic. Where economic performance falls short of poten- tial or where revenue effort is weak, urban services suffer, which affects business activity and quality of life, especially of the poor. Most cities have barely begun to come to grips with the physical and institutional changes required to contain greenhouse gas (GHG) emissions29 and to engineer the resilience demanded by the threat of climatic extremes.30 If a dou- bling of urban populations and an increase in average temperatures by 2° or more are inevitable by mid-century, then delay is becoming increas- ingly costly.31 For cities in most developing countries, certain inescapable facts demand an adequate response—if not immediately, then definitely in the not too distant future—if cities are to reap the urban advantages and sus- tain their long-term viability. Demographic trends and a youth bulge, most notably in the Middle East and South Asia, will necessitate employment- augmenting policies to maintain adequate growth in incomes and social stability.32 For an expanding global economy, energy and resource scarci- ties will be mounting concerns requiring a change in urban design, in modes of transport, and in soft and hard infrastructure. And climate change will expose cities to pressures and shocks rarely experienced before. Few cities will be spared, and many coastal and semiarid locations may only remain habitable through major injections of capital.33 It is possible for metropolitan centers to thrive and for even the most vulnerable to avoid being plunged into vicious spirals leading to steadily worsening conditions. Inevitably, there is no infallible recipe, no sufficient conditions to assure success. However, the collective experience of scores of urban centers, many of which have embarked on innovative policies and introduced new technologies, provides reliable guidelines on how to create a dynamic metropolitan region that would provide most inhabitants with jobs and a decent life. Wealth of Cities and National Policies If cities are truly the drivers of economic growth, how closely is that performance keyed to the national policy and overall national economic conditions? In other words, can cities forge ahead by dint of good urban policies more or less independently of what is happening at the national 148 Geography of Growth level? Singapore surely fits this description, being a city-state, but other cities, even the largest and most prosperous such as Bangkok, São Paulo, Seoul, Shanghai, Tokyo, and the complex of Guangzhou and Hong Kong SAR, China, depend on the enabling matrix of national trade, investment (domestic and foreign),34 fiscal, education, and other policies to provide the springboard for their own development. Even though decentraliza- tion and localization have transferred more administrative and fiscal dis- cretion and policy initiative to subnational governments, and even though cities are at the leading edge of development, fundamental national poli- cies define policy parameters, incentives, and the degrees of freedom available to city managers and determine the fiscal and financial resources they can mobilize. The industrialization of Seoul and Shanghai was enabled by planning, day-to-day decision making conducted by city authorities, and a host of local regulations, rules, standards, and licensing requirements, but the opportunities for the cities were delineated and circumscribed by exchange rate, trade, industrial, labor, education, and technology policies of the central government. Both cities successfully groomed highly competitive export industries that generated economic momentum and employment and catalyzed the development of other sectors of the urban economy. Seoul took a lead in establishing a world- class infrastructure to harness the potential of information and commu- nication technologies (ICTs), with Shanghai now close behind. These measures initiated the process of modernization and integration with the global economy, and the end result as of now are two metropolitan economies that rank among the world’s most vibrant. However, in both instances—and these examples can be multiplied—urban outcomes were prompted and shaped by national policies. The government of the Republic of Korea, once it had embraced export-oriented industrializa- tion, viewed Seoul as the engine room of the economy,35 and urban development complemented other policies—more recently, policies to develop an ICT-supported knowledge economy. The industrialization of the Seoul metro region propelled the Korean economy during the high- growth era starting in the mid-1960s and continues to do so as Korea enters a postindustrial stage. Seoul not only has served as the seat of gov- ernment and the nation’s cultural hub, but also is home to several of Korea’s leading export industries, including textiles, machinery, elec- tronics, and now the creative industries.36 Once China set its sights on reform and catching up with the leading East Asian economies and designated Shanghai as one of the principal Dragonheads,37 the city authorities had the green light to pursue an ambitious urban industrial Urban Development and Growth 149 strategy that was amply supported by resources from the central gov- ernment and supplemented by foreign direct investment (FDI) induced through central policies reinforced by municipal incentives. Shanghai’s development since the early 1990s is the stuff of legend, and it owes much to the vision and energy of a succession of local officials,38 but it was the central government that loosened the corsets binding Shanghai, encouraged the local authorities to raise their sights, and created the policy environment that allowed the city to harness its vast latent capa- bilities and bid for resources from elsewhere in China and from abroad. It is the central government that sets the stage and to a greater or lesser extent choreographs, through policies and other interventions, urban development in positive as well as negative directions. Where central governments are missing in action, passive, obstructive, or predatory, urbanization may continue, as it has in Sub-Saharan Africa and in South Asia, but the urban development that results in growth, exports, and jobs may be slow to materialize, if it surfaces at all. Some cities in Africa, such as Dar es Salaam and Kinshasa, have become more populous over the past decade, but not because of development. Urbanization in Zimbabwe and the Democratic Republic of Congo is the direct outcome of conflict and worsening conditions in rural areas. Development has gone into reverse because the states have faltered or are failing (see World Bank 2011b). Thus, the policy making and administrative capabilities of the state and its urban strategy broadly define the opportunities for urban development. Some cities, especially capitals, are favored over others, and they have a head start; however, it is up to the municipal authorities and other stakeholders to derive maximum benefits from the urban assets at their disposal, to enhance competitive advantage in profitable directions, to augment the local resource base, and to encourage investment that can maximize long-run growth. Notes 1. Annual births average 140 million, and deaths average 57 million, leading to a net population gain of 83 million each year. 2. The most recent projections point to 9.3 billion people by 2050 (UN 2011). According to John Bongaarts (2011), the margin of error for 2050 estimates could be plus or minus 1 billion. 3. An aerial remote-sensing investigation conducted by Jennifer Pournelle, a student of Robert McCormick Adams, suggests that settlements were con- structed on small ridges in marshy areas called “turtlebacks� (Pournelle 2007, 150 Geography of Growth 35), and these eventually grew into cities such as Eridu and Uruk—the birth- place of writing; Vanderbilt (2011a); McCormick Adams (1966). 4. The speeches of Greek and Roman orators are laced with praise for their cities, and such praise was often modeled on praise of individuals. The rec- ognition accorded to the fallen hoplites in Pericles’s “Funeral Oration� is “subsumed into an account of the moral and political virtues of the city of Athens� (Grafton, Most, and Settis 2010, 202). Price and Thoneman (2010) observe that there were more than 300 cities in the Asian part of the Roman Empire alone and that the empire in its entirety contained several thousand cities. 5. Bosker, Buringh, and van Zanden (2008) attribute the cultural and commer- cial retreat of Arab cities from the heights they had scaled through the twelfth century and their lagging performance thereafter relative to European cities to the autonomy and “producer� orientation of the European urban centers in northern Italy, for example, and the grip of predatory states on the Arab and Islamic cities, which became oriented toward “consumption.� 6. Kim (2005). Immigration to the United States contributed to the growth of the cities and supplied the workforce for industrialization. 7. The share of manufacturing production in the United States peaked in 1979. However, cities such as New York and Philadelphia had entered the spiral of deindustrialization in the 1950s. New York’s garment industry, which invented ready-to-wear clothing, gave birth to the Singer sewing machine and accounted for 70 percent of all women’s clothing and 40 percent of mens- wear in the United States in 1910, was battered by the rise of low-cost pro- ducers in East Asia and by the revolution in telecommunications and in transport. The fact that it survives at all is because some products governed by tight schedules require close face-to-face coordination among suppliers, service providers, designers, and those who actually sew the various parts of a garment (Vanderbilt 2011b). 8. Most countries adopted the strategy of import-substituting industrialization. 9. In some countries, land reform accelerated the transfer of power by disman- tling feudal systems of privilege, wealth, and political control, for example, in the Republic of Korea. In others, such as Pakistan, the political sway and social influence of the landowning class has eroded much more slowly. 10. Brazil’s urban population rose from 36 percent in 1950 to 77 percent in 1990. 11. This is the so-called Harris-Todaro effect of higher urban incomes. See Fields (2007). 12. See Smil (2008, 2011) on the energy (and nitrogen fertilizer) intensity of modern agriculture. Urban Development and Growth 151 13. Township and village enterprises blossomed in China with the spread of agri- cultural reforms in the early 1980s. By 1996, they accounted for 26 percent of China’s GDP and employed 30 percent of the rural workforce. But growth slowed thereafter, as urbanization began pulling industry away from rural locations. 14. Most striking is the decline of employment in high-tech manufacturing in the United States between 1990 and 2008. 15. Between 1977 and 2007, the share of services in global GDP rose from 55 to 70 percent, and it rose to 75 percent in the Organisation for Economic Co-operation and Development countries (Francois and Hoekman 2010). 16. See Eichengreen and Gupta (2009, 2011) on the role of services with refer- ence to India; Ghani (2010) on how growth in India could continue to be propelled by services; and Spence and Hlatshwayo (2011) on the contribu- tion of nontradable services to the bulk of the employment created in the United States since 1990. 17. In 2007, the global value of cross-border trade in services amounted to US$3.3 trillion or about a fifth of total trade. However, the share is closer to 50 percent when measured by direct and indirect value added (Francois and Hoekman 2010). Its growth is impeded by regulatory restrictions and by the greater protection accorded to services. 