Infrastructure and Land Policies [1 ed.] 9781558442825, 9781558442511

Citation preview

Infrastructure and Land Policies

Land Policy Series Value Capture and Land Policies (2012) Climate Change and Land Policies (2011) Municipal Revenues and Land Policies (2010) Property Rights and Land Policies (2009) Fiscal Decentralization and Land Policies (2008) Land Policies and Their Outcomes (2007)

Infrastructure and Land Policies Edited by

Gregory K. Ingram and Karin L. Brandt

© 2013 by the Lincoln Institute of Land Policy All rights reserved. Library of Congress Cataloging-in-Publication Data Infrastructure and land policies / edited by Gregory K. Ingram and Karin L. Brandt. pages cm. Includes index. ISBN 978-1-55844-251-1 (alk. paper) 1. Infrastructure (Economics) 2. Land use. I. Ingram, Gregory K. II. Brandt, Karin L. HC79.C314796 2013 333.73—dc23 2013001318 Designed by Vern Associates Composed in Sabon by Achorn International in Bolton, Massachusetts. Printed and bound by Puritan Press Inc., in Hollis, New Hampshire. The paper is Rolland Enviro100, an acid-free, 100 percent PCW recycled sheet. manufactured in the united states of america

CONTENTS List of Illustrations Preface

Infrastructure, Land, and Development 1. Global Infrastructure: Ongoing Realities and Emerging Challenges Gregory K. Ingram and Karin L. Brandt 2. Infrastructure and Urban Development: Evidence from Chinese Cities Yan Song commentary David M. Levinson 3. Mobile Telephony and Socioeconomic Dynamics in Africa Mirjam de Bruijn commentary Anthony M. Townsend

Finance, Regulation, and Taxation 4. Economic Regulation of Utility Infrastructure Janice A. Beecher commentary Timothy J. Brennan 5. The Unit Approach to the Taxation of Railroad and Public Utility Property Gary C. Cornia, David J. Crapo, and Lawrence C. Walters commentary J. Fred Giertz

ix xiii

1

3

21

54

61

83

85 87

123

126

152

vi

Contents

6. The Location Effects of Alternative Road-Pricing Policies Alex Anas commentary Don Pickrell

The Challenges of Large Projects 7. Chicago and Its Skyway: Lessons from an Urban Megaproject Louise Nelson Dyble commentary Richard G. Little 8. Assessing the Infrastructure Impact of Mega-Events in Emerging Economies Victor A. Matheson commentary David E. Luberoff 9. Involuntary Resettlement in Infrastructure Projects: A Development Perspective Robert Picciotto commentary Dolores Koenig

155

182

187

189

212

215

233

236

263

Improving Sustainability and Efficiency

267

10. Sustainable Infrastructure for Urban Growth Katherine Sierra

269

11. Understanding Urban Infrastructure-Related Greenhouse Gas Emissions and Key Mitigation Strategies Anu Ramaswami commentary W. Ross Morrow

296

313

Contents

12. Strengthening Urban Industry: The Importance of Infrastructure and Location Nancey Green Leigh commentary Alain Bertaud 13. What Is the Value of Infrastructure Maintenance? A Survey Felix Rioja commentary Waheed Uddin 14. How and Why Does the Quality of Infrastructure Service Delivery Vary? George R. G. Clarke

vii

318

341

347

366

370

commentary Ahmed Abdel Aziz

407

Contributors

409

Index

411

About the Lincoln Institute of Land Policy

440

ILLUSTRATIONS Tables 1.1 A Majority of Studies Find That Infrastructure Increases Productivity 1.2 Stock Levels Are Highly Correlated, But Performance Varies Within Countries 2.1 Infrastructure Investment Share in Developed Countries, 1950–1990 2.2 Variables and Data Sources 2.3 Descriptive Statistics of Variables Used in the Land Price Function 2.4 Regressive Results in the Land Price Function 2.5 Independent Variables in the Conversion Model 2.6 Regression Results in the Land Conversion Function 4.1 Ownership Structure for Electricity and Water, 2007 4.2 Structural and Regulatory Status of the Public Utility Sectors 4.3 Allocation of Utility Revenue Requirements Under Economic Regulation 4.4 Select Alternatives to Ratebase/Rate-of-Return Regulation 5.1 Property Taxes Paid by Selected Electric Utility Operating Companies, 2011 5.2 State Assessed Taxable Value as a Percentage of Total Locally Taxable Value, 1992 5.3 Public Utility/Centrally Assessed Property as a Percentage of Total Taxable Value, 2011 6.1 Calibrated Elasticities in RELU-TRAN2 (Chicago MSA) 6.2 Percent Changes in Driving-Related Aggregates Under Road-Pricing Policies 6.3 Effects of the Pricing Policies on Jobs, Residences, and Undeveloped Land 6.4 Changes in Welfare Components, Wages, and Rents Under Pricing Policies 8.1 Hosts of the Summer and Winter Olympic Games and FIFA World Cup 8.2 Number of Bids for Summer and Winter Olympic Games 8.3 Examples of Mega-Event ex ante Economic Impact Studies 8.4 Examples of Mega-Event ex post Economic Impact Studies 8.5 Costs of Hosting Mega-Events 10.1 The Urban Transition by the Numbers 10.2 Greening the Local Fiscal Tool Kit 10.3 Using Public Finance to Leverage Private Capital for Sustainable Infrastructure 10.4 Mexico’s Urban Transportation Transformation Program Investment Plan

8 18 23 34 36 37 47 47 97 100 104 110 127 129 129 167 173 174 175 216 218 220 220 223 270 276 281 285 ix

x

Illustrations

11.1 Impact of Different Strategies on Near-Term Transport Sector GHG Mitigation Over Five Years 12.1 Loss of Industrial Land to Rezoning in Select U.S. Cities 12.2 Local Industrial Issues, Policies, Smart Growth Planning 12.3 Comparing the Costs Associated with Greenfield and Brownfield Development in Portland, OR 12.4 Contrasted (Opposed) Paradigms: Large Technical Networked System Versus Sustainable “Techno-ecocycle” 13.1 Additional Operating Costs Due to Using a Road in Bad Condition 13.2 The Condition of Infrastructure 13.3 Economic Rate of Return for Maintenance Projects 13.4 Capital and Operations and Maintenance Expenditures on Public Infrastructure in the United States, 1956–2004 14.1 Correlation of Access Indicators Before Controlling for Macroeconomic Regressors 14.2 Different Measures of Access by Income Level 14.3 Impact of Macroeconomic Variables on Availability of Infrastructure 14.4 Correlation of Access Indicators After Controlling for Macroeconomic Regressors 14.5 Correlation of Price Indicators Before Controlling for Macroeconomic Regressors 14.6 Impact of Macroeconomic Variables on Price of Infrastructure Services 14.7 Correlation of Price Indicators After Controlling for Macroeconomic Regressors 14.8 Correlation of Quality Indicators Before Controlling for Macroeconomic Regressors 14.9 Impact of Macroeconomic Variables on Quality of Infrastructure Services 14.10 Correlation of Quality Indicators After Controlling for Macroeconomic Regressors 14.11 Correlations with Electricity Obstacles 14.12 Correlations with Transportation Obstacles 14.13 Impact of Macroeconomic and Sector Variables on Perceptions About Infrastructure 14.14 Electricity Variables 14.15 Transportation Variables 14.16 Water and Telecommunications Variables 14.17 Macroeconomic Variables 14.18 Pairwise Correlations for Electricity Variables

308 324 325 333 336 348 349 355 357 375 376 377 378 379 380 382 383 384 385 389 391 393 401 402 403 403 404

Illustrations

xi

Figures 1.1 Growth of U.S. Infrastructure as a Percentage of Maximum Network Size, 1850–2005 1.2 Composition of Infrastructure Stock Values Varies with Country Income 1.3 Mobile Cellular Subscriptions and Telephone Lines in Sub-Saharan Africa, 1991–2011 1.4 Private Participation in Infrastructure Versus Development Assistance, 1990–2011 1.5 South-South Non-OECD Investment Growth in Sub-Saharan African Infrastructure, 2001–2009 2.1 Urban Infrastructure Investment: Share of Total Investment and Share of GDP, 1952–2008 2.2 Urban Infrastructure Investment: Share of GDP by Region, 1990–2002 2.3 Urban Infrastructure Investment: Share of Total Infrastructure Investment by Region, 1990–2001 2.4 Distribution of Infrastructure Investments by Type and Region, 2003–2010 2.5 Infrastructure Investment Shift by Region, 1990–2010 2.6 Ratios of Investment in Major Urban Transport Sectors to Total Urban Infrastructure Investments, 2003–2011 2.7 Railway and Highway Mileages, 1988–2010 2.8 Distribution of Rapid Transit in China 2.9 Changes in Funding Sources for Urban Infrastructure Investment, 1980–2010 2.10 Infrastructure Investment as a Share of Central Government Spending, 1991–2006 2.11 Local Government Income from Land Sales, 1999–2008 2.12 Changes in the New Central Business District Area Over Time 2.13 Changes in the Huaqiao Area Over Time 2.14 Changes in the Yantian Area Over Time 2.15 Land Use Changes in Shenzhen, 1994–2005 2.16 Zhengzhou Eastern New CBD C2.1 Life Cycle of U.S. Railroads C2.2 The Magic Bullet C2.3 Bird’s-Eye View of Minneapolis, 1865 C2.4 Bird’s-Eye View of Minneapolis, 1879 C2.5 Bird’s-Eye View of Minneapolis, 1885 C2.6 Bird’s-Eye View of Minneapolis, 1891 3.1 Estimated Mobile-Cellular Subscriptions Worldwide, 2001–2011 3.2 Estimated Mobile-Cellular Subscriptions in Select Regions, 2011

5 6 10 12 12 23 24 25 26 27 28 28 29 30 31 32 40 42 44 46 50 54 55 57 58 59 60 63 64

xii

Illustrations

3.3 3.4 3.5 3.6 3.7 3.8 3.9 4.1 4.2 4.3 4.4 4.5 5.1 6.1 6.2 6.3 7.1 7.2 7.3 7.4 8.1 8.2 8.3 8.4 10.1 11.1 11.2 11.3 C12.1 13.1 C13.1 C13.2 C13.3 14.1

Mobile Phone Mast in Nomadic Lands, Mali, 2005 Celtel Advertisement in N’Djamena, Chad, 2007 MTN and the “Right to Communicate,” Bamenda, Cameroon, 2009 Young Men Selling Airtime for Celtel in N’Djamena, 2007 Young Men Selling Airtime for Zain in N’Djamena, 2009 Appolinaire in His Shop, 2012 Mobile Telephony in Reach of Everybody? Women Carrying Water in Darfur, 2007 Capital Intensity of Major Companies and Public Utilities for 2012 Value Added by Utility Infrastructure to the U.S. GDP, 1977–2011 Compensatory Pricing for Utility Monopolies Regulatory Tools and Incentives Regulatory Lag and Incentives Appraisal Approaches and Techniques Used to Value Utilities and Transportation Companies RELU-TRAN Zones for Chicago MSA Network of Major Roads in RELU-TRAN2 Fuel Intensity Versus Speed Map of Chicago Skyway and Environs Main Span of Chicago Skyway 1960 Chicago Skyway Promotional Map Actual and Projected Traffic on the Chicago Skyway, 1958–1990 Wrigley Field U.S. Cellular Park Bird’s Nest, Water Cube, and Olympics Sports Center in Beijing Soccer City Near Johannesburg C40 Cities’ Mitigation Measures Transboundary Infrastructure GHG Emissions Footprint (TBIF) for Denver Typical Strategies for GHG Mitigation Initiated by the Three Actor Categories Relative Impacts of Key Pathways for Near-Term GHG Mitigation for Cities U.S. Manufacturing Employment Trends, 1939–2011 Relationship Between Growth and the Ratio of Maintenance to New Investment Influence of Life Cycle Phases on Infrastructure: Road Network Example Vehicle Operating Cost as a Function of Pavement Condition on Pavement Serviceability Index Scale of 0 (Failed) to 5 (Excellent) VOC Components Constraints Identified by Firm Managers as Most Difficult

65 67 68 70 71 74 80 94 95 103 107 108 144 158 159 164 192 195 196 198 225 226 227 228 275 300 304 307 343 353 367 368 369 387

PREFACE This volume, based on a conference held in Cambridge, Massachusetts, in June 2012, addresses the links between infrastructure and land, particularly in urban areas. While infrastructure is as old as cities, technological changes and public policies on taxation and regulation produce new issues worthy of analysis, ranging from megaprojects and greenhouse gas emissions to private participation and involuntary resettlement. This is the seventh in a series of volumes that address land policy as it relates to a range of topics including climate change, municipal revenues and value capture, fiscal decentralization, and property rights. In addition to the authors and conference participants, many others have contributed to the design of the conference and the production of this volume. We thank Armando Carbonell, Martim Smolka, and Joan Youngman for their advice on the selection of topics and on program design. The conference would not have been possible without the logistical support of our conference event team, comprising Brooke Burgess, Sharon Novick, Cindy Moriarty, and Melissa Abraham. Our special thanks go to Emily McKeigue for her overall management of the production of this volume, to Vern Associates for the cover design, and to Nancy Benjamin and Judith Riotto for their tireless and reliable copyediting. Gregory K. Ingram Karin L. Brandt

xiii

Infrastructure, Land, and Development

1 Global Infrastructure: Ongoing Realities and Emerging Challenges Gregory K. Ingram and Karin L. Brandt

M

ore than 50 percent of the global population resides in urban areas where the interactions between land policy and infrastructure facilitate economic opportunities, affect the quality of life, and condition patterns of urban development. Infrastructure drives economic and social activities and represents the “wheels” of economic activity. Transportation and telecommunications facilitate business and trade, while energy and water are necessities for production processes. Transportation connections within urban areas expand labor markets, and the provision of basic needs to households—water, sanitation, and electricity—prevent the spread of disease and increase life spans. Infrastructure facilities provide the spatial skeleton supporting the location of residences, commerce, industry, and governmental activities. As cities grow, the demands on infrastructure facilities and services that support economic activity increase. For urban areas, the challenges of balancing growth with infrastructure development and maintenance are reflected in debates about infrastructure’s finance, regulation, and location and about the sustainable levels of its services. Infrastructure sectors include energy (electricity and natural gas); telecommunications (fixed phone lines, mobile phone service, and Internet connectivity); transportation (airports, railways, roads, waterways, and seaports); and water supply and sanitation (piped water, irrigation, and sewage collection and treatment). Infrastructure services have technical features such as economies of scale and economic features including externalities and spillovers from users to nonusers that make many of them difficult to provide as a normal private good. Because of these attributes, much infrastructure is either publicly provided or 3

4

Gregory K. Ingram and Karin L. Brandt

privately provided with regulatory oversight. Infrastructure also delivers economic and poverty-alleviation benefits when it responds to demand and is provided efficiently. Infrastructure provision is a major determinant of the location of economic activities and of the spatial pattern of development of the built environment. Transport infrastructure in the form of canals, railroads, and highways provides improved access to adjacent land, which in turn is developed by businesses seeking to reduce the costs of shipping inputs to their plants and outputs to their customers. Individuals similarly prefer locations well served by transport in order to widen their labor market opportunities and to ease their shopping and social interactions. Infrastructure services such as energy, water supply, and sanitation make the locations they serve more attractive. Community leaders spend much time devising zoning systems, property taxes, impact fees, and extensions of infrastructure services that strongly affect development patterns in urban areas. In developing countries, the challenge is to provide infrastructure before development occurs because it is much less costly to do so beforehand than to retrofit services in an already developed area. The historical development of new infrastructure sectors and services has been driven by technological advances that respond to existing needs. Roads and canals are the world’s oldest infrastructure sectors. In the nineteenth century, new infrastructure technologies lowered the costs and increased the demand for transportation services. In the United States, extensive networks of canals were constructed in the early nineteenth century, soon to be followed and replaced by railways in the mid-nineteenth century during the Industrial Revolution. The paving material of stone and soil macadam improved road quality, allowing for greater vehicle weights and higher speeds. Around the same time, poor sanitation in cities led to the spread of diseases like cholera. The need in urban areas for fire prevention and household sanitation led to the extension of the water supply and sewer lines. The development in the 1880s of alternating current and long-distance transmission enabled larger generators to take advantage of scale economies, reducing the cost of electricity. Telecommunication expansion saw the spread of the telegraph in the second half of the nineteenth century, the diffusion of the land-line telephone system in the first half of the twentieth century, and now the adoption of mobile phones and the Internet in the past two decades. Newly introduced infrastructure technologies expand their networks quickly because of their lower costs and often higher-quality service, displacing previous competing services. In the United States, railways and telegraphs quickly expanded their network coverage at a similar rate, achieving their mature size in about 45 years (figure 1.1). Railways were a substitute for canals, and the telegraph was later largely displaced by telephone services. Although the spread of telephone land lines was briefly slowed by the Great Depression, their expansion continued, and their dominance is now being threatened by mobile phones. The success of mobile phones is partly based on service convenience, but cost reductions have also played an important role. For example, the investment cost

global infrastructure: ongoing realities and emerging challenges

5

Figure 1.1 Growth of U.S. Infrastructure as a Percentage of Maximum Network Size, 1850–2005 100 90 80

Percentage

70 60 50 40 30 20 10 1850 1855 1860 1865 1870 1875 1880 1885 1890 1895 1900 1905 1910 1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005

0 Railways Telegraphs

Phones Oil pipelines

Internet Roads

Sources: Grübler (2003); Norris (2002).

associated with a mobile phone subscription fell from around $700 in 2000 to only $100 today (Chatterton and Puerto 2005). Since the mid-1990s, the Internet Figure network has been growing at a faster rate1.1than other recent infrastructure techLincoln_Ingram_Infrastructure nologies, and its price continues to fall while its service quality increases. The total value of infrastructure facilities across countries varies proportionally with country incomes. Moreover, the sectoral composition of infrastructure varies significantly across country income groups (figure 1.2). As a result, countries with high economic growth rates need to invest higher shares of country income in infrastructure facilities to maintain the balance between infrastructure and output. Industrialization and income growth increase the demand for energy to support development, and the energy sector grows more rapidly with income than other sectors, such as water and sanitation. Roads make up a fairly constant share of total infrastructure value across all country income groups. While infrastructure technology continues to advance, many debates about infrastructure also persist. A key debate concerns the extent of infrastructure’s impact on economic growth, urban development, and social development. Private investment and development assistance for infrastructure have increased in the past two decades, partly in response to evidence showing high rates of return for

6

Gregory K. Ingram and Karin L. Brandt

Figure 1.2 Composition of Infrastructure Stock Values Varies with Country Income 100 90

Percentage of stock value

80 70 60 50 40 30 20 10 0

Low

Lower-middle Upper-middle Country income group Energy

Roads

Telecommunications

High Water and sanitation

Sources: International Telecommunications Union (2010); World Bank (2012).

infrastructure investment. An ongoing debate in the United States concerns the adequacy of infrastructure maintenance and how it should be financed. Debates also continue about other aspects of infrastructure, such as its regulation and taxation. In many countries, the private sector is replacing the public sector in providing infrastructure services. This requires the1.2public sector to take on an explicit Figure regulatory role that often remains implicit when services are publicly provided. Lincoln_Ingram_Infrastructure In developing countries, private participation in infrastructure (PPI) investment has dramatically increased since 1990, and new sources of finance have become available, especially for well-defined projects such as power-generation facilities or mobile phone franchises. Large infrastructure projects continue to be a challenge for infrastructure provision—with political complications and detrimental social impacts that include involuntary resettlement. Mega-events, such as the Olympic Games, often catalyze large infrastructure investment programs that have had decidedly mixed postevent consequences. Finally, climate change concerns have directed attention to improving the sustainability and efficiency of infrastructure.

Infrastructure, Land, and Development Infrastructure looms large in terms of a country’s share of annual investment, its share of the total stock of capital in an economy, its contribution to economic

global infrastructure: ongoing realities and emerging challenges

7

growth, its effect on human welfare, and its influence on spatial development patterns. Annual investment in infrastructure ranges across countries from a low of around 2 percent of gross domestic product (GDP) up to 8 percent—or even higher in rapidly growing economies. The public sector plays a large role as a direct provider of investment, as a provider of services, and as a regulator of services provided by itself and others. At the same time, the share of private investment in infrastructure can be large, ranging up to half or more of annual infrastructure investment in some countries. The value of cumulative infrastructure investment—the physical stock of all existing infrastructure—is also large, with its value averaging around 60 percent of GDP across countries (Ingram, Liu, and Brandt 2013). Infrastructure services increase economic productivity by offering firms and households efficient and inexpensive services that enable them to reduce their transportation, energy, water, sanitation, and communication costs. Lower transportation costs reduce price differences over space and underpin some of the earliest theories about the distribution of land use and agriculture around urban centers (von Thünen 1966). Adequate infrastructure is seen as a key determinant of international competitiveness because it can reduce production costs, ease international transportation, and even facilitate the availability of labor and specialized services to firms. Economists use several approaches to measure the productivity impacts of infrastructure. At the microeconomic level, infrastructure project rates of return are often substantial and higher than those in other sectors such as agriculture and human services (World Bank 1994). At the macroeconomic level, production functions, cost functions, and cross country studies are the main approaches used to estimate the productivity of infrastructure. Production functions estimate the contribution of infrastructure to national output, such as GDP, typically in a single country over time. The results are often summarized in terms of the percentage increase in national output that is associated with a 1 percent increase in the amount of infrastructure (the elasticity of output with respect to infrastructure). The amount of infrastructure is measured either by constructing a perpetual inventory (adding up depreciated annual investment amounts from the past) or by using current data on physical quantities of infrastructure, such as miles of paved road or installed generating capacity, multiplied by their unit cost. Data for the perpetual inventory approach are not as widely available as data on physical quantities of infrastructure, and private investment is particularly elusive, so the production function approach tends to focus on public investment. Some analysts argue that the costs of public infrastructure investment may include extra expenditures for inefficiencies such as overcapacity or side payments and therefore are biased upward (Pritchett 2000). Cost functions explore the extent to which infrastructure reduces costs at the firm level, and these are also usually estimated based on data from a single country over time. Cross country studies relate national amounts of infrastructure (typically using physical quantities) to national outputs, usually at a single point in time.

8

Gregory K. Ingram and Karin L. Brandt

Table 1.1 A Majority of Studies Find That Infrastructure Increases Productivity Infrastructure Effect (% of studies) Overall results By infrastructure proximity Public capital Physical indicator By approach Production function Cost function Cross country regression

Number of Studies

Negative

None

Positive

5.7

31.4

62.9

140

10.8 1.3

40.0 24.0

49.2 74.7

65 75

2.9 7.7 13.8

36.2 15.4 37.9

60.9 76.9 48.3

69 13 29

Source: Straub (2008).

Straub (2008), who reviewed and summarized 140 empirical studies examining the effect of infrastructure on productivity, reported that nearly two-thirds of the studies found a positive relation (table 1.1). Studies using physical measures of infrastructure were more likely to find positive effects than those based on accumulated past investment. Typical results using production functions and physical measures report elasticities of output with respect to infrastructure around 0.1 (Calderon, Moral-Benito, and Serven 2011). This indicates that the rate of return to infrastructure investment is around 16 percent if infrastructure stocks average 60 percent of GDP, meaning an investment that increased infrastructure stocks by 1 percent would cost 0.6 percent of GDP and increase output by 0.1 percent of GDP. Most analyses of returns to infrastructure focus on countries, and few estimate returns to metropolitan areas. However, investment data are available for two unique metropolitan areas: the special administrative region of Hong Kong and the city-state of Singapore. Their average economic growth rates from 2000 to 2010 are 4.4 percent and 5.9 percent, respectively. Between 2007 and 2010, Hong Kong’s infrastructure expenditure averaged 2.56 percent of its GDP, and Singapore’s averaged 6.44 percent, supporting the view that higher growth rates are associated with higher infrastructure expenditures (albeit over a short period of time in these cases). As Yan Song points out in chapter 2, China—which has one of the most rapidly growing economies—proposes to invest around US$1.03 trillion on urban infrastructure during its 12th Five-Year Plan from 2011 to 2015. This is slightly more than 3 percent of its total GDP projected

global infrastructure: ongoing realities and emerging challenges

9

over the five-year period—a share larger than many countries routinely invest in all national infrastructure needs. In addition to increasing economic productivity, infrastructure promotes human welfare by reducing the cost of accessing markets: helping farmers get their crops to market and workers get to jobs. Infrastructure investment also shapes urban growth patterns, which are linked to transportation networks that serve local demand and connect cities to expand labor markets and increase incomes. In developing countries, rural roads also allow easier access to health care and education, facilitating human capital formation. Adequate infrastructure can reduce consumption costs for households. For example, the cost of piped water in underserved informal settlements is a fraction of the cost of trucked-in water, and for illumination, electric lights cost much less than candles. Household consumption of infrastructure services improves household productivity, but measuring this productivity increase is difficult because household production is not included in standard measures of national product. Thus, increases in household productivity are not normally included in estimates of the productivity impacts of infrastructure, such as those summarized in table 1.1. Infrastructure is likely to have collateral benefits that may be of particular importance to the poor. Improved infrastructure may benefit the poor by increasing the value of their assets. For example, improved rural transportation can increase the value of agricultural land in rural areas, much of which may be owned by low-income households. Better and safer roads raise school attendance; access to electricity allows more time to study and makes computer use possible; and improved water and sanitation reduce child mortality. Empirical work is finding that infrastructure development is associated with reduced income inequality (Calderon and Serven 2008; Lopez 2004). Infrastructure improvements may not only increase output, but may also provide a host of benefits to low-income households. For example, infrastructure services create developmental and social impacts at the individual level. Clean water and sanitation service have improved the health and extended the life expectancies of urban residents, while social relationships have been transformed by telecommunications, particularly mobile phones in developing countries. Communication in rural areas in the global south is changing rapidly as mobile phones supplant fixed phone lines. This change is most notable in sub-Saharan Africa (figure 1.3). The revolution in mobile telephony that is sweeping through Africa has made it much easier for residents in rural areas and small towns to become knowledgeable about market prices, to learn of opportunities in other locations, and to keep in touch with distant relatives and friends. The mobile technology itself creates new jobs: running charging stations where electricity is not readily available, selling mobile credit for minutes, and repairing mobile phones. This transformation is changing relationships as people once spatially isolated become connected through mobile communication. In chapter 3 Mirjam de Bruijn illustrates this impact with anthropological case studies in four sub-Saharan African countries.