18. Gill and Goh (2010); Glaeser and Gottlieb (2009); Rosenthal and Strange (2004); World Bank (2008). Geoffrey West (2010) compares large cities to big animals whose size is a source of scale economies. When a city doubles in size, the resources required to sustain it grow 85 percent. See Lehrer (2010). 19. Rosenthal and Strange (2007) note that a doubling of city size can lead to an increase in productivity of between 3 and 8 percent. 20. Cohen (2004) presents data underlining the inexorable increase in average city size over the past two centuries. In 1800, the largest 100 cities in the world had an average population of 200,000. By 1990, the population of the top 100 had risen to 5 million. Beijing was the only city with 1 million inhab- itants at the beginning of the nineteenth century, and 100 years later only 16 cities had attained this size. By 1950, their number had swelled to 86. 21. Superlinear scaling refers to a faster than exponential rate of increase. Thus, as cities grow, according to Geoffrey West and his coworkers at the Santa Fe Institute, such superlinearity is evident in telecommunications traffic, patent- ing, and pedestrian speed. See SENSEable City Lab MIT (2009). 22. UN-HABITAT (2010) points to the emergence of the mega region—an end- less city. However, the bulk of the urban population resides in mid-size and small cities. 152 Geography of Growth 23. See McGee et al. (2007) on the rise of the region comprising Guanghzhou and Hong Kong SAR, China; Berger and Lester (1997) on the transfer of industry from Hong Kong SAR, China, to emerging cities on the mainland of China. 24. Eventually some of these will end up as ghost towns, as younger people migrate, revenues decline, services atrophy, and infrastructure deteriorates. 25. The concentration of people in coastal areas is described by Rapaport and Sachs (2003). The migration of the U.S. population toward the south and away from colder areas is described by Glaeser (2011). 26. The contribution of geography to city formation in Europe is analyzed in depth by Bosker and Buringh (2010). 27. Smaller household size, increasing numbers of older people, and the explosion in relational networking are among the factors influencing lifestyles and demands on urban infrastructure and services. Per capita consumption of energy is greater in smaller households. http://www.statcan.gc.ca/pub/11 -526- s/2010001/part-partie1-eng.htm. 28. Inequality is greatest in African cities (Gini coefficients of 0.58), but it is ris- ing most rapidly in Asia (UN-HABITAT 2010). Although the percentage of those living in urban slums is estimated to have declined—from 39 to 32 percent between 2000 and 2010—the absolute numbers have risen. If current trends continue, almost 900 million people will be living in slums by 2020 (UN-HABITAT 2010). 29. Cities account for 80 percent of all GHG emissions, with the top 50 cities releasing 2.6 trillion tons of GHGs per year (Oxford Analytica 2011). 30. The topic of urban resilience has brought forth a considerable literature. See Newman, Beatley, and Boyer (2009); World Bank (2008); see the ICLEI, Local Governments for Sustainability website: http://resilient-cities.iclei.org /bonn2011/about/. 31. Heat island effects will only exacerbate the problem for cities, a foretaste of which was experienced by Chicago in 1995 and by Europe in 2003. 32. Recent unrest in the Middle East, sparked by unemployment and growing inequality of incomes and opportunities, has demonstrated the seriousness of the challenge. 33. In some instances, this will include expenditures on infrastructure for aug- menting the water supply with the help of transfers from other parts of the country, as in China, and desalination of seawater. 34. FDI is an important source of capital and technology transfer for industrial- izing countries and is likely to remain a vital conduit. Singapore was the leading urban recipient of FDI projects in 2009, followed by Shanghai, London, and Dubai. In Latin America, Bogotá, Mexico City, and São Paulo led the field. See FDI Intelligence (2011). Urban Development and Growth 153 35. Even though the Korean government was painfully aware of Seoul’s vulnera- bility to an attack from the Democratic People’s Republic of Korea given that it was just 30 miles from the demilitarized zone, it acknowledged and exploited the city’s strategic location and long-standing role in the national economy. 36. These include online video games, multimedia, and publishing. See World Bank (2008); Yusuf, Nabeshima, and Yamashita (2008). 37. Its past history made Shanghai a logical choice. See Yusuf and Nabeshima (2006, 2010); Yusuf and Wu (1997). 38. Some of the mayors who contributed to Shanghai’s resurgence were Wang Daohan (mayor, 1981–85); his protégé and successor, Jiang Zemin (mayor, 1985–89, and later Party chief of Shanghai); and Zhu Rongji (mayor, 1989–91). References Berger, Suzanne, and Richard Lester. 1997. Made by Hong Kong. New York: Oxford University Press. Bongaarts, John. 2011. “One Minute with...� NewScientist, April 2, p. 40. Bosker, Maarten, and Eltjo Buringh. 2010. “City Seeds: Geography and the Origins of the European City System.� CEPR Discussion Paper 8066, Centre for Economic Policy Research, London. Bosker, Maarten, Eltjo Buringh, and Jan Luiten van Zanden. 2008. “From Baghdad to London: The Dynamics of Urban Growth in Europe and the Arab World, 800–1800.� CEPR Discussion Paper 6833, Centre for Economic Policy Research, London. Cohen, Barney. 2004. “Urban Growth in Developing Countries: A Review of Current Trends and a Caution Regarding Existing Forecasts.� World Development 32 (1): 23–51. Currid, Elizabeth. 2007. “How Art and Culture Happen in New York.� Journal of the American Planning Association 73 (4): 454–67. Eichengreen, Barry, and Poonam Gupta. 2009. “The Two Waves of Service Sector Growth.� NBER Working Paper 14968, National Bureau of Economic Research, Cambridge, MA. ———. 2011. “The Service Sector as India’s Road to Economic Growth.� NBER Working Paper 16757, National Bureau of Economic Research. Cambridge, MA. FDI Intelligence. 2011. “Manufacturing Makes a Comeback: FDI Global Outlook Report 2011.� FDI Special Report, Financial Times, London, April–May. Fields, Gary S. 2007. “The Harris-Todaro Model.� Working Paper 21, Cornell University, Ithaca, NY. http://digitalcommons.ilr.cornell.edu/working papers/21. 154 Geography of Growth Fiksel, Joseph. 2009. Design for Environment: A Guide to Sustainable Product Development. New York: McGraw Hill. Francois, Joseph, and Bernard Hoekman. 2010. “Services Trade and Policy.� Journal of Economic Literature 48 (3): 642–92. Ghani, Ejaz, ed. 2010. The Service Revolution in South Asia. New York: Oxford University Press. Gill, Indermit S., and Chor-Chung Goh. 2010. “Scale Economies and Cities.� World Bank Research Observer 25 (2): 235–62. Glaeser, Edward L. 2011. Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter, Greener, Healthier, and Happier. New York: Penguin Press. Glaeser, Edward L., and Joshua D. Gottlieb. 2009. The Wealth of Cities: Agglomeration Economies and Spatial Equilibrium in the United States.� Journal of Economic Literature 47 (4): 983–1028. Grafton, Anthony, Glenn W. Most, and Salvatore Settis. 2010. The Classical Tradition. Cambridge, MA: Harvard University Press. Henderson, J. Vernon. 2010. “Cities and Development.� Journal of Regional Science 50 (1): 515–40. Jha, Abhas, and Henrike Brecht. 2011. “Building Urban Resilience in East Asia.� Eye on East Asia and Pacific 8, Washington, DC, World Bank. http://siteresources .worldbank.org/INTEASTASIAPACIFIC/Resources/226262-129112673 1435/EOEA_Abhas_Jha_April2011.pdf. Johnson, Steven. 2010. Where Good Ideas Come From: The Natural History of Innovation. New York: Penguin Group. Kim, Sukkoo. 2005. “Industrialization and Urbanization: Did the Steam Engine Contribute to the Growth of Cities in the United States?� Explorations in Economic History 42 (4): 586–98. Lall, Somik, and Hyoung Gun Wang. 2011. “China Urbanization Review: Balancing Urban Transformation and Spatial Inclusion.� Eye on East Asia and Pacific 6, World Bank, Washington, DC. Lehrer, Jonah. 2010. “A Physicist Solves the City.� New York Times, December 19. http://www.nytimes.com/2010/12/19/magazine/19Urban_West-t.html. McCormick Adams, Robert. 1966. The Evolution of Urban Society: Early Mesopotamia and Prehispanic Mexico. Chicago: Aldine Publishing. McGee, Terry, George C. S. Lin, Mark Wang, and Andrew Marton. 2007. China’s Urban Space: Development under Market Socialism. Routledge Studies on China in Transition. London: Routledge. Melo, Patricia C., Daniel J. Graham, and Robert B. Noland. 2009. “A Meta- Analysis of Estimates of Urban Agglomeration Economies.� Regional Science and Urban Economics 39 (3): 332–42. Urban Development and Growth 155 Newman, Peter, Tim Beatley, and Heather Boyer. 2009. Resilient Cities: Responding to Peak Oil and Climate Change. Washington, DC: Island Press. Oxford Analytica. 2011. “International: Cities Spearhead Climate Initiatives.� Oxford Analytica, Oxford, March 24. http://www.oxan.com/display.aspx?Item ID=ES167025. Pournelle, Jennifer. 2007. “KLM to CORONA: A Bird’s Eye View of Cultural Ecology and Early Mesopotamian Urbanization.� In Settlement and Society: Essays Dedicated to Robert McCormick Adams, ed. E. C. Stone, 29–62. Cotsen Institute of Archaeology, University of California, Los Angeles; Oriental Institute of the University of Chicago, Chicago, IL. Prasad, Neeraj, Federica Ranghieri, Fatima Shah, Zoe Trohanis, Earl Kessler, and Ravi Sinha. 2009. Climate Resilient Cities: A Primer on Reducing Vulnerabilities to Disasters. Washington, DC: World Bank. Price, Simon, and Peter Thoneman. 2010. The Birth of Classical Europe: A History from Troy to Augustine. London: Penguin. Rapaport, Jordan, and Jeffrey D. Sachs. 2003. “The United States as a Coastal Nation.� Journal of Economic Growth 8 (1): 5–46. Ravallion, M., S. Chen, and P. Sangraula. 2007. “New Evidence on the Urbanization of Global Poverty.� Population and Development Review 33(4): 667–701. Rosenthal, Stuart S., and William C. Strange. 2004. “Evidence on the Nature and Sources of Agglomeration Economies.� In Handbook of Urban and Regional Economics, Vol. 4, ed. J. Vernon Henderson and J. F. Thisse, 2119–71. Amsterdam: Elsevier. ———. 2007. “The Micro-Empirics of Agglomeration Economies.� In A Companion to Urban Economics, ed. R. J. Arnott and D. P. McMillen. Oxford: Blackwell Publishing. SENSEable City Lab MIT (Massachusetts Institute of Technology). 2009. “City Gravity: Visualizing Communications Traffic across the US Reveals Surprising Patterns of Innovation and Urban Growth.� Seed (February): 24–26. http:// www.carloratti.com/publications/on_us/20090101SeedMagazine.pdf. Smil, Vaclav. 2008. Energy in Nature and Society: General Energetics of Complex Systems. Cambridge, MA: MIT Press. ———. 2011. “Nitrogen Cycle and World Food Production.� World Agriculture 2: 9–13. Spence, Michael, and Sandile Hlatshwayo. 2011. “Evolving Structure of the American Economy and the Employment Challenge.� Working Paper, Council on Foreign Relations, New York. Tomlinson, Bill. 2010. Greening through IT: Information Technology for Environmental Sustainability. Cambridge, MA: MIT Press. 156 Geography of Growth UN (United Nations). 2011. “World Population to Reach 10 Billion by 2100 if Fertility in All Countries Converges to Replacement Level.� Press Release, United Nations, New York, May 3. http://esa.un.org/unpd/wpp/other- information/Press_Release_WPP2010.pdf. UN (United Nations)-HABITAT. 2010. State of the World’s Cities 2010/2011: Cities for All; Bridging the Urban Divide. Nairobi: UN-HABITAT. Vanderbilt, Tom. 2011a. “Did the First Cities Grow from Marshes?