10

Gregory K. Ingram and Karin L. Brandt

Figure 1.3 Mobile Cellular Subscriptions Bypass Telephone Lines in Sub-Saharan Africa, 1991–2011 60 Mobile cellular subscriptions Telephone lines

Percentage

50 40 30 20 10

12

11

20

10

20

09

20

08

20

07

20

06

20

05

20

04

20

03

20

02

20

01

20

00

20

99

20

98

19

97

19

96

19

95

19

94

19

93

19

92

19

19

19

91

0

Source: World Bank (2012).

Finance, Regulation, and Taxation The economic and social benefits of infrastructure are great, but so are the costs of developing and maintaining infrastructure. Governments continue to experiment with various financing mechanisms, including fee-for-service pricing arrangements and related revenue-collection schemes. Finance methods vary widely Figure 1.3 across infrastructure sectors and countries, although user charges, subsidies from Lincoln_Ingram_Infrastructure general funds, and borrowing are the primary sources used to pay for operating costs and investment. User fees are common in many sectors, including telephone use, power consumption, transit fares, and travel on toll roads. Examples of indirect cost recovery include fuel taxes, vehicle registration fees, some property taxes, and land-based revenues. Borrowing from public and private lenders is advantageous given the long life of many infrastructure facilities. Industrialized countries have long used sovereign bond finance to fund hydroelectric dams and road systems, and municipal bonds to finance local infrastructure. These practices are being adapted to the needs of developing countries. For example, Johannesburg’s “Jozi bonds” support the provision of infrastructure services in poorly serviced townships that were incorporated into municipalities following the end of apartheid. The need for innovative financing schemes to support rapid growth has drawn attention to the use of land-based finance. Land value capture—the practice of using land value increases associated with service provision to finance infrastructure—was the subject of the Lincoln Institute’s previous Land Policy

global infrastructure: ongoing realities and emerging challenges

11

Series volume.1 Value capture is a common practice in Latin America, and variations of this method are used in other countries. In Bogotá, Colombia, betterment levies have supported infrastructure development since the 1930s, including roads, water and sewer, and more recently sidewalks and public parks (Borrero et al. 2011). Transit companies in Hong Kong and Tokyo used value capture to finance transit projects with revenue from the codevelopment of residential and commercial areas served by transit, and Ahmedabad, India, is known for its Town Planning Scheme, which uses a land readjustment process to finance infrastructure for newly urbanized land (Ingram and Hong 2012). Proceeds from the sale of urban land is a major revenue source used to fund infrastructure in China, and the sale of development rights is a growing source of revenue in Brazil. While high- and middle-income countries are more concerned with financing infrastructure maintenance or infrastructure that supports growth, many lowincome countries face large infrastructure needs. In developing countries, the projected annual cost of infrastructure investment (US$450 billion) and maintenance needs (US$305 billion) totals US$755 billion, which is nearly 5 percent of the countries’ aggregate GDP (Ingram, Liu, and Brandt 2013). External funding commitments for infrastructure have grown substantially and now cover about 46 percent of projected investment in developing countries. This total comprises bilateral development assistance (US$22 billion in 2010); regular and concessional financing from multilateral development banks (US$23.7 billion in 2010); and private participation in infrastructure (US$161.7 billion in 2011) (figure 1.4). Some developing countries are now investing in infrastructure in other developing countries. This so-called south-south financing has been growing, as enterprises in one country invest in those in other countries, a practice that has been under way for some time in Latin America. More recently, some developing countries, especially China, have been investing in infrastructure projects in sub-Saharan Africa, often in arrangements where loans are repaid with commodity exports (figure 1.5). The movement in developing countries toward greater private sector involvement in infrastructure development is not unprecedented. Private finance and ownership of infrastructure—including railroads, transit, and canals—were common in the nineteenth century. Much infrastructure finance shifted to public sources in the mid-twentieth century and then back to public-private partnerships starting in the 1980s. In Europe, power and telephone services were largely publicly provided, whereas they have been privately provided in the United States and regulated in terms of service quality, universal service obligations, and price (usually based on the utility’s rate of return). The United States has more than 100 years of experience with the regulation of utility infrastructure, as Janice A.

1. See the sixth volume in the Land Policy Series, Value Capture and Land Policies, ed. Gregory K. Ingram and Yu-Hung Hong (Cambridge, MA: Lincoln Institute of Land Policy, 2012).

Figure 1.4 Private Participation in Infrastructure Versus Development Assistance, 1990–2011 200,000 180,000

Amount (million US$)

160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000

Bilateral development assistance

11

10

20

09

20

08

20

07

20

06

20

05

20

04

20

03

20

02

20

01

20

00

20

99

20

98

Private participation in infrastructure

19

97

19

96

19

95

19

94

19

93

19

92

19

91

19

19

19

90

0

Multilateral development banks

Sources: OECD (2012); World Bank (2012); World Bank and Public-Private Infrastructure Advisory Facility (2011).

Figure 1.5 South-South Non-OECD Investment Growth in Sub-Saharan African Infrastructure, 2001–2009

Amount (billion US$)

15

Figure 1.4 Lincoln_Ingram_Infrastructure

12 9 6 3 0 2001

2002

2003

Private participation in infrastructure

2004

2005 Official development assistance

2006

2007

2008

non–Organisation for Economic Co-operation and Development

Sources: Foster et al. (2008); World Bank (2012); World Bank and Public-Private Infrastructure Advisory Facility (2011).

12

2009

global infrastructure: ongoing realities and emerging challenges

13

Beecher points out in chapter 4. The first U.S. regulatory agency for infrastructure was the Interstate Commerce Commission (ICC), which was established in 1887 to address railway price discrimination. Other regulatory agencies then followed to control monopoly power and pricing. Technological advances and the advent of competing services (e.g., trucking that competes with rail) have led to deregulation. Thus, the ICC was abolished in 1995, and deregulation has advanced in telecommunication, trucking, and air travel in the United States. Privately owned infrastructure service providers in the United States, such as airlines, gas and electric utilities, railroads, water companies, and telecommunications firms, are typically subject to property taxation, often at rates that are much higher than residential rates. This is an often-overlooked topic in property taxation, and most service users are unaware that their utility bills cover such taxes. This tax base is of great importance to any local government that hosts utility plants and equipment. Utility property is generally assessed by state agencies using methods that differ markedly from those employed by local assessors, with results that are often controversial and complex, as noted by Gary C. Cornia, David J. Crapo, and Lawrence C. Walters in chapter 5. Some utilities pay a substantial portion of their revenues in property taxes, such as Consolidated Edison Co. of New York, which pays 12 percent of its gross operating revenues in property taxes. Just as physical infrastructure affects urban development patterns, the taxation of motor vehicle use and the pricing of road infrastructure shape land policy by affecting location choices and land values. Researchers and planners have debated the impact of tolls, congestion charges, and gasoline taxes on metropolitan development patterns. The locational setting and spatial coverage of toll systems, congestion pricing, or fuel and vehicle taxes have implications for residential and employment location. In theory, congestion charges focused on central locations could either attract jobs and residents to these areas or displace them. Careful simulations by Alex Anas in chapter 6 suggest that congestion tolls in central areas are likely to move jobs and residents to suburban areas, whereas a metropolitanwide gasoline tax is likely to cause both jobs and residents to concentrate in the city.

The Challenges of Large Projects Extremely large infrastructure projects, such as a new airport, subway system, or ring road, can produce large changes in land values, location decisions, and development patterns. The United States has seen many such projects in recent decades, including the Washington, DC, Metro, the Denver airport, and Boston’s depression of the Central Artery (“the Big Dig”). These were all multibillion-dollar projects whose spatial impacts are still under way long after their completion. For example, the construction of residential, commercial, and office space and the general concentration of economic activity around Metro stations in Washington is still ongoing, 25 years after the bulk of the Metro system was opened.

14

Gregory K. Ingram and Karin L. Brandt

In addition, infrastructure megaprojects are frequently well over budget. The new Denver airport’s final cost was nearly three times its original cost estimate of $1.7 billion, and the Big Dig’s cost of $14.6 billion was three and a half times its original cost estimate (all in 2003 dollars). Because megaprojects take a long time to build and their construction spans several terms of governors, mayors, or department secretaries, they inevitably become steeped in political lore. Most megaprojects are long-lived after they begin operation, and their public and private financing and operation take many forms, as facilities can be transferred back and forth between public and private operators. A common publicprivate instrument is the flexible build-operate-transfer (BOT) scheme. For example, Kuala Lumpur, Malaysia, has developed three urban rail lines, each with varying BOT agreements and government guarantees for domestic debt. While the financing instrument is important, the outcome of such projects is dependent on numerous factors, many related to the political environment, which facilitate or hinder implementation. In recent years large public facilities have been sold to private firms, a transfer often motivated by government’s need for funds. In chapter 7 Louise Nelson Dyble examines the history of political leadership throughout the Chicago Skyway’s development as a public project and assesses the lessons from its eventual privatization. Like megaprojects, mega-events such as the Olympics or the World Cup are associated with large infrastructure investments and promises of high economic returns. However, the evidence from host economies shows that these sporting events typically bring high costs with low rewards, as Victor A. Matheson notes in chapter 8. For the London 2012 Summer Games, a fifth of the total budgeted cost for the new Wembley Soccer Stadium was for infrastructure improvements, including new roads and a renovated transit station to accommodate Olympic traffic. Mega-event facilities may even serve one event’s single purpose and not be suitable for other regular events. For the Beijing 2008 Summer Games, the National Aquatic Center, or “Water Cube,” opened for public swimming after the Olympics and was later transformed into a large and extraordinarily expensive water park. Some cities have reused sports infrastructure by modifying them to serve universities or professional teams. Nonetheless, the high costs of these short-term mega-events raise questions of economic feasibility for host cities in developing countries. A perennial issue with megaprojects and mega-events is the involuntary resettlement and forced displacement that are often associated with infrastructure development. Mega-events involve large stadium and transportation projects to support an influx of visitors, and they often cause the displacement of local, lowincome residents. An estimated 720,000 low-income workers, renters, and squatters were forcibly evicted to make room for the 1988 Olympic Games in Seoul, South Korea (Davis 2007). In Atlanta, Georgia, the controversial demolition of Techwood Homes, one of the first U.S. public housing projects, located between the Olympic venue and the Georgia Institute of Technology campus, displaced all

global infrastructure: ongoing realities and emerging challenges

15

of its residents. Similarly, with mega-projects, particularly dams, the objectives of infrastructure development and human development frequently collide, as Robert Picciotto notes in chapter 9. For example, Ghana’s Akosombo Hydroelectric Project created the world’s largest artificial lake, Lake Volta, which covers nearly 4 percent of the country’s land and displaced 80,000 people, nearly 1 percent of the population. The dam provides a substantial amount of electric power to the West African nation and its neighbors, Togo and Benin. The reservoir behind the Three Gorges Dam in China displaced over a million people as of 2008. Differences between resettlement standards promulgated by international agencies and the practices applied by many governments remain unresolved.

Improving Sustainability and Efficiency Infrastructure facilities and their services produce a variety of spillovers and externalities for nearby activities, for metropolitan areas, and for the environment. Electric power and transportation are energy-intensive sectors that produce emissions that have local consequences (particulates and smog) and global consequences (greenhouse gases). Because of the large scale of emissions, infrastructure will bear a major burden when it comes to reducing emissions, but this burden can become an asset when credits are granted for emission reductions—an outcome that is being realized in some countries. Infrastructure also produces spatial externalities because its location often helps determine the location of residential, commercial, and industrial activities. Accordingly, locating infrastructure is an instrument that can be used to affect spatial development patterns in both urban and rural areas. Infrastructure also has fiscal externalities because its influence on the location of economic activities has consequences for local employment and tax revenues. Finally, because most infrastructure facilities have long lives, the management and stewardship of infrastructure assets are a critical determinant of their efficient use. Sustainability increasingly affects infrastructure investment patterns, particularly in cities in developing countries where urban populations will more than double—increasing by 2.6 billion people—by 2050. Some governments provide economic incentives to support long-term investments in sanitation and drainage systems that include green infrastructure such as green roofs, rain gardens, and permeable pavements. Energy-efficient power generation from wind, solar, and water sources is growing, and improved technology will decrease prices. Institutional partnerships are critical in creating finance strategies for sustainable infrastructure in cities, as illustrated by Katherine Sierra in chapter 10 with three case studies of cities facing climate change challenges. The regulation of carbon emissions has created a market for financing sustainable energy and transportation infrastructure in developing countries with carbon credits—payments for activities that sequester carbon or reduce emissions. Uganda’s West Nile Electrification Project (WNEP) is the World Bank’s first sub-Saharan project

16

Gregory K. Ingram and Karin L. Brandt

to issue carbon credits. The WNEP includes a 3.5-megawatt hydroelectric power plant and a 1.5-megawatt heavy fuel oil–fired power plant that issued more than 20,000 carbon credits. Other strategies identify environmental needs and form responsive public-private partnerships such as the Arab Financing Facility for Infrastructure (AFFI). The Middle East and North Africa region historically receives the smallest share of private participation in infrastructure, and the AFFI seeks to address the pressing needs of rapid population growth and unmet investment needs by supporting public infrastructure services and public-private partnerships that follow Islamic-compliant financing. Analyzing the economic and environmental impacts of urban infrastructure—whether its contribution to productivity or its generation of greenhouse gas emissions—is very challenging because urban or metropolitan infrastructure is seldom self-contained. Cities almost always import infrastructure services (e.g., power and water) from outside their boundaries, and allocating the costs and benefits of intercity transportation to specific cities is a difficult accounting problem. Accounting for emissions from industrial production and household consumption raises the issue of how to assign the carbon emissions of intermediate inputs and final production goods that originate outside the city. Promising approaches that address this problem specify a boundary around the city or metropolitan area and carefully analyze all transboundary flows of goods and services to analyze both infrastructure and all other movements, as noted by Anu Ramaswami in chapter 11. Results indicate that improvements in energy efficiency and reductions in greenhouse gas emissions will be achieved more readily for buildings (responsible for nearly half of urban greenhouse gas emissions) than for transportation (responsible for nearly a quarter of greenhouse gas emissions). The location of infrastructure has a major influence on the location of residential, commercial, and industrial activity. In the United States, much has been written about the relationship between highway networks and the spread of lowdensity residential and commercial development. Less obvious is the major role that infrastructure has played in the location decisions of industrial firms. The growth of trucking and the spread of highway facilities in the United States unmoored many industrial firms from locations next to rail sidings or port facilities. The growing use of assembly lines and related production processes that work best in a one-story rather than a multistory building increased the land area required for many manufacturing facilities. Accordingly, the use of trucking and the need for larger sites stimulated industrial firms to move to suburban and even exurban locations along interstate highways. This move has reduced city tax revenue and employment in many cities as industrial enterprises and their associated suppliers relocated. Some analysts think that a trend toward a third industrial revolution, based on smart manufacturing, will provide incentives for smaller industrial firms to relocate to developed urban areas and that cities need to plan for this possibility by preserving areas for future industrial development,

global infrastructure: ongoing realities and emerging challenges

17

as Nancey Green Leigh notes in chapter 12. If cities fail to do this, they will have little success in attracting these new smart manufacturing firms and their associated jobs. To increase productivity and to achieve related environmental goals, infrastructure facilities must produce high-quality services efficiently. This means that the facilities must be maintained so that they achieve their planned useful lives and targeted operating costs. But adequate maintenance faces many challenges, particularly in the United States. Maintenance has large economic returns and reduces long-term investment requirements and consumer costs, but it is often undervalued by revenue-constrained governments. The most recent annual infrastructure “report card” from the American Society of Civil Engineers (2009) issued an overall grade of D for delayed maintenance and underfunding in many categories and estimated that poor road conditions impose costs of $67 billion annually on U.S. motorists. Moreover, repairing neglected roads is two to three times more costly than performing appropriate ongoing maintenance. In chapter 13 Felix Rioja reviews research findings and empirical studies that indicate that optimal levels of maintenance expenditures can increase a country’s growth rate and have significant and positive effects on productivity. He also sheds light on why industrialized and developing governments neglect maintenance despite its positive effects. While research offers some promising insights about how to improve maintenance and the overall performance of infrastructure provision, countries may be able to learn from one another how to improve performance or may even be able to transfer lessons from their strongly performing infrastructure sectors to those with weaker performance. As noted earlier, the value of infrastructure stocks across countries is closely associated with country incomes. However, the performance of the various infrastructure sectors is not strongly associated with country incomes. This is not because some countries are just better at managing infrastructure than others. It is because the performance across infrastructure sectors varies greatly within countries. The low correlation of performance within countries means that if a country is performing well in one sector, such as electric power, it tells one very little about how well it is performing in another sector, such as providing telephone service. Table 1.2 uses data from 83 countries to show that infrastructure stock levels are highly correlated across sectors, but performance in one sector is weakly correlated with, and is a poor predictor of, performance in other sectors within countries. In chapter 14 George R. G. Clarke explores some reasons for this lack of correlation of performance across sectors, and he also assesses which sector’s performance is most important to private firms.

18

Gregory K. Ingram and Karin L. Brandt

Table 1.2 Stock Levels Are Highly Correlated, But Performance Varies Within Countries Correlation Matrices Performance Levels (average correlation = 0.34) Phone-Line Faults Phone-line faults Electric loss Unpaved road share Lack of cell service

Electric Loss

Unpaved Roads

Lack of Cell Service

1.00 0.42 0.22

1.00 0.44

1.00

Phone Lines

Generating Capacity

Paved Road Share

Mobile Subscriptions

1.00 0.86 0.83 0.82

1.00 0.81 0.75

1.00 0.66

1.00

1.00 0.01 0.22 0.75

Stock Levels (average correlation = 0.79) Phone lines Generating capacity Paved road share Mobile subscriptions Source: Authors’ calculations.

Conclusions Infrastructure, characterized by Jawaharlal Nehru as a principal element of the commanding heights of the economy, continues to have a key role in economic development. The value of infrastructure facilities increases in step with national income, which is now growing more rapidly in developing countries than in industrial countries. The unprecedented expansion of the urban population of developing countries will require substantial investments in the infrastructure of cities for the next 40 years. At the same time, there may be surprises that stem from technological change, such as when a new technology displaces or reduces the prominence of another. For example, in the past two decades we have witnessed the explosive growth of mobile phones relative to fixed-line telephony. Compelling evidence indicates that infrastructure investment often increases economic productivity, although this is not true for poorly conceived projects, such as bridges to nowhere. Growing cities in the United States that face demands on infrastructure facilities and services can use land policy to increase density and make use of economies of scale. Infrastructure, particularly transit, has a strong relationship with density, and reports show that cost-effective transit requires density levels of 15 people

global infrastructure: ongoing realities and emerging challenges

19

per hectare (Angel 2012). In the United States the development of highways promoted density reductions of households (Alonso 1964) and of industrial firms (Moses and Williamson 1967). This pattern of urban development has generated ongoing challenges to modify existing zoning, taxation, and fee structure systems that support low-density development, though some U.S. cities are now gradually increasing density levels. The lesson for cities in developing countries is to build infrastructure and implement land policy that support dense patterns of urban development, a point highlighted in chapter 2’s case study of Chinese cities. More recent research is finding that inadequate infrastructure is associated with income inequality. This is likely linked to the benefits that good infrastructure services deliver to households in the form of direct health benefits, improved access to education, and enhanced economic opportunities. Because much infrastructure is energy intensive, efforts to reduce greenhouse gas emissions will need to address infrastructure, particularly electric power and transport. It is difficult to foresee what impacts policies to reduce greenhouse gas emissions will have on these sectors, but they could have large effects on households and firms. Bringing the management of infrastructure up to levels of good practice has a large economic payoff, and performance levels vary dramatically between and within countries. Conveying the large economic returns from improving infrastructure performance, and particularly maintenance, to policy makers and voters is a necessary, but so far unmet, challenge.

references Alonso, W. 1964. Location and land use. Cambridge, MA: Harvard University Press. American Society of Civil Engineers. 2009. Report card for America’s infrastructure, 2009. Washington, DC: American Society of Civil Engineers. Angel, S. 2012. Planet of cities. Cambridge, MA: Lincoln Institute of Land Policy. Borrero, O., E. Durán, J. Hernández, and M. Montaña. 2011. Evaluating the practice of betterment levies in Colombia: The experience of Bogotá and Manizales. Working Paper. Cambridge, MA: Lincoln Institute of Land Policy. Calderon, C., E. Moral-Benito, and L. Serven. 2011. Is infrastructure capital productive? A dynamic heterogeneous approach. Policy Research Working Paper No. 5682. Washington, DC: World Bank. Calderon, C., and L. Serven. 2008. Infrastructure and economic development in subSaharan Africa. Policy Research Working Paper No. 4712. Washington, DC: World Bank. Chatterton, I., and O. S. Puerto. 2005. Estimations of infrastructure investment needed in the South Asia region. Washington, DC: World Bank. Davis, M. 2007. Planet of slums. New York: Verso Books. Foster, V., W. Butterfield, C. Chen, and N. Pushak. 2008. Building bridges: China’s growing role as infrastructure financier for Africa. Washington, DC: PPIAF, World Bank. Grübler, A. 2003. Technology and global change. Cambridge, U.K.: Cambridge University Press.

20

Gregory K. Ingram and Karin L. Brandt

Ingram, G. K., and Y.-H. Hong, eds. 2012. Value capture and land policies. Cambridge, MA: Lincoln Institute of Land Policy. Ingram, G. K., Z. Liu, and K. Brandt. 2013. Metropolitan infrastructure and capital finance. In Financing metropolitan governments in developing countries, ed. R. W. Bahl, J. F. Linn, and D. L. Wetzel. Cambridge, MA: Lincoln Institute of Land Policy. International Telecommunications Union. 2010. World telecommunications development report. Geneva: International Telecommunications Union. Lopez, J. H. 2004. Macroeconomics and inequality. World Bank Research Workshop, Macroeconomic Challenges in Low Income Countries (October). Moses, L., and H. F. Williamson Jr. 1967. The location of economic activity in cities. American Economic Review 57(2):211–222. Norris, P. 2002. The digital divide: Civic engagement, information poverty, and the worldwide Internet. Cambridge, U.K.: Cambridge University Press. Organization for Economic Cooperation and Development. 2012. Quick wizard for international development statistics, Paris. Pritchett, L. 2000. The tyranny of concepts: CUDIE (cumulated, depreciated, investment effort) is not capital. Journal of Economic Growth 5(4):361–384. Straub, S. 2008. Infrastructure and growth in developing countries: Recent advances and research challenges. Policy Research Working Paper No. 4460. Washington, DC: World Bank. von Thünen, J. H. 1966. The isolated state. Oxford, U.K.: Pergamon (Hamburg, Germany: Perthes, 1826). World Bank. 1994. World development report, 1994: Infrastructure for development. New York: World Bank and Oxford University Press. ———. 2012. World development indicators. Washington, DC: World Bank. World Bank and Public-Private Infrastructure Advisory Facility. 2011. Private Participation in Infrastructure (PPI) project database. Washington, DC: World Bank.