� Science 331 (6014): 141. ———. 2011b. “Long Live the Industrial City.� Wilson Quarterly (Spring): n.p. West, Geoffrey. 2010. “A Physicist Solves the City.� New York Times, December 17. http://www.nytimes.com/2010/12/19/magazine/19Urban_West-t.html? pagewanted=all. World Bank. 2008. Climate Resistant Cities: A Primer on Reducing Vulnerabilities to Climate Change Impacts and Strengthening Disaster Risk Management. Washington, DC: World Bank, Global Facility for Disaster Reduction and Recovery (GFDRR), United Nations International Strategy for Disaster Reduction (ISDR), June. http://siteresources.worldbank.org/EASTASIAPAC IFICEXT/Resources/climatecities_fullreport.pdf. ———. 2011a. World Development Indicators, 2011. Washington, DC: World Bank. ———. 2011b. World Development Report: Conflict, Security, and Development. Washington, DC: World Bank. Yusuf, Shahid, and Kaoru Nabeshima. 2006. Postindustrial East Asian Cities: Innovation for Growth. Washington, DC: World Bank. ———. 2010. Two Dragon Heads: Contrasting Development Paths for Beijing and Shanghai. Washington, DC: World Bank. Yusuf, Shahid, Kaoru Nabeshima, and Shoichi Yamashita. 2008. Growing Industrial Clusters in Asia: Serendipity and Science. Washington, DC: World Bank. Yusuf, Shahid, and Weiping Wu. 1997. The Dynamics of Urban Growth in Three Chinese Cities. New York: Oxford University Press for the World Bank. Zhu, Yu, Xinhua Qi, Huaiyou Shao, and Kaijing He. 2009. “The Evolution of China’s in Situ Urbanization and Its Planning and Environmental Implications: Case Studies from Quanzhou Municipality.� In Urban Population-Environment Dynamics in the Developing World: Case Studies and Lessons Learned, ed. Alex de Sherbinin, Atiqur Rahman, Alisson Barbieri, Jean-Christophe Fotso, and Yu Zhu. Paris: Committee for International Cooperation in National Research in Demography (CICRED). http://www.ciesin.columbia.edu/repository/pern /papers/urban_pde_zhu_etal.pdf. CHAPTER 8 Elements for Future Success of Metropolitan Regions Size and agglomeration economies can influence urban fortunes through productivity, but there are too many examples of metropolitan regions that are punching far below their weight. There are megacities where the development of industry and tradable services is creeping along or in retreat, where growth is stagnating, and where the supply of housing and public services is struggling to keep up with the demand because the productive economic base and revenue effort are both weak. Cairo, Johannesburg, Karachi, Manila, and São Paulo belong to this category of cities that are deriving few advantages from size and suffer instead from the diseconomies of unbridled agglomeration. What differentiates these cities from metropolitan regions that are economically dynamic and reg- istering high growth rates? For low- and middle-income countries that are experiencing lagging urban development in the face of rising urbaniza- tion, the missing ingredient is exploding business activity represented by the entry and growth of firms producing tradables (whether manufac- tured products or services), creating good jobs,1 generating exports, and serving as a channel for new technologies absorbed from overseas and supplemented by own adaptation and innovation. Bangalore, Bangkok, and Shenzhen owe their dynamism to the continual value-adding and growth-enhancing churning of the business scene, with new—domestic 157 158 Geography of Growth and foreign—firms serving as a conveyor belt for investment and technol- ogy and competitive pressures sharpened by exposure to global markets continually weeding out the laggards. The entry of new firms and the growth of the most entrepreneurial firms are the lifeblood of the metropolitan region.2 The dynamic cities not only benefit from high rates of entry, but also, as in Beijing or in Dongguan, encourage the formation of clusters, which give rise to technological spill- overs, stimulate productivity, and create conditions conducive to the for- mation of new firms.3 Entry, cluster formation, and growth of the more productive firms can promote exports, which, in turn, further stimulate economic expansion.4 In fact, urban industrialization in the current con- text and for all but the largest countries is inseparable from participation in the international market.5 This broadens market opportunities for venturesome firms, which are a minority everywhere—but an important one—and spurs productivity growth. Firms with the greatest managerial, organizational, and technical capabilities grow, and in both East Asia and Latin America, participation in international value chains has provided firms with technology and growth ladders. The experience of Taiwan, China, in particular, highlights this process of urban industrialization through a proliferation of small and medium enterprises, their entry into trade, their proactive absorption of technology, and their emergence as globally competitive entities that drive the economies of cities in Taiwan, China, and the nation. Once urban development takes off, the large metropolitan region has several advantages. The medium-size peripheral cities are likely to grow quickly—a worldwide trend—to have a large youthful population that can provide entrepreneurial dividends, and to have lower-priced land to encourage new starts, especially in manufacturing. The core city, with a concentration of services and unskilled workers, offers a different range of opportunities, with many more niches for new start-ups and easier access to financing for existing firms or clusters of firms and for small and medium enterprises (SMEs).6 The core city is better supplied with busi- ness development services, which can be valuable for start-ups. The core city is also the focus of academic and cultural activities. Together, the con- centration of universities, research and consulting services, and recreational facilities provides the opportunities for knowledge workers with diverse skills to exchange and breed new ideas, some of which are enriched by being at the intersection of two or several disciplines. The metropolitan region that combines the advantages of medium-size and large cities has strong economic credentials, but its full developmental Elements for Future Success of Metropolitan Regions 159 potential is only realized when certain other criteria are met in whole or in part. These are as follows: (a) industrial composition and clustering, (b) connectedness, (c) compactness, (d) urban smarts, (e) governance, and (f) sustainability and resilience. These criteria or attributes were not uppermost in the minds of city planners, managers, and developers; when metropolitan cities were taking shape in the twentieth century, fuel was cheap, abundant land was avail- able for development, pollution and population pressures were less obtrusive, and sprawling low-rise cities appeared appropriate for the fore- seeable modes of economic activity and lifestyles. Few if any city author- ities seriously considered adopting a holistic approach, which seems warranted from the vantage point of current knowledge. But in the future, to succeed in attracting resources and talent and to maintain adequate growth rates, cities will need to monitor progress with reference to the above, moving farther along some axes than others, depending on circumstances, without neglecting any one. The Industrial Matrix It is appropriate to start with industrial composition because this is of immediate relevance for growth, employment, and exports, and the cur- rent mix foreshadows future options for a metropolis. The type and competitiveness of productive activities dominating the city’s economy are the principal indicators of growth prospects through sales in domestic and foreign markets. They indicate the gains to be derived from produc- tivity, from innovation, or from technological catch-up. Figure 8.1, on total factor productivity (TFP) by industrial sector in the United States between 1960 and 2007, demonstrates the edge that certain high-tech industries, but also services benefiting from information technology (IT), have over other activities. Figure 8.2 points to the research and develop- ment (R&D) intensity of key subsectors among 10 Organisation for Economic Co-operation and Development (OECD) countries, which is a reliable indicator of innovation and productivity growth. And the indus- trial composition points to employment elasticities and the types of skills likely to be in demand. When firms cluster in ways that promote spill- overs, the productivity bonus can be larger. The IT-enabled services sector in Bangalore7 and in Gurgaon, the sec- ond largest city in Haryana and located about 30 kilometers south of New Delhi, are clusters of proven competitiveness and export success, employ- ing highly skilled workers and diversifying into more complex services 160 Geography of Growth Figure 8.1 Industry Contributions to Productivity Growth in the United States, 1960–2007 0.80 0.70 average annual contribution (%) 0.60 0.50 0.40 0.30 0.20 0.10 0.00 –0.10 –0.20 1960–95 1995–2000 2000–07 Domar weight = ratio of industry output to aggregate value added non-IT industries IT-using industries IT-producing industries Source: Jorgenson, Ho, and Samuels 2010. offering larger rewards. This kind of industry, with good long-term poten- tial and significant local linkages, is an asset for the metropolis, not the least because it is an industry with low entry barriers, which encourages the proliferation of businesses in societies where demonstration effects can uncork pent-up entrepreneurial energies. Dongguan, one of the fastest-growing metro cities in China, is the center of manufacturing covering a spectrum ranging from textiles to electronics.8 These industries provide jobs for skilled and unskilled work- ers, and the diversity is fertile soil for new businesses. Manufacturing activities in Dongguan target foreign markets, and major multinationals such as Foxconn and Nike have located their main manufacturing assets in the city. This further enriches the industrial ecology of the city because large factories owned by multinational corporations (MNCs) exploit scale economies and buy inputs from or subcontract with thousands of specialized suppliers.9 The MNCs nourish the ecosystem with capital and production technologies and boost the development of local research and testing facilities.10 No less important from the productivity angle are the Elements for Future Success of Metropolitan Regions 161 Figure 8.2 R&D Intensity, by Industry Average across 10 OECD Countries (average across 10 countries in %) Services sector Paper, paper products, and printing Metal products Food, beverages, and tobacco Textiles, apparel, and leather Petroleum refineries and products Nonmetallic mineral products Wood products, furniture, other manufacuring, nec Iron and steel Nonferrous metals Shipbuilding and repairing Rubber and plastic products Chemicals excluding drugs Other transport equipment Nonelectrical machinery Motor vehicles Aircraft Professional goods Drugs and medicines Office and computing machinery Subtotal electrical-electronic 0 5 10 15 20 25 30 R&D intensity Source: van Pottelsberghe 2008. Note: nec = not elsewhere classified. managerial, design, and marketing techniques and the habit of many- faceted, incremental innovations that the MNCs introduce. That manu- facturing productivity is increasing by 10 percent or more in cities such as Dongguan testifies to the speed at which technologies are being dis- seminated, and this helps to absorb rising wages, while maintaining healthy profit margins.11 Bangkok is yet another example of a dynamic industrial metropolis. The core city is richly supplied with services, and around it has sprung a necklace of secondary cities crowded with manufacturing firms that rely on the providers of IT, finance, management, marketing, logistics, and human resource management services located in Bangkok.12 The metro- politan economies and the advantages accruing from the presence of the central government are such that efforts to disperse economic activities to other cities in Thailand have largely failed. Other cities, such as Cairo, 162 Geography of Growth Johannesburg, or Rio de Janeiro, with a modest suite of tradable activi- ties pay a price. Cairo’s manufacturing sector is smaller, mainly low-tech, and low also on the scale of competitiveness. Services cater mostly to domestic demand. This constrains productivity gains, technological change, diversification, and growth. Rio de Janeiro is in a similar pre- dicament, having deindustrialized and failed to substitute departing industries with tradable services other than those serving tourists.13 Rio de Janeiro, for all its natural beauty, is a city without the leading export and research-intensive sectors that can deliver high rates of growth and employment and lessen the city’s dependence on budgetary transfers from the center.14 In spite of its strong mining and engineering sectors, Johannesburg also has to cope with slow growth, largely because of the decline in mining and manufacturing activities,15 which tend to be skill intensive and offer few jobs for South Africa’s legions of unemployed, youthful, unskilled workers. Growth prospects of the Johannesburg-Gauteng region, while adequate for the near term, look increasingly dim over a longer horizon unless industrial trends are reversed. What we learn from Chinese and some Southeast Asian metropolitan centers is that, for low- and middle-income countries, a broad manufac- turing base—complemented, as in Bangkok, Shanghai, and Taipei, by the densification of services industries—promises growth and the scope for diversification. Analysis using the Hausmann-Rodrik-Hidalgo product space mapping technique indicates that production systems lying on the periphery of the product space without many links to other product categories, as in the case of Johannesburg and Rio de Janeiro, face diffi- culty in acquiring the richly networked core activities that contribute to a deepening of industrial capabilities with better longer-term growth prospects.16 Figures 8.3 and 8.4 present the sectors with the highest rates of pro- ductivity increase in the United States over a 47-year period. These industries also rank high in the trade statistics, and the manufacturing industries on the list are among the most research intensive, underscoring the likely longer-term growth potential. The list of activities is illustrative: these are some of the subsectors that performed well in the recent past and boosted city economies. However, productive highfliers can be stingy sources of jobs. Whether it is finance and insurance or electronics and biotechnology, leading innovative activi- ties are more productive and more intensive in their use of capital and skills. If recent trends persist, and there is reason to think that they will Elements for Future Success of Metropolitan Regions 163 Figure 8.3 Contributions to Productivity Growth in the United States, by Industry, 1960–2007 0.6 0.5 annual Domar-weighted 0.4 contribution (%) 0.3 0.2 0.1 0 1960–2007 1960–95 1995–2000 2000–07 –0.1 Information and data-processing services Computer systems design and related services Computer and peripheral equipment manufacturing Communications equipment manufacturing Semiconductor and other electronic components Software publishing Source: Jorgenson, Ho, and Samuels, 2010. for at least a decade, metropolitan centers with a large share of the trad- ables in the most productive categories could grow faster than other centers, but if they wish to produce an abundance of jobs, they will need to nurture many nontradables, mainly services. Connectivity A highly connected metropolitan region enhances productivity and maximizes the benefits from increased trade and capital flows, the circu- lation of talented people, and the collaborative efforts of researchers in different countries. Connectedness has several facets, but the two that deserve the most attention are the quality of the information and com- munication technology (ICT) and transport infrastructure and the links they help to create. A wealth of research has pieced together evidence mainly from devel- oped countries showing that the cross-sectoral absorption of ICT in myriad activities has raised productivity and induced innovation. Erik Brynjolfsson believes that IT is changing the innovation process itself. He claims, “IT is setting off a revolution on four dimensions simultaneously: measurement, experimentation, sharing, and replication that reinforce 164 Geography of Growth Figure 8.4 Industry Contributions to Productivity in the United States, 1960–2007 Wholesale trade Retail trade Semiconductor and other electronic component Computer and peripheral equipment manufacturing Securities commodity contracts and invesments Farms Broadcasting and telecommunications Software publishing Textile mills and textile product mills Air transportation Rail transportation Motor vehicles bodies and trailers and parts Truck transportation Federal general government Machinery Miscellaneous manufacturing Other electronic products Miscellaneous professional scientific and technical services Fabricated metal products Other transportation and support activities Plastics and rubber products Accommodation Information and data processing services Food and beverage and tobacco products Warehousing and storage Communications equipment manufacturing Other transportation equipment Petroleum and coal products Furniture and related products Social assistance Water transportation Waste management and rmediation services Food services and drinking places Pipeline transportation Chemical products Electrical equipment appliances and components Mining except oil and gas Apparel and leather and allied products Nonmetalic mineral products Paper products Performing arts and spectator sports museums and related Administrative and support services Printing and related support activities Motion picture and sound recording industries Amusements, gambling, and recreation industries Household Wood products Federal government enterprises Support activities for mining Computer systems design and related services Transit and ground passenger transportation Forestry, fishing, and related activities Educational services Primary metals Management of companies and enterprises S&l government enterprises Funds, trusts, and other financial vehicles Insurance carriers and related activities Other services except government S&l general government Newspaper; periodical; book publishers Legal services Utilities Ambulatory health care services Rental and services and lessors of intangible Hospitals, nursing, and residential care facilities Oil and gas extraction Real estate Federal reserve banks, credit intermediation, and related Construction –0.10 –0.05 0 0.05 0.10 0.15 0.20 contribution to productivity (%) Source: Jorgenson, Ho, and Samuels, 2010. and magnify each other� and permit the rapid scaling up of innovations (quoted in Hopkins 2010, 52). The United States has been the leader in this regard, although European countries have also benefited, and some developing countries are catching up. Elements for Future Success of Metropolitan Regions 165 The centrality of capital for growth in the world as a whole since 1980 has been highlighted by Jorgenson and Vu Khuong (2009), who show that capital was the source of 54 percent of growth in 1989–95 and that its share was still as high as 41 percent during 2000–06, exceeding the contribution of other factors. The compelling development since the mid- 1990s is the increasing importance of total factor productivity, which accounted for 36 percent of growth in 2000–06, compared with less than a fifth in 1989–95. If this trend were to persist—and Jorgenson, Ho, and Samuels (2010) acknowledge that productivity will be vital for maintain- ing U.S. living standards over the longer term—then TFP could draw abreast or even pull ahead of capital as a driver of growth. However, as Jorgenson and his coauthors observe, “With only replication [of estab- lished technologies] and without innovation, output will increase in proportion to capital and labor inputs. By contrast, the successful intro- duction of new products and new or altered processes, organization struc- tures, and business models generates growth of output that exceeds the growth of capital and labor inputs. This results in growth of multifactor productivity or output per unit of input� (Jorgenson, Ho, and Samuels 2010, 13–14). Innovation, or the successful exploitation of new ideas and technology, is becoming the cornerstone of growth, and ICT is the leading driver of such growth.17 As Jorgenson, Ho, and Samuels (2009) show,18 the contribution to growth of IT-producing and IT-using industries in the United States has increased steadily, and IT capital now makes a substan- tial contribution to gains in productivity. Additional evidence on the salience of ICT is provided by Brynjolfsson and Saunders (2010) and the research conducted by the Information Technology and Innovation Foundation on urban activities deriving productivity gains from IT.19 The point to be noted is that the use of ICT for industrial, commercial, or social purposes is to a great extent an urban phenomena and, because frequent exchanges via electronic media also increase face-to-face encounters (Leamer and Storper 2001), a metro region well furnished with ICT infrastructure and recreational amenities is the ideal setting for the circulation of information, the testing of ideas, and the fruiting of innovation. Seoul is a classic example of a city with state-of-the-art ICT infrastruc- ture providing locals with unparalleled access to the Internet and the latest advances in mobile telecommunications. Seoul’s edge over most other cities derives from the government’s ambitious plans, launched in 1995, to wire the nation in enlightened anticipation of a tectonic shift in communications and the use of media20 and subsequent initiatives to 166 Geography of Growth develop IT-based activities. Such activities include the Digital Media City to support the growth of the digital content industry, a major source of high-value-adding jobs in the metro area. Productivity gains aside, the large strides made in weaving ICT into the fabric of urban life in the Republic of Korea have spurred innovation, as evidenced by increasing patent output and, more important, the rise in international collaboration between Korean and foreign researchers. Domestic connectivity strengthened urban civil society and energized social and intellectual activities. International connectivity is tightening the links that Korea needs to sustain its competitiveness. Singapore is another city that has leveraged ICT to maximize the gains from globalization and made its business environment the envy of other countries in the region and beyond. Singapore is a leader in technologies to expedite the operations of its busy container port and its world-class airport.21 It also uses electronic pricing to smooth traffic flows and to minimize congestion. Singapore’s e-government platform is the bench- mark for other cities, and the government is continuously searching for ways to prune transaction costs further. Through these investments in ICT as well as others in education and health care, Singapore has strengthened connectivity and