2 Infrastructure and Urban Development: Evidence from Chinese Cities Yan Song

I

nfrastructure services—including power, transportation, telecommunications, provision of water and sanitation, and safe disposal of wastes—are central to economic production and urban growth. It is commonly agreed that infrastructure plays an important role in stimulating urban land development and private economic activity (Démurger 2001; Gramlich 1994). The adequacy of infrastructure—which can contribute to diversifying production, expanding trade, coping with population growth, reducing poverty, or improving environmental conditions—helps determine a country’s success (World Bank 2004) by accommodating economic and urban growth (Calderon and Serven 2004). Infrastructure has been used as a tool to stimulate the growth of human settlements in many urban areas. Policy makers and planners have used infrastructure systems to attract private investments for housing and economic development. Despite this, the link between infrastructure and urban growth remains understudied, and infrastructure research has developed in isolation from the large literature on urban growth. This chapter discusses the links between infrastructure provision and urban expansion, the relationship between levels of infrastructure and land prices, and the mechanisms used to finance infrastructure. Data and case studies from developed and developing cities in China provide empirical evidence about the extent to which the provision of infrastructure affects urban development and shapes development patterns. China was chosen as the case study for this chapter because it provides sufficient dynamics and variation to enable the investigation of these research 21

22

Yan Song

questions. In three decades of market-oriented reforms, China has been one of the world’s fastest-growing economies, with per capita real incomes more than quadrupling since 1978. During this period, China has made substantial investments in infrastructure and has improved access to services such as safe water, sanitation, electric power, telecommunications, and transportation (J. Zhang 2011). Today, China is set to accelerate the construction of urban public infrastructure by investing as much as 7 trillion yuan (US$1.03 trillion) during its 12th Five-Year Plan from 2011 to 2015. The scale of infrastructure investment and the extent to which infrastructure has transformed the urban landscape in China might seem remarkable. However, problems persist in the form of insufficient provision of infrastructure, discrepancies in the level of infrastructure across regions, deficiencies in cost recovery, inadequate sources of financing, and the lack of incorporation of sustainable principles in shaping urban growth. This chapter describes these challenges in infrastructure provision in China and explores the causes of some of the existing problems.

An Overview of Infrastructure Development in China THE SCOPE OF INFRASTRUCTURE INVESTMENT IN CHINESE CITIES

Since implementing economic reform with the adoption of its opening-up policy in 1978, China has made substantial investments in infrastructure, improving access to services such as clean water, sanitation, electricity, telecommunications, and transportation (Economic Research Institute for ASEAN and East Asia 2007). Urban infrastructure investment in China has grown exponentially in the last several decades. From 1978 to 2008, urban infrastructure investment as a percentage of gross domestic product (GDP) fluctuated between 0.33 and 3.29 percent, while urban infrastructure investment as a percentage of total investment ranged between 1.79 and 8.02 percent (figure 2.1). Despite this marked increase, infrastructure investment in China relative to both GDP and total investment was lower than it was in other developing countries, as reported in a World Bank survey, and it was below the levels recommended by the United Nations (World Bank 2004), as shown in table 2.1. Although China spends about 50 percent of its GDP on fixed investment (compared to the world average of less than 20 percent or the U.S. figure of 15 percent), only a small percentage is spent on urban infrastructure. As shown in figure 2.1, China’s investment in infrastructure in recent years has increased, relieving economic and social development pressures caused by limited infrastructure. This investment has been driven by the high demand for infrastructure services, which has been fueled by steady economic growth. Increased public expenditures in infrastructure are also related, in part, to the proactive fiscal policy adopted by the government to minimize the impact of the financial crisis (Liu 2010). Many problems still exist with the current practice of infrastructure development. Inadequate infrastructure is evident in some regions,

Figure 2.1 Urban Infrastructure Investment: Share of Total Investment and Share of GDP, 1952–2008 9 8 7

Percentage

6 5 4 3 2 1 0 1952 1960 1968 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Urban infrastructure share of total investment

Urban infrastructure share of GDP

Sources: Ministry of Housing and Urban-Rural Development of China (2000–2011); National Bureau of Statistics of China (1996–2011).

Table 2.1 Figure 1950–1990 2.1 Infrastructure Investment Share in Developed Countries,

Lincoln_Ingram_Infrastructure Japan Germany

Infrastructure investment as a percentage of GDP Infrastructure investment as a percentage of fixed assets investment

USA (1950–1983)

(1960–1980)

(1976–1980)

Developing Countries Percentage Recommended by UN

Developing Countries Percentage Surveyed by World Bank (1980–1990)

1.2–1.8

2.1–4.2

1.7–1.9

3.0–5.0

2.0–8.0

6.0–10.2

6.4–12.9

7.3–9.0

>10.0

20.0

Source: World Bank (2004).

23

24

Yan Song

especially the western region, despite the recent increase in investment. In addition, insufficient investment in infrastructure could hinder urban and economic growth (Economic Research Institute for ASEAN and East Asia 2007). REGIONAL DISCREPANCIES IN URBAN INFRASTRUCTURE INVESTMENT

To analyze regional discrepancies in urban infrastructure investment, data are agglomerated for three urban regions—east, central, and west—to compare levels of infrastructure investment. For the purposes of comparison, the eastern area includes the cities of Beijing, Shanghai, Tianjin, Guangzhou, Nanjing, Shenyang, Qingdao, Jinan, Shenzhen, Xiamen, Dalian, Hangzhou, and Ningbo; the central area includes Wuhan, Changchun, and Ha’erbin; and the western area includes Chongqing, Chengdu, and Xi’an. Figures 2.2 and 2.3 reveal several regional patterns of infrastructure investment. First, as the more industrialized and developed part of the country, eastern cities entered an advanced stage of economic development with a relatively stable or slightly declining ratio of urban infrastructure investment to GDP. Around 1998, infrastructure investment in eastern cities started to drop after having risen Figure 2.2 Urban Infrastructure Investment: Share of GDP by Region, 1990–2002 14 12

Percentage

10 8 6 4 2 0 1990

1991

1992

1993

1994

1995

Urban investment in eastern region Urban investment in central region Urban investment in western region

1996

1997

1998

1999

2000

2001

2002

Urban investment by government in eastern region Urban investment by government in central region Urban investment by government in western region

Sources: Ministry of Housing and Urban-Rural Development of China (2000–2011); National Bureau of Statistics of China (1996–2011).

Figure 2.2

infrastructure and urban development: evidence from chinese cities

25

Figure 2.3 Urban Infrastructure Investment: Share of Total Infrastructure Investment by Region, 1990–2001 40 35 30

Percentage

25 20 15 10 5 0 1990

1991

1992

1993

1994

Urban investment in eastern region Urban investment in central region Urban investment in western region

1995

1996

1997

1998

1999

2000

2001

Urban investment by government in eastern region Urban investment by government in central region Urban investment by government in western region

Sources: Ministry of Housing and Urban-Rural Development of China (2000–2011); National Bureau of Statistics of China (1996–2011).

continuously for years: the ratio of urban infrastructure investment both to GDP and to investment in fixed assets hasFigure dropped, 2.3 falling below the levels of the central and western cities (figure 2.2). This can be attributed to a comparatively Lincoln_Ingram_Infrastructure high GDP and high amounts of fixed capital, which hold the ratio investment at a relatively low level. The ratio of governmental infrastructure investment to GDP and to investment in fixed assets, however, remains relatively high (figure 2.3), consistently surpassing levels in the central and western cities. Second, from 1997 to 2002, western cities had the highest ratio of investment in urban infrastructure to GDP. Driven by the national Develop the West strategy, the ratio in western cities increased rapidly after a long-term downturn and surpassed the ratio of the eastern and central cities. In addition to exhibiting the lowest ratio of investment in urban infrastructure to GDP, cities in the central region had the lowest infrastructure investment per capita (National Bureau of Statistics of China 1996–2011). Through the mid-1990s, the ratio in the central cities increased moderately but remained below that of the east.

26

Yan Song

A different classification of regions sheds light on how different types of infrastructure investments vary across China. Figure 2.4 illustrates the distribution of several types of infrastructure investments in the east, central, west, and northeast regions. Infrastructure investments in interior areas have grown significantly in recent years following several policy regimes, including Develop the West, Revitalize Old Industrial Bases in the Northeast, and Boost the Equalization of Public Services. From 2003 to 2010, on average, 40.5 percent of China’s railway investment was channeled to the west. Over the same period, road investment in the west comprised 32.7 percent of the national total. This heavy investment in the western region’s transportation infrastructure has decreased businesses’ transportation costs and has promoted the redistribution of industry from coastal areas to the interior. The investment in public services is highest in the east, followed by the western, central, and northeastern regions. The east’s higher financial capacity as compared to other regions accounts for its larger share of investment in public facilities, such as drainage systems within cities. In general, infrastructure investment is shifting from more developed cities to developing cities. Figure 2.5 compares the distribution of infrastructure in-

Figure 2.4 Distribution of Infrastructure Investments by Type and Region, 2003–2010 50 45 40

Percentage

35 30 25 20 15 10 5 0

Railroad

Roadway East

Central

Water West

Public facility Northeast

Source: State Council Development Research Center Information Network (1980–2010).

Figure 2.4 Lincoln_Ingram_Infrastructure

infrastructure and urban development: evidence from chinese cities

27

Figure 2.5 Infrastructure Investment Shift by Region, 1990–2010 70 60

Percentage

50 40 30 20 10

10

09

20

08

20

07

20

06

20

05

20

04

20

03

20

02

20

01

20

00

Interior region

20

99

20

98

19

97

19

96

19

95

19

94

19

93

19

92

19

91

19

19

19

90

0

Eastern region

Source: National Bureau of Statistics of China (1996–2011).

vestment in the eastern and inland regions as a whole between 1990 and 2010. Figure 2.5 Infrastructure investment in the eastern region was greater until 2007, when it was surpassed by investment Lincoln_Ingram_Infrastructure in the interior. TRANSPORTATION INFRASTRUCTURE INVESTMENT

Since the late 1980s, China’s investment in major urban transportation infrastructure, including railroads, roadways, aviation, and public facilities, has increased significantly with rapid urbanization in the country (Economic Research Institute for ASEAN and East Asia 2007). Figure 2.6 demonstrates investment trends by type from 2003 to 2011. During this period, total investment in major transportation infrastructure accounted for 13.2 to 17 percent of total urban investment, or 15.7 percent on average. Contributing to the rise in government spending in 2009 was an increase in investment to address the financial crisis, resulting in a small peak in investment in railways and public facilities. More specifically, railway mileage increased from 51,700 kilometers to 78,000 kilometers, a total increase of 50 percent and an average annual increase of 1.4 percent from 1978 to 2007 (figure 2.7). During this period, road mileage tripled, and civil aviation mileage increased 15-fold. According to the Interim and Long Term Railway Network Plan adopted by the State Council in 2004, the target is to add 120,000 kilometers of railway to the nation’s existing 91,000kilometer network by 2020, with an investment of 700 billion RMB (approximately

28

Yan Song

Figure 2.6 Ratios of Investment in Major Urban Transport Sectors to Total Urban Infrastructure Investments, 2003–2011 10

Percentage

8 6 4 2 0

2003

2004

2005

2006

Aviation

2007

Railroad

2008 Public facility

2009

2010

2011

Roadway

Source: National Bureau of Statistics of China (1996–2011).

Figure 2.7 Railway and Highway Mileages, 1988–2010

Figure 2.6 Lincoln_Ingram_Infrastructure

100,000

Mileage (km)

80,000 60,000 40,000 20,000 0 1988

1990

1992

1994

1996

1998 Railway

2000

2002

2004

2006

2008

2010

Highway

Source: National Bureau of Statistics of China (1996–2011).

US$113 billion) in the 12th Five-Year Plan period between 2011 and 2015. The objective set in the 12th Five-Year Plan is to construct an additional 120,000 kilometers of railway before 2015, 45,000 kilometers of which would be highspeed railway. China currently has 13,000 kilometers of high-speed railway. A high proportion of these planned infrastructure projects will connect cities in the

Figure 2.7 Lincoln_Ingram_Infrastructure

infrastructure and urban development: evidence from chinese cities

29

western region, such as from Xining to Lanzhou, or will connect cities in the west to those in the central or eastern regions. Intra-urban railway transportation infrastructure is also developing at an accelerated rate. In Shanghai, for example, the State Development Reform Commission has approved a plan to increase subway mileage to 850 kilometers by the end of 2015, which would far exceed the total subway mileage in New York or London. Spatially, these new intra-urban railways are concentrated; as shown in figure 2.8, 95 percent of urban railway transportation is concentrated in 11 provinces and cities in the east and is developing at an exceptionally rapid pace in the Yangtze River Delta. Figure 2.8 Distribution of Rapid Transit in China

Source: Freemark (2010). Reproduced with permission.