raised total factor productivity. Other cities taking note of the benefits accruing to Seoul and Singapore have begun investing in infrastructure and training, but they frequently neglect to adopt a com- prehensive approach, which is the key to mutually reinforcing gains from several interlocking activities. A major metropolis seeking greater connectivity must also look to its airport and port facilities (if it is a coastal city). An urban economy reliant on trade—and the foremost metropolitan regions depend on trade to boost domestic sources of demand by a few percentage points—must enlarge and grease the channels through which trade flows.22 The eco- nomic significance of ports has long been recognized. A busy port has a large footprint, employing tens of thousands and consuming a wide assortment of services produced locally.23 The contribution of a major international airport equals and may exceed that of a port. By value, close to one-third of global trade is now shipped by air.24 This includes high- value electronic products and pharmaceuticals, cut flowers, meat, and other farm products requiring a cold chain—and the percentages are ris- ing as the cost of air transport declines in relative terms with the intro- duction of larger fuel-efficient aircraft. In addition, airports serve as the gateways for the export of tourism and business travel services, which cities such as Bangkok, Cairo, Cape Town, and Rio de Janeiro depend on Elements for Future Success of Metropolitan Regions 167 for the large slice of their earnings from trade. As air transport has increased its share of trade, major airports with space around them are becoming the foci of industrial and services clusters. A classic example is Dulles airport serving the Washington, DC, area, which is the axis of IT, telecommunications, and defense industry clusters and the driver of growth for the metropolitan region.25 Other cities are also discovering that airports can stimulate clustered industrial activities through connec- tivity and induced employment. Songdo City,26 which is sprouting IT activities adjacent to Seoul-Incheon airport, is one example; Bangkok’s new Suvarnabhumi airport is another. Both cities see these airports as hubs for new activities with a high trade component. The Smarter Metropolis The globally connected metropolis, which is a “smart city� like Seoul or Singapore or San Francisco or San Jose, is doubly advantaged because it has the capabilities to exploit the opportunities arising from globaliza- tion. There is no precise definition of the smart city. Being smart is associ- ated with several attributes, including a large percentage of the population with a college degree, state-of-the-art ICT infrastructure,27 and the early adoption of environmentally friendly and green technologies. However, for our purposes, urban “smarts� or intelligence derives from a concentra- tion of skills and the quality of governance. In other words, being smart has to do with the brainpower a city can marshal to manage and acceler- ate its development with the help of innovation at many different levels. Alongside depth and quality of human capital, these cities require insti- tutional mechanisms and basic research for generating ideas and ways of debating, testing, and perfecting these ideas. The smart city can achieve rapid and sustainable growth of industry by bringing together and fully mobilizing four forms of intelligence: the human intelligence inherent in local knowledge networks enriched by the inmigration of people with diverse talents;28 the collective intelligence of institutions that support innovation through a variety of channels and serve to urbanize technologies, shaping them to suit the environment and making them easily available to users; the production intelligence of its industrial base; and the collective intelligence that can be derived from the effective use of digital networks and online services—a kind of invol- untary crowd sourcing that contributes to problem solving and a progres- sive upgrading of the urban environment (Komninos 2008). Cities positioning themselves to become innovative hotspots are open to ideas 168 Geography of Growth and thrive on the heterogeneity of knowledge workers drawn from all over the country and the world. Moreover, such cities are closely inte- grated with other global centers of research and technology develop- ment—they are a part of the global innovation system—and their teaching and research institutions must compete with the best for talent and vali- dation of their own ideas. Last, but not least, because smart cities are at the leading edge of the knowledge economy, their design, physical assets, attributes, and governance need to reflect their edge over others. Industrial cities can become innovative cities, and a strong manufacturing base can be an asset, as it is for Munich, Seattle, Seoul, Stuttgart, Tokyo, and Toulouse. But industry is not a necessary condition: Cambridge (United Kingdom), Helsinki, Kyoto, and San Francisco are not industrial cities; they are innovative cities that have acquired significant high-tech or IT production capabilities. As long as a city is part of a metro region or adja- cent to one, size can be a secondary consideration and overridden by the advantages of livability. Medium-size industrial cities, by exploiting local- ization economies, can promote the formation of vibrant industrial clus- ters. And because they tend to be less congested, medium-size cities can appeal to younger age groups, who are concerned about cost of living and environmental quality,29 as well as to members of the creative class, who place a high premium on quality of life (see tables 8A.1–8A.5 at the end of this chapter, which rank cities with respect to quality of life and creativ- ity and highlight the lead enjoyed by medium-size cities). Of course, only a subset of mid-size cities are potential winners, but those that exploit their location and strategically develop the assets that contribute to long- term prosperity can equal or exceed the innovation and productivity advantages of the most dynamic large cities.30 A city with an abundance of skills is better positioned to maintain industrial competitiveness, to move up the value chain by assimilating technologies and reinforcing catch-up with innovations, and to diversify into more profitable activities as existing ones enter the stage in their life cycle when commoditization lowers entry barriers, pares profit margins, and triggers migration to lower-cost locations. Glaeser (2011) singles out Boston as a skilled city that has flourished because its world-class univer- sities and urban ambience have made it “sticky� for talented people (on stickiness, see Markusen 1996). The wide base of skills has nurtured entrepreneurs and led to the proliferation of firms offering jobs for skilled workers; with the universities generating so many ideas, Boston has recov- ered from downturns and bouts of deindustrialization by pursuing new technological opportunities using its unique labor pool and financing Elements for Future Success of Metropolitan Regions 169 these with the help of highly experienced, locally based venture capital- ists. Boston is not alone. Other cities such as Bangalore, Beijing, Singapore, and Taipei are adopting similar models of development to good effect. The leading smart cities have not only deep pools of skills, but also high-quality skills. Growth regressions are uncovering a robust relation- ship between the quality of schooling, as captured by test scores of middle school students, and the increase in gross domestic product (GDP) (Hanushek 2010; Hanushek and Woessman 2010). These have been capped by related findings highlighting the significance of the numbers of students in the upper tail of the distribution of test scores (see Pritchett and Viarengo 2010). A country—or city—with many students with sci- ence and math scores in the highest percentiles has the strongest growth prospects. Singapore, which is top-ranked by test scores, also has impres- sive competitiveness and innovation capacity rankings. It has successfully diversified and sustained an average growth rate of 5 percent since 1995. Shanghai, which topped the 2009 Programme for International Student Assessment results, is en route to becoming a smart metropolis the equal of Seoul and Tokyo. Shanghai is a magnet for talent from throughout China, and this inflow augments its own base of high-quality skills. As traditional light manufacturing industries transfer to cities in Shanghai’s hinterland or to the interior, new and more skill-intensive activities are enabling Shanghai to expand in fresh directions appropriate for a city with per capita GDP that is five times the average for China. Mexico City and São Paulo trail Shanghai’s performance, and their prospects are less bright because they have not set their sights on becoming smart cities with human capabilities as the prime source of growth. Governing the Metropolitan Center A metropolis will struggle to accumulate and retain talent and create new business lines if urban planning, management, and financing do not pro- vide the necessary preconditions for development. That is, smart urban governance complements other forms of urban intelligence. Suffice to say that the selection and empowerment of city managers are one of the req- uisites. Smart cities plan ahead, establish realistic monitorable targets, and place a premium on rapid and efficient implementation of policies.31 Cities such as Seoul, Singapore, and Tokyo draw their governance capa- bilities from the quality of a well-paid municipal workforce and an insti- tutional infrastructure that evolves with changing developmental imperatives and is quick to incorporate IT as well as other technologies to 170 Geography of Growth improve services delivery. The enduring characteristic of smart cities is the awareness of competition and the commitment to incremental progress through benchmarking and learning from other cities. Smart cities such as Singapore are not caught unawares by the hollowing out of traditional industries; instead, they seek to anticipate and avert or neutralize trends that can lead to the entrenching of slums and environmental decay— physical as well as social. Cape Town, Karachi, and Rio de Janeiro have sacrificed many of the advantages that could be derived from a concentra- tion of skills because the environment in both cities is rendered perilous by widespread unemployment, serious security concerns, and the obtru- siveness of slums, whether in the core city areas or on the outskirts. Being smart is all about defining ambitious but achievable develop- ment objectives, mobilizing resources using a frequently sharpened set of incentives to deliver results, thinking ahead so as to minimize the risk of being caught napping, and solving problems expeditiously. Smart cities can raise their game by making full use of technological opportunities as they arise and by inculcating a culture of innovation. However, high-tech and IT intensity is not the answer for most cities, or at best it is a partial answer. Smart urban development in Cairo and Karachi would be low- tech yet innovative at the outset, while aiming for longer-term growth based on skills and technological capabilities that would narrow the vast gaps in productivity between these cities and some of their competitors in East Asia. The Resilience Imperative A metropolis that is deemed smart and successful must also meet the test of sustainability and resilience. Metropolitan economies in low- and middle-income countries must strive after decades of growth in the 5 to 8 percent range to generate enough employment, raise living