30

Yan Song

THE FINANCING MECHANISMS OF URBAN INFRASTRUCTURE IN CHINA

Funding for urban infrastructure in China comes from seven sources: central budgetary allocation, local budgetary allocation, domestic loans, bonds, foreign investment, self-financing, and other funds. The funding sources have become much more diversified since the beginning of China’s economic reform in the early 1980s. New financing channels for urban infrastructure construction emerged with the introduction of foreign capital in 1985, bond financing in 1996, and local budgetary allocation in 2001. Figure 2.9 shows the evolution of China’s different funding sources as a proportion of total urban infrastructure investment since 1980. Central budgetary allocation gradually decreased as a percentage of total investment, dipping below 10 percent in 1988 and remaining there, except for 1999 and 2000, when the government increased investment in infrastructure to cope with the impacts of Asian financial turmoil. In these two years, central budgetary allocation reached 11.96 percent and 12.75 percent of total funding, respectively. Nonetheless, the Figure 2.9 Changes in Funding Sources for Urban Infrastructure Investment, 1980–2010 80 70 60

Percentage

50 40 30 20 10

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

0

Central budgetary allocation Local budgetary allocation

Domestic loans Bond

Foreign investment Self-financing

Source: Statistics from State Council Development Research Center Information Network (1980–2010).

Figure 2.9

Other sources

infrastructure and urban development: evidence from chinese cities

31

central government has not been a major source of funding for urban infrastructure since the 1980s. Figure 2.9 also shows that urban infrastructure funding has gradually evolved. Self-financing and domestic loans are now the primary sources of funding, supplemented by a wide range of additional sources, including central government investment, foreign investment, and bond projects. The extent of private and foreign investment in infrastructure development has been very small, with the foreign direct investment (FDI) accounting for less than 2 percent (Sahoo, Dash, and Nataraj 2010). The proportion of the central budget in financing urban infrastructure has been decreasing rapidly in recent years, largely due to the implementation of a tax-sharing policy in 1994. At the same time, the increased availability of local financing through bank loans has enabled local governments to invest more in urban public facilities. Figure 2.10 shows that infrastructure investment as a share of central government spending has shown a downward trend, dropping to just 15 percent in 2006 from 34 percent in 1991. The same trend can be observed for state-owned investment as a share of total urban investment, which fell from 64.3 percent in 2003 to just 35.6 percent in 2011. As mentioned, local financing through bank loans has become increasingly available. According to a National Audit Office report (2011), there are approximately 6,500 platforms for local financing and 10.7 trillion yuan (US$1.72 trillion) in outstanding debt. Local financing platform debt is much higher than total local government revenue. Some local governments bear a heavy burden of obligation for debt repayment. By the end of 2010, 78 municipal governments Figure 2.10 Infrastructure Investment as a Share of Central Government Spending, 1991–2006 40 35

Percentage

30 25 20 15 10 5 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Source: National Bureau of Statistics of China (1996–2011).

Figure 2.10

32

Yan Song

(19.9 percent of all local governments) were bearing more than 100 percent of the debt ratio. Debt repayment by local financing platforms mainly relies on local finance and land revenues, along with purpose taxes, fees, and charges from public services and asset management. In recent years, local fiscal revenue has increased rapidly, with annual growth exceeding 20 percent, a strong guarantee for the repayment of the debt by local financing platforms. In addition, local government extra-budgetary revenue has increased substantially. Revenues from land sales (also known as land transfer fees) are the largest extra-budgetary sources of income. Figure 2.11 shows that land revenue accounts for about 80 percent of the extra-budgetary revenue of local governments, making it a major funding source of debt repayment for local government financing platforms. It is widely agreed that the local municipalities are relying too much on revenue from land sales for repayment. Revenue from land sales is not a sustainable source (Ding 2003). An alternative source of funding for local governments is levying taxes, such as property and real estate taxes. However, resistance to establishing such taxes is considerable (C. Zhang 2011). Another potential source of funding for local financing platforms is charging fees for the services that public infrastructure provides. This approach raises an important question about how to price public services in China without causing social unrest. China’s current pricing mechanism for rail, urban rail transit, and other public services has evolved from past practices of the central planning system. The lack of a more Figure 2.11 Local Government Income from Land Sales, 1999–2008 100

Percentage

80 60 40 20 0

1999

2000

2001

Revenue from land sales/ local budgetary revenue

2002

2003

2004

2005

Real estate–related revenue/ local budgetary revenue

Source: Ministry of Land and Resources of China (2000–2009).

Figure 2.11 Lincoln_Ingram_Infrastructure

2006

2007

2008

Revenue from land sales/ local extra-budgetary revenue

infrastructure and urban development: evidence from chinese cities

33

flexible pricing mechanism may increase the risk of local governments defaulting on their debts. For example, at present, Shanghai Metro fare box revenue barely covers interest payments; consequently, local government fiscal revenue is required for the repayment of principal (C. Zhang 2011). The development of other public service pricing mechanisms, such as land value capture, is important to ensure that local governments will be able to repay their loans.

Infrastructure, Urban Scale, and Urban Land Prices: A Cross-Sectional Analysis Infrastructure development in China has affected urban expansion and land prices in various ways. Many studies have examined the driving forces of urban land expansion across Chinese cities (He, Ke, and Song 2011; Ke, Song, and He 2009; Liu, Zhan, and Deng 2005; Song and Zenou 2006) and have found that China’s urban expansion exhibits a great number of spatial differences, resulting from different levels of demographic change, economic growth, and changes in land use policies and regulations (Liu, Zhan, and Deng 2005). However, few attempts have been made to investigate how infrastructure provision can affect urban expansion and urban land prices. Démurger (2001) provides empirical evidence demonstrating the links between infrastructure investment and economic growth in China using panel data from a sample of 24 Chinese provinces between 1985 and 1998. The results indicate that besides differences in terms of reforms and openness, geographical location and infrastructure endowment accounted significantly for observed differences in growth performance across space and that transportation facilities are a key differentiating factor in explaining the growth gap. This chapter examines the links between infrastructure investment and urban growth; specifically, this section offers empirical evidence suggesting how different levels of infrastructure provision contribute to variations in urban scale and land prices across Chinese cities. This chapter applies a consolidated monocentric model that was previously developed by He, Ke, and Song (2011). The model is developed to account for both “closed” and “open” city features in a developing country, where permanent urban residents and migrants interact in the informal goods market and the land market and yield a distinctive equilibrium pattern. The theoretical model (He, Ke, and Song 2011) is applied to an empirical analysis of urban scales and land prices across all Chinese cities for 2010. Table 2.2 describes the variables used in two regressions of urban scale and land prices. The dependent variables are urban scale and land price, respectively, in 2010. Of main interest, the analysis includes a set of variables accounting for infrastructure level in 2005: number of city buses per thousand residents (Bus), street length per resident (Street), number of express highways (ExpHwy), capacity of railroad center (RRCenter), capacity of airport (AirPort), and percentage of infrastructure investment compared to GDP (Infra_GDP). Together, these variables characterize

Table 2.2 Variables and Data Sources Variable

Description

Urban scale

Defined as Built _ distr / p where Built_distr is the area of the urban built district Defined as land sale values divided by areaa Number of city buses per thousand residents Street length per resident Number of express highways passing through the jurisdictional territory For cities at the prefecture level and above, RRCenter is 0 if total railway passenger throughput was zero in year 2000; non-zero throughput values are grouped into six quantiles, and numerical values 1–6 are assigned accordingly; for cities at the county level, RRCenter is assigned the value 1 if there is a railway station within 30 km from the urban center, and 0 otherwise For cities at the prefecture level and above, AirPort is 0 if total airline passenger throughput was zero in year 2001; non-zero throughput values are grouped into six quantiles, and numerical values 1–6 are assigned accordingly; for cities at the county level, AirPort is assigned the ordinal values 0–4 if number of daily flight departures is no greater than 0, 200, 300, 500, or greater than 500 from airports within 100 km from the urban center Percentage of infrastructure investment compared to GDP Population of permanent residents with nonrural residence permits in the jurisdictional territory Average annual salary income of permanent residents GDP_Ag if GDP_Ag ! 0; otherwise as Defined as Urban_area-Built_distr GDP Urban_area Defined as Price in the previous year Available land for transactions in the marketa Dummy for provincial capital cities or directly governed cities (30 in total) Dummy for prefecture-level cities (249 in total) Dummy for resource-extraction cities (16 in total) Dummy for cities in the central region, including the following provinces: Anhui, Heilongjiang, Henan, Hubei, Hunan, Inner Mongolia, Jiangxi, Jilin, and Shanxi

Land price Bus Street ExpHwy RRCenter

AirPort

Infra_GDP UrbPop Salary GDP_AgriLand

Price_lag Land_Supply Capital Prefecture Resource Central

Units km ¥10,000/hectare NA km/person NA

NA

NA NA 10,000 ¥1,000

¥10,000/km2 ¥10,000/hectare hectare NA NA NA

NA (continued)

34

infrastructure and urban development: evidence from chinese cities

35

Table 2.2 (continued) Variable

Description

Western

Dummy for cities in the western region, including the following provinces: Chongqing, Gansu, Guizhou, Ningxia, Qinghai, Shaanxi, Sichuan, Tibet, Xinjiang, and Yunnan

Units

NA

NA = not applicable. a For details on variable construction, see He, Ke, and Song (2011).

the capacity of intracity infrastructure levels and long-range transportation infrastructure in each city. It is hypothesized that a greater level of infrastructure provision and investment corresponds to better accessibility, hence larger cities and higher land prices. There are 660 officially designated cities in China. Excluding observations with missing data on key variables, 638 observations are retained for the urban scale equation. Data on land prices are available only for cities at the prefecture level or higher. Of these, 260 of the 280 have complete data and are used in the price equation. The data are extracted from the China City Statistical Yearbook, 2011; China Statistical Yearbooks for Urban Construction, 2011; and China Statistical Yearbooks for Land and Resources, 2010. The regressions include a set of control variables as suggested by previous studies (He, Ke, and Song 2011), such as population (UrbPop), average annual salary income of permanent residents (Salary), and average productivity of agricultural land (GDP_AgriLand ) as a proxy for the price of agricultural land (Brueckner 1990). In addition, a set of dummy variables are used to define city types, such as provincial capital cities (Capital), prefecture-level cities (Prefecture), resource-extraction cities (Resource), and cities in different regions (Central or Western). For land price estimation, lagged land price (Price_lag) and land supply (Land_Supply) are also included. These variables all have large standard deviations relative to the means and show great dispersion, suggesting significant disparities among Chinese cities in urban scale and land prices, and the attributing factors (table 2.3). Following He, Ke, and Song (2011), the log-log form is used to estimate the urban scale and land price equations. The results are shown in table 2.4. The estimates show that the provision of urban infrastructure affects both urban expansion and land prices. Specifically, in the urban scale regression, of the six variables that are used to account for infrastructure level, Street, ExpHwy, RRCenter, and Infra_GDP have significant and positive parameter estimates, while Bus and AirPort are not significant determinants of urban scale. This indicates that both intracity and intercity transportation investments, including streets within cities and intercity express highways and railways connecting cities, have a positive impact on urban growth. The general measure of

36

Yan Song

Table 2.3 Descriptive Statistics of Variables Used in the Land Price Function Variable Urban scale Land price Bus Street ExpHwy RRCenter AirPort Infra_GDP UrbPop Salary GDP_AgriLand Price_lag Land_Supply Capital Prefecture Resource Central Western

Descriptive Statistics Mean

S.d.

Min

Max

3.46 698 0.06 8.25E-04 1.42 1.79 0.84 2.69% 55.31 15.89 3.87E+04 527 149 0.046 0.383 0.025 0.372 0.194

2.03 540 0.06 4.11E-04 1.31 1.48 1.76 2.08% 110.34 5.77 2.99E+06 457 201 0.21 0.486 0.155 0.484 0.395

0.94 28.49 0.015 1.67E-04 0 0 0 1.97% 2.01 6.53 7.29 15.78 2.44 0 0 0 0 0

21.45 3868.35 0.26 3.41E-03 11 6 5 5.87% 1398.36 37.93 4.06E+06 3564.71 2048.9 1 1 1 1 1

Note: For most variables, the statistics are calculated for the scale equation sample, which includes 638 cities. Statistics for Land price, Price_lag, and Land_Supply are reported for the price equation sample, which includes only 260 cities.

infrastructure investment as a percentage of GDP is also positive and significant, indicating infrastructure’s important role in stimulating urban growth. The estimates show that a 10 percent increase in Street increases the urban radius by 1.39 percent, a 10 percent increase in ExpHwy increases the urban radius by 0.78 percent, a 10 percent increase in RRCenter increases the urban radius by 0.39 percent, and a 10 percent increase in Infra_GDP increases the urban radius by 0.05 percent. In the land price regression, of the six variables used to account for infrastructure level, ExpHwy, AirPort, and Infra_GDP have significant and positive parameter estimates, while Bus, Street, and RRCenter are not significant determinants of urban land prices. The insignificant estimate of Bus, Street, and RRCenter suggests that much of the convenience provided for intracity and intercity

infrastructure and urban development: evidence from chinese cities

37

Table 2.4 Regressive Results in the Land Price Function Variable (Intercept) Bus Street ExpHwy RRCenter AirPort Infra_GDP UrbPop Salary GDP_AgriLand Price_lag Land_Supply Capital Prefecture Resource Central Western Adj-R2 Wald test

Urban Scale

Land Price

Estimate

t value

Pr(>|t|)

Estimate

t value

Pr(>|t|)

0.867* 0.009 0.139*** 0.078*** 0.039*** 0.004 0.005** 0.313*** 0.174*** –0.021***

2.285 0.781 10.973 9.736 4.573 0.95 2.627 20.99 7.472 –5.909

0.02 0.31 0 0 0 0.32 0.01 0 0 0

0.135*** 0.164*** 0.045 0.051** 0.03 0.816 24.5**

3.695 5.98 1.335 2.672 1.11

0 0 0.18 0.01 0.22

2.750*** 0.008 0.065 0.058* 0.023 0.121** 0.004* 0.237* 0.681*** 0.028 0.389*** –0.116 0.264* 0.497*** –0.035 0.0379 0.058 0.606 20.3**

3.789 0.972 0.785 2.164 1.312 2.711 2.166 2.291 5.882 1.112 11.381 –1.333 2.289 3.762 –0.282 1.109 1.255

0 0.34 0.32 0.03 0.22 0.01 0.03 0.02 0 0.27 0 0.18 0.02 0 0.78 0.27 0.22

*p < .05 **p < .01 ***p < .001

commuters by streets, public transit, and railways has not been capitalized into land prices. The variables of RRCenter and AirPort perform differently in the urban scale and land price equations. The RRCenter variable is significant in explaining urban scale, while the AirPort variable explains more price variations. The different results can be explained by the differences between China’s rural-to-urban migrants and intercity migrants. Rural-to-urban migrants, contributing to urban expansion, are more likely to use the railways than the airlines. Thus, capacity of railway infrastructure is valued more in affecting urban scale. However, intercity migrants are usually better-paid professionals. In a few highly developed Chinese

38

Yan Song

cities, intercity migrants have dominated high-tech industry and financial business. Many white-collar migrants prefer air travel, and thus the capacity of the aviation infrastructure exerts a much greater influence on urban land prices (He, Ke, and Song 2011). Most of the other variables are consistent with expectations. In the urban scale regression, urban population plays the dominant role in determining urban spatial scale. The estimated elasticity of urban scale with respect to average income (Salary) is 0.17 and very significant. The estimates also provide evidence that government planning is important in affecting urban scale. Estimated coefficients for Capital and Prefecture are both significant and positive, indicating that holding everything else constant, a capital city or a prefecture-level city will be larger than an otherwise comparable county-level city because of the greater number of governmental functions contained in capital and prefecture-level cities. The estimated coefficient of the regional dummy Central is significant, indicating that cities in central China on average use slightly more land than those in the east. In the land price regression, results also show that land prices are very responsive to urban population and average income. Other things being equal, a 10 percent increase in average income induces a 6.8 percent increase in land prices. A 10 percent increase in nonagricultural urban population drives up land prices by 2.4 percent. Price_lag is significant, indicating that lagged land price induces an increase in the current price. Estimated coefficients for Capital and Prefecture are both significant and positive, indicating that holding everything else constant, a capital city or a prefecture-level city will have higher land prices than an otherwise comparable county-level city. In summary, by examining the determinants of urban scale and land prices across Chinese cities, it is evident that when controlling for other factors, greater urban infrastructure investment contributes to a higher level of growth of human settlements and higher land prices. Because the analysis included in this section is a cross-sectional analysis, the results suggest only that earlier investment in infrastructure is correlated with more expansive urban growth and higher land prices across cities. It is also possible that earlier decisions on infrastructure investments were made because of expected urban and economic growth in selected cities (World Bank 2004). In the next section, time-series data are used to explore how earlier infrastructure shapes later land development.

Infrastructure and Urban Land Conversions: A Time-Series Case Study Studies focusing on the links between urban infrastructure and urban spatial land development are needed. This section presents an analysis of land conversion using geographic information systems (GIS) and remote sensing data to explore the temporal and spatial characteristics of land use/cover change and urban land

infrastructure and urban development: evidence from chinese cities

39

development from 1994 to 2005 in Shenzhen City, Guangdong Province. Then a land conversion probability analysis is used to explore whether infrastructure exerts a role in inducing land development. Shenzhen was chosen for the case study on links between infrastructure and spatial development because it is a developed and yet dynamic city that provides sufficient changes in both infrastructure level and land development. Shenzhen is the first city in China to establish a Special Economic Zone (SEZ) to attract foreign technology. Since the 1980s, comprehensive plans have been drafted and implemented to provide urban infrastructure and to construct urban development (Bruton, Bruton, and Li 2005; Sun 1991). Figures 2.12 to 2.14 illustrate how land development occurs and how land use changes over time in three fast-growing areas in Shenzhen. As these figures show, transportation infrastructure projects are planned and implemented to attract future private investments in urban development (Sun 1991). This section employs Shenzhen as a case, uses remote sensing data at two time periods (1994 and 2005), and constructs a land conversion model to examine the impact of infrastructure on land use conversions. Previous studies have confirmed that the decision to change the existing use is influenced by economic, social, political, and personal considerations. There are many studies of land use conversion and the factors that influence the timing and location of this phenomenon (Carrion-Flores and Irwin 2004; Irwin and Geoghegan 2001; Liu, Wang, and Long 2008; Mertens and Lambin 1997; Veldkamp and Fresco 1996; Xiao et al. 2006). This section focuses on land use changes within an urban area over time and the impact of transportation infrastructure on land developments. The first step is to detect land use changes using remotely sensed land use/ cover data. Two scenes of Landsat images are collected for analyzing land use/ land cover change between 1994 and 2005. Both are Landsat 7ETM+image data, cloud free, and filtered with a 3×3 median kernel to exclude noise.1 To detect land use changes, a number of tasks must be performed: 1. Create and prepare a training dataset to support the satellite image classification. The classification system designed to categorize the land use properties of the study area included nine classes: urban/built-up, residential, crop field, vegetable field, forest/trees, orchard, grass, water body, and barren/sandy lands. The supervised classification method Maximum Likelihood was used to detect the land cover types.

1. Orbiting satellites capture reflected electromagnetic waves in bands (ranges of wavelengths) and vary in the number of bands of data they collect, the spatial resolution at which they capture data, and the spatial scale covered. Landsat is a commonly used data source for analyzing landscape change. It has global coverage and captures data in seven bands, at a resolution of 30 meters, in scenes that are approximately 180 km2. Landsat 5 and Landsat 7 detect blue, green, and red light in the visible spectrum as well as near-infrared, mid-infrared, and thermalinfrared radiation that human eyes cannot perceive. Landsat records this information digitally, and it is downlinked to ground stations, processed, and stored in a data archive.

40

Yan Song

Figure 2.12 Changes in the New Central Business District Area Over Time a. 1986

b. 1998

(continued)

infrastructure and urban development: evidence from chinese cities

41

Figure 2.12 (continued) c. 2008

Source: Image from Urban Planning, Land and Resources Commission of Shenzhen Municipality (2012).

2. Derive a signature file containing spectral characteristics of land cover classes of interest. 3. Perform a supervised classification of Landsat satellite imagery. 4. Identify urban areas within the study area at two time periods (1994 and 2005). 5. Detect and quantify the observed change in urban extent between 1994 and 2005. Figure 2.15 depicts the changes from nonurban to urban land use by each cell of 30 by 30 meters. The land use change analysis yielded a total of 8.2 percent of the cells that changed between 1994 and 2005. Figure 2.15 also shows the spatial distribution of changes, most of which occurred along the coastline and highway or arterial corridors. Following previous studies on identifying determinants of land use changes (Bockstael 1996; Wilson and Song 2011), a discrete choice probabilistic approach was used wherein the dependent variable is the probability of observing land use change from nonurban to urban use between 1994 and 2005. The land

42

Yan Song

Figure 2.13 Changes in the Huaqiao Area Over Time a. 1998

(continued)

use conversion model was estimated using all cells of the landscape that as of 1994 could be considered buildable in urbanized use. The dependent variable is a 0, 1 variable indicating whether the cell was actually converted between 1994 and 2005. As shown in table 2.5, the main set of predictors is a series of variables capturing infrastructure investments. For each cell, distances to the nearest existing roads, newly added roads, and subway corridor were calculated. In addition, road density in 1995 in buffers of different sizes was included to test whether these factors affect land changes between 1994 and 2005 with the aim to test the spatial extent of the hypothesized positive effect exerted by infrastructure density. To do so, a sensitivity analysis was implemented to consider how the parameters and fit of the statistical model respond to variation on the different distance thresholds used to derive the infrastructure effect measure. A distance threshold of two miles was chosen as the upper limit for the sensitivity analysis. As a point of reference, the mean distance from all cells to the nearest infrastructure was calculated to be 0.18 miles. This value formed the basis for the lower bound of the distance radii for the sensitivity analysis. Two more distance thresholds were also selected to partition these two endpoints and lend greater detail to the sensitivity analysis: one-half mile and one mile.

infrastructure and urban development: evidence from chinese cities

43

Figure 2.13 (continued) b. 2008

Source: Image from Urban Planning, Land and Resources Commission of Shenzhen Municipality (2012).

The second set of predictors was designed to capture market and neighborhood influences on land conversion. The average housing value per square foot at the beginning of the study period controls for disparities in real estate values in each cell. Distance to city center was included to account for access to aggregated economic activities by each cell. The third measure of land market conditions focuses on the supply of land and was operationalized as the proportion of total nonurbanized and buildable area in each cell at the beginning of the study period. Table 2.6 presents regression results. Most of the explanatory variables are highly significant and of the expected sign. For the infrastructure variables, cells that are closer to the nearest existing roads, the nearest newly added roads between 1995 and 2005, and the subway corridor are more likely to be developed.

44

Yan Song

Figure 2.14 Changes in the Yantian Area Over Time a. 1998

(continued)

Cells with denser roads nearby in 0.18 mile and one-half mile are more likely to be converted, with the one-half mile variable being the most significant among four sizes of buffers in the sensitivity analysis. The variable of road density becomes insignificant when it is measured at the one-mile or two-mile buffers. For the control variables, the higher housing values in 1995 increase the likelihood

infrastructure and urban development: evidence from chinese cities

45

Figure 2.14 (continued) b. 2008

Source: Image from Urban Planning, Land and Resources Commission of Shenzhen Municipality (2012).

that a cell will be converted, mainly because of the expected higher return for the real estate developers (Bockstael 1996). The availability of undeveloped land in 1995 increases the likelihood of the cell being urbanized by 2005. The distance to the city center is not significant, possibly because of the uniform distribution of economic activities across the city. In summary, through examining the determinants of urban land conversions, it is evident that when controlling for other factors, greater urban infrastructure investment increases the likelihood of land being converted for urbanized developments, indicating that access provided by streets and subway transit stimulates land conversions.

46

Source: Image from Urban Planning, Land and Resources Commission of Shenzhen Municipality (2012) and GIS calculations by the author.