standards of the vast majority to socially acceptable levels, and find the resources to address legacy problems and upcoming challenges, not to mention envi- ronmental and economic shocks. Both governance and skills directly impinge on sustainability. Governance affects development because it is a determinant of the urban business climate and the level of business activity. Skills likewise have a powerful bearing on development, as discussed above. The point to be noted is that the sustainability of a metropolitan economy is inseparable from growth. If growth stalls or goes into reverse, as happens when a key Elements for Future Success of Metropolitan Regions 171 industrial or mining activity implodes—as has happened in Detroit, Pittsburgh, and some cities in Eastern Europe—sustainability is imperiled because industrial decline is followed by rising poverty and social unrest and by an exodus of capital and skills. Avoiding such a contingency is central to the notion of sustainability. A diversified metropolitan region is at lower risk than smaller specialized urban centers, but, as even New York discovered in the mid-1970s, a narrowing of the industrial base and excessive dependence on a few services subject to crisis-induced swings can become problematic. Both Seoul and Shanghai have industrial breadth, as do Karachi, Lagos, and Mumbai, but the plight of the latter three cities draws attention to two other facets of sustainability: urban finance and urban design. Financing Urbanization Urban development assumes the provision of an array of services for businesses and households. If these dip below minimum standards of adequacy, development is impeded and the urban economy begins to stall and unravel, as happens in conflict and immediate postconflict situ- ations. Infrastructure services, public health services, education services, and policies are among the basics. Scarcity of water, for example, can seriously constrain urban development, and poor sewage, waste disposal, and sanitation severely compromise the health and living conditions of the majority. Whether a metropolitan region can build and maintain physical infra- structure, finance basic services, supply affordable housing, and offer recreational amenities is ultimately a function of finances. Transfers from central and provincial governments (both general and specific) are the source of revenues, but sustainability requires that these constitute a relatively minor source of income and that the local tax base is the pri- mary source of revenues. At least five criteria must be met for a city to be broadly self-sufficient with regard to revenue. First, revenue generation is a function of the scale of economic activity and how this translates into earnings of residents, the distribution of incomes, and the value of taxable assets. Second, the revenue actually raised depends on the degree of local tax autonomy and taxes assigned to local authorities. Other fees collected by municipalities supplement taxes, but income and real property taxes generally constitute the bulk of local revenues. To meet expenditure assignments, subnational governments often look to central governments 172 Geography of Growth to bridge any gaps, but a sustainable metropolis should, in principle, be self-sufficient. Third, the selection and use of tax instruments need to be efficient and to derive the maximum advantage by maintaining incentives for busi- nesses and households to remain in the jurisdiction (see Inman 2007). Moreover, local authorities need to be able to enforce and collect taxes, especially property and real estate taxes, and to assess properties and adjust rates regularly. Fourth, a metropolis spanning multiple jurisdictions must be able to coordinate regional development so as to optimize the provision of infra- structure and internalize scale economies where these exist. Equally important is the coordination of tax instruments and rates so as to avoid distorting incentives and inducing tax arbitrage and Tiebout shopping.32 Fifth, fiscal responsibility laws can serve to underscore local respon- sibilities, minimize moral hazard, and induce fiscally prudent behav- ior.33 Furthermore, the fiscal performance and service delivery of local governments can be bolstered by procedures for evaluating perfor- mance. Bangkok, much like other metropolitan centers in developing countries, relies on a mix of transfers and locally sourced revenues, but efficiency is compromised by the large number of local government organizations and an inability to analyze the data collected to improve monitoring and performance. Tax revenues can partially finance infrastructure, but most long-lived capital-intensive facilities call for additional financing, which can come from development grants provided by the center or raised by issuing bonds that are guaranteed by the center or provincial governments. Whether it is tax revenues or financing through public-private part- nerships or the financial market, sustainability first and foremost assumes that industrial development is on track and that trends are pointing in the right direction. Where the development impetus is weak or failing, finan- cial sustainability can prove elusive. Governance mechanisms—central and local—are an equally important determinant of sustainability, affect- ing not just corruption and malfeasance but also legislative log rolling, a common problem in U.S. cities, which is when legislators avoid the risk of policy gridlock by indiscriminately voting for all new initiatives.34 Designing for Sustainability Today’s metropolitan regions emerged in most instances with the mini- mum of planning and attention being given to resource constraints or long-term environmental considerations. Low energy prices, transport Elements for Future Success of Metropolitan Regions 173 subsidies, cheap land, low property taxes, the lure of automotive mobility, and the emergence of powerful lobbies composed of real estate develop- ers and auto manufacturers together led to horizontal, sprawling urban development. This process is continuing in industrializing economies such as China, Indonesia, Malaysia, Nigeria, and South Africa and also in North America, which provided the model of the sprawling metropolitan region.35 This form of development, while it surely provides city-dwellers with more living space, requires costly investment in transport, water, sewage, and energy infrastructure and greatly increases dependence on private automobiles.36 Sprawl also goes hand-in-hand with eating and exercise habits that are injurious to health (Frumkin, Frank, and Jackson 2004). The sprawling metropolis, with its low densities and emptiness,37 poses a huge challenge for sustainable development. Sustainability is predicated on energy and resource conservation and on the building of robust and resilient infrastructures. The model of a resource-frugal city is compact and vertical, with high population densities that permit the efficient use of public transport.38 This model, attractive to efficiency- and resource-conscious planners, may be coming into vogue, but it should not take the form of the “tower in the park� model so popular in China, which is much more energy intensive and isolating than the mixed-use neighborhoods it is displacing. A doubling of urban populations demands a rethinking of how people can be accommodated, especially if there is a growing need to conserve both energy and fertile farmland adjacent to cities. The need to invest in facilities to protect the more vulnerable cities from the consequences of climate change is another factor that will be harder to realize given the declining trend in global savings linked to aging populations in the devel- oped world as well as in some of the industrializing countries. The immi- nence and seriousness of each of these trends can be debated. Legacy housing, transport, and public utility infrastructure and the force of iner- tia are huge obstacles to changing the pattern of urban development, but they cannot be ignored, and retrofitting them is unavoidable. Resistance to an increase in energy and water prices, or the price of externalities aris- ing from unchecked private automobile use,39 reluctance to raise and collect real property taxes, and reluctance to modify zoning and floor area regulations affecting land use (Mumbai is a frequently cited example) are fierce in all countries.40 The political economy of urban development in virtually all countries favors endless delay. This is because politicians have short time horizons and few incentives to champion radical policies, interest groups with a stake in the status quo forcefully oppose actions 174 Geography of Growth that would jeopardize the rents they gain from existing arrangements, and households reflexively oppose higher taxes and prices. Even severe fiscal crises, the threat of spiraling energy prices, and the increasing frequency of severe weather events seem unable to persuade metropolitan residents in advanced and developing countries that delay is fast becoming an unaf- fordable luxury. The issue of urban sustainability is here to stay, and with each passing year it will only become more pressing. In different ways—sometimes obliquely, sometimes directly—it is being debated in crisis-ridden advanced countries in a state of political paralysis such as the United States, in industrializing countries currently with deep pockets where urbanization is approaching a midpoint, such as China, and in low- income countries in the crosshairs of climate change, such as Pakistan, that are struggling with acute resource scarcities, limited organizational capabilities, and dysfunctional governance. Reluctantly and later rather than sooner, the great metropolitan centers throughout the developing world will translate the concept of sustainable urbanization into practice through a physical redesign of cities and the widespread incorporation of green technologies and resource-frugal ways of living. Legacy infrastruc- ture cannot be wished away overnight; however, through a process of deconstruction, retrofitting, adaptation, and new construction based on green templates, cities will have to be transformed if they are to remain livable and economically dynamic. It may be too late to contain carbon dioxide concentration to the desired 450 parts per million, but mankind will need to adapt to the 550 parts per million atmosphere toward which we are heading. Metropolitan Futures The terms smart and IT-enabled, compact, vertical, mixed use, green, and livable define the vision of the future for some, but no one knows quite how these terms can define a coherent and holistic long-term development plan for a Beijing, a Karachi, or a São Paulo, what kind of organizations could manage urbanization across several dimensions, what it would cost to implement, the amount of dislocation it would entail, and the viability of the eventual outcome in the world that future generations will inherit. The advantages—and also the drawbacks—of the compact city have been aired for many years. The technologies— hard and soft—that can make a city “greener� have been taking shape and Elements for Future Success of Metropolitan Regions 175 are being tested piecemeal. No one of the tiny experimental green cities currently under construction has been put to the test and its carbon neutrality convincingly established.41 The livability of compact and green cities and how they would accommodate diverse industrial activi- ties are also not known. The technologies coming off the drawing boards and some being commercialized are perhaps decades away from wide- spread application once they have been debugged and made more affordable. However, building sustainability cannot wait. Cairo, Dhaka, Karachi, São Paulo, and Shenzhen are daily pouring more concrete into the ground, accommodating more people, and building more roads. Instead of becoming more dense, urban areas are becoming less dense. Bangkok’s urbanized area grew sixteenfold between 1944 and 2002, and that of Accra grew 153 percent between 1985 and 2000. In spite of recurrent fiscal debacles, local politicians and city managers are unable to learn enduring lessons, and acres of literature on urban fiscal policies have failed to improve urban tax systems substantially worldwide. These are frightening trends and missed opportunities. Left unchecked they will make a rationalization of urban development far harder. Some economists are of the view that price adjustments reflecting energy and water scarcities, increased vulnerability of cities near rivers to flooding and coastal locations to rising sea levels,42 and inland areas to droughts and firestorms will bring about the redistribution of the population, force a refashioning of the urban landscape, and demand the building of passive and active coastal defenses, as in the Netherlands. Economists rightly underscore the strength of the market mechanism but are apt to minimize its failings, as evidenced by the devastating financial crisis of 2008 and 2009 and the many real estate bubbles. From the perspective of urban sustainability and green development, market-induced changes might be too slow, too myopic, and too piece- meal, and the market might not promote the kind of fast-paced innova- tion that is urgently needed or provide the insurance required by the inhabitants of vulnerable cities in developing countries. On the current trajectories, Karachi and Lagos could become the world’s two largest cities by mid-century, assuming that the availability of water permits such growth. A doubling of populations with no change in layout will lead to metropolitan regions that could be unsustainable and ungovernable over the longer term and forced to confront painful crises. Advanced countries may have the resources to indulge in wasteful sprawling urban regions, and they may even endure deindustrialization 176 Geography of Growth for several decades by living off their accumulated fat. But industrializing countries need to learn quickly and avoid the costly decisions made when energy, land, and water were relatively cheap, green technologies were unknown, and global warming was a scientific curiosity. Low-income countries have even less room for maneuver because they are lacking the growth momentum of the leading middle-income nations, the techno- logical capabilities, and the resources; in addition, they must cope with rapidly expanding populations. With so much urbanization still lying ahead and the stakes rising, the design and implementation of forward-looking urban development strat- egies are taking on added importance. Whether countries make rapid strides on the economic front will depend on one or a small handful of metropolitan centers. And whether these are smart, sustainable, eco- nomically dynamic, and livable will also depend on how cities develop organizational and technical skills, assure revenue autonomy, create agile (soft and hard) infrastructure, and make the best use of evolving practical ideas and technologies to take existing and budding metropolitan regions boldly into an uncertain future. Annex: City Rankings Table 8A.1 Mercer Quality of Living Ranking of Cities Worldwide, 2010 City and country Rank Index Vienna, Austria 1 108.6 Zurich, Switzerland 2 108.0 Geneva, Switzerland 3 107.9 Vancouver, Canada 4 107.4 Auckland, New Zealand 5 107.4 Dusseldorf, Germany 6 107.2 Frankfurt, Germany 7 107.0 Munich, Germany 7 107.0 Bern, Switzerland 9 106.5 Sydney, Australia 10 106.3 Copenhagen, Denmark 11 106.2 Amsterdam, Netherlands 12 105.7 Ottawa, Canada 13 105.5 Brussels, Belgium 14 105.4 Source: Mercer LLC, http://www.mercer.com/qualityof living. Note: Index base city: New York, United States = 100. Elements for Future Success of Metropolitan Regions 177 Table 8A.2 Ranking of Creative Cities in the United States, by Arts Employees per Capita, 2008 Arts employees City,State Rank per 1,000 residents Population Atlanta, GA 1 47.7 537,958 San Francisco, CA 2 39.7 808,976 Seattle, WA 3 36.1 598,541 Washington, DC 4 34.4 591,833 Minneapolis, MN 5 33.5 382,605 Boston, MA 6 32.7 609.023 Los Angeles, CA 7 31.4 3,833,995 New York, NY 8 28.0 8,363,710 Portland, OR 9 27.5 557,706 Philadelphia, PA 10 27.4 1,447,395 Source: Americans for the Arts, http://www.artsusa.org; U.S. Census Bureau, http://www.census.gov. Table 8A.3 Ranking of Innovative Cities in the United States, 2008 City and state Rank Population Portland, OR 1 557,706 Chicago, IL 1 2,853,114 Seattle, WA 1 598,541 New York, NY 1 8,363,710 San Francisco, CA 1 808,976 Minneapolis, MN 6 382,605 Boston, MA 6 609,023 Los Angeles, CA 6 3,833,995 Baltimore, MD 9 636,919 Sacramento, CA 9 466,488 San Diego, CA 9 1,279,329 Dallas, TX 12 1,279,910 Source: SustainLane, http://www.sustainlane.com/us-city-rankings/categories/innovation; population statistics from the U.S. Census Bureau, http://www.census.gov. Table 8A.4 Top 10 Innovation Cities in the World, 2010 City and country Global rank Boston, United States 1 Paris, France 2 Amsterdam, Netherlands 3 Vienna, Austria 4 New York, United States 5 Frankfurt, Germany 6 San Francisco, United States 7 Copenhagen, Denmark 8 Lyon, France 9 Hamburg, Germany 10 Source: 2thinknow Innovation TM Cities Program, www.innovation-cities.com. 178 Geography of Growth Table 8A.5 Ranking of Innovation Cities in the Americas, 2010 Rank in the City Americas Global rank Boston, MA 1 1 New York, NY 2 5 San Francisco, CA 3 7 Toronto, Canada 4 12 Washington, DC 5 23 Philadelphia, PA 6 30 Montreal, Canada 7 34 Seattle, WA 8 35 Austin, TX 9 44 Minneapolis–St. Paul, MN 10 45 Source: 2thinknow Innovation TM Cities Program, www.innovation-cities.com. Notes 1. All those pouring into cities are looking for “good jobs,� if not for themselves, then for their children. Banerjee and Duflo (2011, 228). 2. Firms test their competitiveness by selling in the domestic market, which is frequently sheltered by tariffs, transport costs, local regulations, cultural pre- dispositions of consumers, and complexities of marketing and logistics that foreign firms have difficulty mastering. Lenovo, the Chinese manufacturer of personal computers, and Haier, the producer of white goods, have established and maintained a lead in the domestic market by catering more effectively to local preferences and using domestic marketing channels effectively. 3. See McGee et al. (2007) on the globally oriented industrialization of Dongguan; see Yusuf, Nabeshima, and Yamashita (2008) on the international experience with clusters. 4. Larger, more capital-intensive, and more productive firms are more likely to venture into the export market. See Bernard et al. (2007); Iacovone and Javorcik (2010). On the relationship between trade and growth, see the sur- vey by Lopez (2005); Greenaway, Morgan, and Wright (1999). 5. Some evidence suggests that successful SMEs begin orienting toward global markets from the very outset. See the papers in Lloyd-Reason and Sear (2007). 6. Much depends on the availability of affordable accommodation for small firms and their employees. In cities such as London, New York, Paris, and the cities in Silicon Valley, such space is becoming hard to find, which is squeez- ing out the most dynamic elements of the urban economy. Elements for Future Success of Metropolitan Regions 179 7. See Heitzman (2004) on the development of Bangalore; see also http: //www.nytimes.com/2006/03/20/business/worldbusiness/20bangalore .html?ex=1300510800&en=993a 11e65908ab91&ei=5088. 8. With a population of almost 7 million in 2008, including nearly 5 million migrants, Dongguan is the fourth ranked Chinese city in terms of exports. 9. Now Chongqing is attempting to create a similar eco-system by inducing Foxconn and Hewlett Packard to establish production facilities in the city, with the promise that the city will work with them to attract suppliers to the inland metropolis. Together, the two companies will be investing US$3 billion. http://www.chinadaily.com.cn/business/2009-08/05/content_8528616.htm. 10. MNCs account for 87 percent of China’s exports of electronic devices and 88 percent of the exports of telecommunications equipment (Moran 2011). 11. In spite of rising wages, new entry and export growth continued in the Pearl River Delta during 2009–10. 12. Government investment in port and highway infrastructure and incentives for developers contributed to the growth of these cities and the transfer of some of the automobile, electronic, machinery, and other industries from the core city areas. See Yusuf and Nabeshima (2010). 13. A software industry serves the domestic market, but it lacks the large firms that account for the performance of Indian IT centers. Cape Town is in a similar predicament: the software-IT industry caters mostly to the domestic finance and insurance industry, which constrains its growth prospects. 14. The discovery of huge offshore pre-salt oil deposits will increase the revenues accruing to the state, depending, of course, on the terms negotiated with the center. 15. Engineering industries are transferring some of their operations to Australia. 16. See the discussion of the product space and core-periphery issues in Hidalgo et al. (2007). 17. The U.S. Department of Commerce estimates that technological innovation is responsible for as much as three-fourths of U.S. growth since World War II (Ezell and Atkinson 2010). This is not necessarily inconsistent with the find- ings of Jorgenson, Ho, and Samuels (2010) because the contribution of capital is heavily determined by embodied technological advances. 18. http://www.economics.harvard.edu/faculty/jorgenson/files/Houston_ productivity_DJA.pdf. 19. http://www.itif.org/. 20. See Farivar (2011); Lee (2005); http://www.itu.int/ITU-D/finance/work- cost-tariffs/events/tariff- seminars/kuala-lumpur-05/presentation-lee.PDF. 21. See http://www.portnet.com/WWWPublic/pdt_portnet.html on Singapore’s Portnet IT-based business-to-business system. 180 Geography of Growth 22. São Paulo’s Santos port has long been a bottleneck, even though the cost of its inefficiency and its roots are well known. See Doctor (2002). 23. Cities with major ports are coming to recognize the air and water pollution that is caused by shipping, but they have been slow to take remedial action, although some are preparing to offer power sources to run the systems of docked ships. 24. On the importance of air cargo services especially for high-value goods, see Leinbach and Bowen (2004). 25. This has given rise to Internet Alley in a four square mile area called Tyson’s Corner, a short drive from Dulles airport. See Ceruzzi (2008). 26. http://www.songdo.com/songdo-international-business-district/the-city /master-plan.aspx. 27. Cisco, IBM, and Siemens are among the companies working to create smart networked cities where computer monitoring and control of activities will increase the efficiency of everything from transport systems to energy and water use. Cisco’s Connected Urban Development approach and how it affects the workplace, transport, energy consumption, and businesses using IT are described by Villa and Mitchell (2009). 28. Many of these individuals are likely to be attracted by the presence of major universities. See Winters (2011). 29. Depending on the type of industry and environmental regulations, mid-size cities can be more or less polluted. 30. The relationship between size and innovation is analyzed by Carlino, Chatterjee, and Hunt (2007) and by Carlino and Hunt (2009). 31. The grave weaknesses of governments in industrializing countries are not so much in the making of policies as in their implementation. See Hallward- Driemeier, Khun-Jush, and Pritchett (2010). 32. Philadelphia has suffered from a lack of coordination on taxation, land use, and transport development among the 238 municipalities making up the greater metro area. See Pugh O’Mara (2002). 33. The bailouts of Rio de Janeiro and São Paulo highlight the problem of moral hazard. Discouraging cities from using long-term debt to finance current expenditures is a key objective. For a review of international experience with fiscal responsibility laws, see Liu and Webb (2011). 34. See Inman (2007), who cites a study showing that a doubling in the size of a city council results in a 20 percent increase in spending per city resident. 35. North America is the model of the sprawling industrial and science parks, which have also proliferated in developing countries (O’Mara 2007). 36. It also imposes a heavy burden on the poor living on the fringes of the city, who must engage in long and costly daily commutes, as, for instance, in Johannesburg and Rio de Janeiro. Elements for Future Success of Metropolitan Regions 181 37. The architect Rem Koolhaas remarks, “There are city centers around the world in which no one seems to be a full-time resident,� quoted in Heathcote (2010, 4). 38. This point is strongly championed by Ed Glaeser (2011). Although Manhattan is compact and densely populated, the New York metro area covers 3,000 square miles (Greater London is 600 square miles, Paris is 1,000 square miles), and it is significantly less dense than Los Angeles—the supposed epitome of a sprawling metropolis (7,738 residents per square mile compared with 5,728 per square mile for New York). But for all its density, Los Angeles is not a walkable city (Rybczynski 2011). Metropolitan São Paulo covers 8,000 square kilometers, while the Cape Town City region spans 100 kilome- ters (UN-HABITAT 2010). 39. The vision of “mobility on demand� (see http://cities.medi.mit.edu/) offered by the MIT Media Lab is alluring, and bit by bit, some elements of this are taking shape. Whether it or something like it is a part of the metropolitan future, not just in a few enlightened cities but worldwide, remains to be seen. 40. For example, a recent World Bank (2009) report notes that in China, the fragmentation of land on the fringes of cities is growing worse, land use is not being coordinated with the development of urban transport, and floor area ratios are increasing much too slowly. In fact, the gross floor area ratios are far lower in Chinese cities than in Seoul or Tokyo and much lower than in Manhattan. 41. Some incredible specimens of the green city are taking shape in Abu Dhabi (Masdar), Seoul-Incheon, Shanghai, and Tianjin, but their economic and social viability and carbon neutrality have yet to be put to the test. 42. See also Kahn (2010). See Jha et al. (2011) on both the magnitude of the problems and remedial measures. References Banerjee, Abhijit V., and Esther Duflo. 2011. Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. New York, NY: Public Affairs. Bernard, Andrew B., J. Bradford Jensen, Stephen J. Redding, and Peter K. Schott. 2007. “Firms in International Trade.� Journal of Economic Perspectives 21 (3): 105–30. Brynjolfsson, Erik, and Adam Saunders. 2010. Wired for Innovation: How Information Technology Is Reshaping the Economy. Cambridge, MA: MIT Press. Carlino, Gerald A., Satyajit Chatterjee, and Robert M. Hunt. 2007. “Urban Density and the Rate of Invention.� Journal of Urban Economics 61 (3): 389–419. Carlino, Gerald A., and Robert M. Hunt. 2009. “What Explains the Quantity and Quality of Local Inventive Activity?� Working Paper 09-12, Federal Reserve Bank of Philadelphia, Philadelphia, PA. 182 Geography of Growth Ceruzzi, Paul E. 2008. Internet Alley: High Technology in Tysons Corner, 1945–2005. Cambridge, MA: MIT Press. Doctor, Mahrukh. 2002. “Business and Delays in Port Reform in Brazil.� Brazilian Journal of Political Economy 22 (2): 79–101. http://www.rep.org.br/pdf /86-5.pdf. Ezell, Stephen J., and Robert D. Atkinson. 2010. The Good, The Bad, and the Ugly (and the Self-Destructive) of Innovation Policy: A Policymaker’s Guide to Crafting Effective Innovation Policy. Washington, DC: Information Technology and Innovation Foundation, October. http://www.itif.org/files/2010-good -bad-ugly.pdf. Farivar, Cyrus. 2011. The Internet of Elsewhere: The Emergent Effects of a Wired World. Piscataway, NJ: Rutgers University Press. Frumkin, Howard, Lawrence D. Frank, and Richard Jackson. 2004. Urban Sprawl and Public Health: Designing, Planning, and Building for Healthy Communities. Washington, DC: Island Press. Glaeser, Edward L. 2011. Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter, Greener, Healthier, and Happier. New York: Penguin Press. Greenaway, David, Wyn Morgan, and P. W. Wright. 1999. “Exports, Export Composition, and Growth.� Journal of International Trade and Economic Development 8 (1): 41–52. Hallward-Driemeier, Mary, Gita Khun-Jush, and Lant Pritchett. 2010. “Deals Versus Rules: Policy Implementation Uncertainty and Why Firms Hate It.� NBER Working Paper 16001, National Bureau of Economic Research, Cambridge, MA. Hanushek, Eric A. 2010. “The High Cost of Low Educational Performance: The Long-Run Impact of Improving PISA Outcomes.� Programme for International Student Assessment, Organisation for Economic Co-operation and Development, Paris. Hanushek, Eric A., and Ludger Woessmann. 2010. “Education and Economic Growth.� In Economics of Education, ed. Dominic J. Brewer and Patrick J. McEwan, 60–67. Amsterdam: Elsevier. Reprinted in International Encyclopedia of Education, ed. Eva Baker, Barry McGaw, and Penelope Peterson, 245–52. Amsterdam: Elsevier. Heathcote, Edwin. 2010. “Urban Evolution.� Financial Times, September 8, 2010. Heitzman, James. 2004. Network City: Planning the Information Society in Bangalore. New York: Oxford University Press. Hidalgo, Cesar A., Bailey Klinger, Albert-László Barabási, and Ricardo. Hausmann. 2007. “The Product Space Conditions the Development of Nations.� Science 317 (5837): 482–87. Elements for Future Success of Metropolitan Regions 183 Hopkins, Michael S. 2010. “The Four Ways IT Is Revolutionizing Innovation.� MIT Sloan Management Review, May 21. Iacovone, Leonardo, and Beata S. Javorcik. 2010. “Getting Ready: Preparation for Exporting.� University of Oxford. http://www.economics.ox.ac.uk/members /beata.javorcik/Tequila.pdf. Inman, Robert P. 2007. “Financing Cities.� In A Companion to Urban Economics, ed. R. J. Arnott and D. P. McMillen. Oxford: Blackwell. Jha, Abhas, Jessica Lamond, Robin Bloch, Namrata Bhattacharya, Ana Lopez, Nikolas Papachristodoulou, Alan Bird, David Proverbs, John Davies, and Robert Barker. 2011. “Five Feet High and Rising: Cities and Flooding in the 21st Century.� Policy Research Working Paper 5648, World Bank, Washington, DC. Jorgenson, Dale W., Mun S. Ho, and Jon D. Samuels. 2010. “Information Technology and the Productivity Expansion.� Paper presented at the 2010 Sixth Biennial North American Productivity Workshop, Rice University, Houston, TX, June 2–5. Jorgenson, Dale W., and M. Vu Khuong. 2009. “Information Technology and the World Economy.� http://www.frbsf.org/economics/conferences/0511/6 _ITAndWorldEconomy.pdf. Kahn, Matthew. 2010. Climatopolis: How Our Cities Will Thrive in the Hotter Future. New York: Basic Books. Komninos, Nicos. 2008. Smart Cities and Globalization of Innovation Networks. New York: Routledge. Leamer, Edward E., and Michael Storper. 2001. “The Economic Geography of the Internet Age.� NBER Working Paper 8450, National Bureau of Economic Research, Cambridge, MA. Lee, Byoung Nam. 2005. “Korean Government-Driven ICT Policy: IT 839 Strategy.� Paper presented at the Regional Seminar on Costs and Tariffs for the TAS Group Member Countries, Cyberjaya, Malaysia, May 31–June 3. Leinbach, Thomas R., and John T. Bowen Jr. 2004. “Air Cargo Services and the Electronics Industry in Southeast Asia.� Journal of Economic Geography 4 (3): 299–321. Liu, Lili, and Steven B. Webb. 2011. “Laws for Fiscal Responsibility for Subnational Discipline.� Policy Research Working Paper 5587, World Bank, Washington, DC. Lloyd-Reason, Lester, and Leigh Sear. 2007. Trading Places: SMEs in the Global Economy: A Critical Research Handbook. Cheltenham, U.K.: Edward Elgar. Lopez, Ricardo. 2005. “Trade and Growth: Reconciling the Macroeconomic and Microeconomic Evidence.� Journal of Economic Surveys 19 (4): 623–48. 184 Geography of Growth Markusen, Ann. 1996. “Sticky Placers in Slippery Space: A Typology of Industrial Districts.� Economic Geography 72 (3, July): 293–313. http://www.jstor.org /stable/144402. McGee, Terry, George C. S. Lin, Mark Wang, and Andrew Marton. 2007. China’s Urban Space: Development under Market Socialism. Routledge Studies on China in Transition. London: Routledge. Moran, Theodore. 2011. “Foreign Direct Investment and Development: Launching a Second Generation of Policy Research.� Peterson Institute, Washington, DC. O’Mara, Margaret. 2007. “Landscapes of Knowledge: History and the Evolving Geography of High Technology.� In special issue on “The Future Metropolitan Landscape,� ed. Peter Bosselman, Places 19 (1, Spring 2007). Pritchett, Lant, and M. Viarengo. 2010. “In Brief … Producing Superstars for the Economic World Cup.� CentrePiece Summer, London School of Economics, London. http://cep.lse.ac.uk/pubs/download/cp310.pdf. Pugh O’Mara, Margaret. 2002. “Learning from History: How State and Local Policy Choices Have Shaped Philadelphia’s Growth.� Greater Philadelphia Regional Review (March). Rybczynski, Witold. 2011. “Dense, Denser, Densest.� Wilson Quarterly (Spring): 46–50. UN-HABITAT. 2010. State of the World’s Cities 2010/2011: Cities for All; Bridging the Urban Divide. Nairobi: UN-HABITAT. van Pottelsberghe, Bruno. 2008. “Europe’s R&D: Missing the Wrong Targets?� Bruegel Policy Brief 2008/03, Bruegel, Brussels, February. Villa, Nicola, and Shane Mitchell. 2009. “Connecting Cities: Achieving Sustainability through Innovation.� Paper presented at the Fifth Urban Research Symposium 2009, Connected Urban Development, Cisco Systems, Internet Business Solutions Group. Winters, John V. 2011. “Why Are Smart Cities Growing? Who Moves and Who Stays.� Journal of Regional Science 51 (2): 253–70. World Bank. 2009. World Development Report 2009: Reshaping Economic Geography. Washington, DC: World Bank. Yusuf, Shahid, and Kaoru Nabeshima. 2010. Two Dragon Heads: Contrasting Development Paths for Beijing and Shanghai. Washington, DC: World Bank. Yusuf, Shahid, Kaoru Nabeshima, and Shoichi Yamashita. 2008. Growing Industrial Clusters in Asia: Serendipity and Science. Washington, DC: World Bank. What makes certain cities more competitive than others? Why do countries often find talent concentrated in a few regions rather than evenly spread across the country? What are the economic drivers that make cities more productive? Geography of Growth: Spatial Economics and Competitiveness answers these and many other questions. This book focuses on the forces that give rise to geographic concentrations of population and economic activity. It examines urban concentration at the developed and developing country levels. It also presents a typology of cities and evaluates how and why they work by examining innovative cities, smart cities, green cities, knowledge cities, creative cities, and global cities, as well as the growth and sustainability of metro regions. This book will be of interest to policy makers, researchers, and students of urbanization and population change. ISBN 978-0-8213-9486-1 SKU 19486