Figure 2.15 Land Use Changes in Shenzhen, 1994–2005

infrastructure and urban development: evidence from chinese cities

47

Table 2.5 Independent Variables in the Conversion Model Variables

Hypothesized Effects

Data Source

Infrastructure Level Distance to nearest existing roads in 1995 (miles) Distance to nearest newly added roads between 1995 and 2005 (miles) Distance to subway corridor (miles) Infrastructure density measured as street length in 4 different sizes of buffers in 1995 (miles)

+ + + +

GIS calculations GIS calculations GIS calculations GIS calculations

+ + +

Planning bureau GIS calculations GIS calculations

Market and Neighborhood Character Average house value in 1995 (dollars) Distance to city center Proportion nonurban uses within quarter mile in 1995

Conclusions Since the economic reform began in China, infrastructure investment has increased to attract private investment, accommodate economic growth (Sahoo, Dash, and Nataraj 2010), and cope with economic crisis. This chapter provides empirical evidence that infrastructure has important effects on urban expansion rates, land prices, and spatial land development. Given this set of established links, it is especially important to examine whether recent infrastructure development could help cities grow toward a sustainable future. Despite the accelerated rate of infrastructure development, there are several challenges associated with the current practice of infrastructure development in China. First, the general level of infrastructure development in China is still low. Although infrastructure development has been advanced in developed cities in China, in many developing cities, the average infrastructure capacity per capita is comparatively low due to the large size of the population and the underdevelopment of infrastructure (Lin 2001). This insufficient level of infrastructure could impede efforts to accommodate both spatial and economic growth. Infrastructure must be improved, not only to facilitate economic growth (Sahoo, Dash, and Nataraj 2010), but also to overcome geographic barriers and increase western growth (Démurger et al. 2002). To address this issue, local municipalities, specifically the planning bureaus, need to design an infrastructure inventory system to accurately evaluate existing and predicted capacities. Such a system will

48

Yan Song

Table 2.6 Regression Results in the Land Conversion Function Variables

Estimates

Significance

0.1683 0.5781 0.2094 0.3855

*** *** * ***

1.0413 0.1588 –0.342

***

Infrastructure Distance to nearest existing roads in 1995 (miles) Distance to nearest newly added roads between 1995 and 2005 (miles) Distance to subway corridor (miles) Infrastructure density measured as street length in half mile in 1995 (miles) Market and Neighborhood Character Average house value in 1995 (dollars) Distance to city center Proportion nonurban uses within quarter mile in 1995

*

Model Summary Log-likelihood: –2834.23 Likelihood ration test (distributed Chi square): 3490.93 *p < .05 **p < .01 ***p < .001

enable infrastructure planners to avoid wasteful investment and to more effectively expand infrastructure development to accommodate urban and economic growth. Second, regional infrastructure development is imbalanced. Infrastructure in the eastern region of China is more developed (Loo 1999) than in the western and central regions despite the recent increase in infrastructure investment in the west. This regional imbalance is a barrier to the socioeconomic development of the hinterlands. In particular, many of these areas still have inadequate transportation infrastructure, as well as inadequate telecommunications, water supply, drainage, and electricity supply (Economic Research Institute for ASEAN and East Asia 2007; Li and Shum 2001). As regional equity is particularly important to maintaining social stability, measures need to be designed and implemented to lessen regional differences. Finally, financing sources for infrastructure provision are still limited in China. On the one hand, increasingly decentralized central-local fiscal relations are allowing municipalities a great degree of freedom for resource mobilization through a wide range of mechanisms that greatly expand extra-budgetary revenue (Wu 1999). In other words, China has succeeded in addressing urban infrastructure backlogs by opening up new avenues for financing. But on the other hand,

infrastructure and urban development: evidence from chinese cities

49

financing problems have emerged, prompted by debt-laden local governments in China in the aftermath of the global financial crisis (Tsui 2011). Several key institutions (the cadre evaluation system, the land management regime, and the banking sector) have created an environment that draws local governments into the trap of relying on unconventional sources, such as land transfer fees. The high levels of debt that have resulted may impede China’s efforts to mitigate structural imbalances in its economy. Cities of different administrative ranks have significant variation in financial capacities. The ingenious nature of extra-budgetary and off-budgetary resource collection by local authorities has resulted in high levels of intercity and intracity inequalities, further unbalancing the distribution of infrastructure (Démurger 2001; Wang et al. 2011). Efforts have been made to diversify financing sources. Much of the money raised through foreign investment and commercial loans is used for infrastructure construction in response to insufficient public financing mechanisms. However, the repayment terms for infrastructure loans are relatively long, and banks face the risk of incurring bad debts. In some regions, commercial bank loans account for 80 percent of the total investment in transportation (Economic Research Institute for ASEAN and East Asia 2007). However, experiences with other market-oriented financing tools and taxes (such as land value capture or property taxes) are still very limited and need to be expanded. The evidence on the link between infrastructure and land prices suggests that a more efficient land value capture tool can be designed to finance public infrastructure projects. Using infrastructure projects to guide sustainable spatial development is challenging in most cities. The priorities set by many cities on infrastructure investment focus on promoting economic growth and attracting private investment. Nevertheless, evidence shows that infrastructure does have an impact on urban scale and urban development patterns. However, when infrastructure development neglects other goals, such as efficient urban form and sustainable communities, an unsustainable form of urban land development could result. Two examples illustrate this issue. t

Many transportation infrastructure projects allocate land uses according to arbitrarily planned geometries such as axes, cores, and circles. Legacies from past central planning schemes have granted more power to the governments to determine where to locate infrastructure in China. City image projects exemplify institutional interference in the process of city growth. The layout design for Zhengzhou’s new central business district (CBD) is an example of emphasizing city image and neglecting principles of sustainable design and planning. Figure 2.16 shows that in the center of the newly constructed CBD is a circular highway system, along which an international convention center, a culture and arts center, and office and residential buildings are sparsely located. This layout of infrastructure falls short in promoting walkability, accessibility, and dense developments endorsed by smart growth principles.

50

Yan Song

Figure 2.16 Zhengzhou Eastern New CBD

t

The current infrastructure planning system does not ensure effective mechanisms for guiding the timing, location, and intensity of urban developments. For example, Huilongguan, a bedroom community in Beijing, was planned and constructed outside the fifth ring road and about 15 kilometers from downtown Beijing. Huilongguan is characterized by its enormous residential capacity, housing about 300,000 residents. However, there is no infrastructure concurrency requirement in terms of transportation infrastructure connecting the community to the city core. With a rapid increase in the number of private passenger cars, the current level of access roads and services is insufficient for 300,000 people. An additional concern with this suburban neighborhood is the lack of land use and transportation integration. More than 60 percent of residents in Huilongguan commute to downtown or other areas in Beijing for work. The community generates more commuting and non-commuting trips (especially external trips during peak hours), which worsens the existing transportation system not only for the lower-occupancy passenger cars, but also for the higheroccupancy bus transit. Current land use design does not consider the provision of more efficient modes of public transportation.

infrastructure and urban development: evidence from chinese cities

51

The failure to incorporate sustainable principles in infrastructure development has caught the attention of China’s planners and policy makers. The Ministry of Housing and Urban and Rural Development is calling for more efficient and green infrastructure development in the next era of urban growth. Existing infrastructure systems and land development in many local cities have been evaluated to identify unsustainable planning practices. Advanced planning techniques are being explored and developed to improve and transform current infrastructure provision practices into a more integrative, sustainable, and inclusive public policy-making process.

references Bockstael, N. E. 1996. Modeling economics and ecology: The implications of a spatial perspective. American Journal of Agricultural Economics 79(5):1168–1180. Brueckner, J. K. 1990. Analyzing third world urbanization: A model with empirical evidence. Economic Development and Cultural Change 38(3):587–610. Bruton, M. J., S. G. Bruton, and Y. Li. 2005. Shenzhen: Coping with uncertainties in planning. Habitat International 29(2):227–243. Calderon, C. A., and L. Serven. 2004. The effects of infrastructure development on growth and income distribution. Policy Research Working Paper No. 3400. Washington, DC: World Bank. http://ssrn.com/abstract=625277. Carrion-Flores, C., and E. G. Irwin. 2004. Determinants of residential land-use conversion and sprawl at the rural-urban fringe. American Journal of Agricultural Economics 86:889–904. Démurger, S. 2001. Infrastructure development and economic growth: An explanation for regional disparities in China? Journal of Comparative Economics 29(1): 95–117. Démurger, S., J. D. Sachs, W. T. Woo, S. Bao, G. Chang, and A. Mellinger. 2002. Geography, economic policy, and regional development in China. Asian Economic Papers 1(1):146–197. Ding, C. 2003. Land policy reform in China: Assessment and prospects. Land Use Policy 20(2):109–120. Economic Research Institute for ASEAN and East Asia. 2007. Infrastructure development in China. www.eria.org/googlesearch.html?q=infrastructure+in+china&ie= UTF-8&oe=UTF-8&hl=ja&btnG.x=0&btnG.y=0. Freemark, Y. 2010. China expands its investment in rapid transit, paving way for future urban growth. www.thetransportpolitic.com/2010/05/13/china-expands-its -investment-in-rapid-transit-paving-way-for-future-urban-growth. Gramlich, E. M. 1994. Infrastructure investment: A review essay. Journal of Economic Literature 32(3):1176–1196. He, M., S. Ke, and Y. Song. 2011. A consolidated model of monocentric cities and an empirical analysis of urban scales and land prices in Chinese cities. Journal of Regional Science 2:371–398. Irwin, E. G., and J. Geoghegan. 2001. Theory, data, methods: Developing spatially explicit economic models of land use change. Agriculture, Ecosystems and Environment 85(1):7–24.

52

Yan Song

Ke, S., Y. Song, and M. He. 2009. Determinants of urban spatial scale: Chinese cities in transition. Urban Studies 46:2795–2813. Li, S.-M., and Y.-M. Shum. 2001. Impacts of the National Trunk Highway System on accessibility in China. Journal of Transport Geography 9(1):39–48. Lin, S. 2001. Public infrastructure development in China. Comparative Economic Studies 43(2):83–109. Liu, J., J. Zhan, and X. Deng. 2005. Spatio-temporal patterns and driving forces of urban land expansion in China during the economic reform era. AMBIO: A Journal of the Human Environment 34(6):450–455. Liu, Y., L. Wang, and H. Long. 2008. Spatio-temporal analysis of land-use conversion in the eastern coastal China during 1996–2005. Journal of Geographical Sciences 18(3):274–282. Liu, Z. 2010. Planning and policy coordination in China’s infrastructure development: A background paper for the EAP infrastructure flagship study. http://siteresources .worldbank.org/INTEAPINFRASTRUCT/Resources/855084-1137106254308 /China.pdf. Loo, B. P. Y. 1999. Development of a regional transport infrastructure: Some lessons from the Zhujiang Delta, Guangdong, China. Journal of Transport Geography 7(1):43–63. Mertens, B., and E. F. Lambin. 1997. Spatial modelling of deforestation in southern Cameroon: Spatial disaggregation of diverse deforestation processes. Applied Geography 17(2):143–162. Ministry of Housing and Urban-Rural Development of China. 2000–2011. China urban construction statistical yearbook. Beijing: China Statistics Press. Ministry of Land and Resources of China. 2000–2009. China land and resources statistical yearbook. Beijing: China Land Press. National Audit Office of China. 2011. Report of local government debt audit result. www.audit.gov.cn/n1992130/n1992150/n1992500/2752208.html. National Bureau of Statistics of China. 1996–2011. China statistics yearbook, 1996– 2011. Beijing: China Statistics Press. Sahoo, P., R. K. Dash, and G. Nataraj. 2010. Infrastructure development and economic growth in China. Institute of Developing Economies (IDE) Discussion Paper No. 261. www.ide.go.jp/English/Publish/Download/Dp/pdf/261.pdf. Song, Y., and Y. Zenou. 2006. Property tax and urban sprawl: Theory and implications for US cities. Journal of Urban Economics 60(3):519–534. State Council Development Research Center Information Network. 1980–2010. Research database. www.drcnet.com.cn/www/integrated. Sun, H. 1991. Urban development in Shenzhen SEZ. Habitat International 15(3): 25–31. Tsui, K. 2011. China’s infrastructure investment boom and local debt crisis. Eurasian Geography and Economics 52(5):686–711. Urban Planning, Land and Resources Commission of Shenzhen Municipality. 2012. Aerial photo database. Veldkamp, A., and L. O. Fresco. 1996. CLUE: A conceptual model to study the conversion of land use and its effects. Ecological Modelling 85(2):253–270. Wang, D., L. Zhang, Z. Zhang, and S. X. Zhao. 2011. Urban infrastructure financing in reform-era China. Urban Studies 48:2975–2998.

infrastructure and urban development: evidence from chinese cities

53

Wilson, B., and Y. Song. 2011. Do large residential subdivisions induce further development? Journal of the American Planning Association 77(1):5–22. World Bank. 2004. Infrastructure for development. In World Development Indicators. New York: Oxford University Press. Wu, W. 1999. Reforming China’s institutional environment for urban infrastructure provision. Urban Studies 36:2263–2282. Xiao, J., Y. Shen, J. Ge, R. Tateishi, C. Tang, Y. Liang, and Z. Huang. 2006. Evaluating urban expansion and land use change in Shijiazhuang, China, by using GIS and remote sensing. Landscape and Urban Planning 75(1):69–80. Zhang, C. 2011. The scale, structure and efficiency of the infrastructure investment. China Economic Observer 28:2–10. Zhang, J. 2011. China’s infrastructure investment: Current situation and evaluation. China Economic Observer 28:11–26.

commentary David M. Levinson History doesn’t repeat itself, but it rhymes. —attributed to mark twain The growth of infrastructure in China is the investment story of the early twentyfirst century. Yan Song’s chapter documents much of what is happening. The plot underlying this story has played out previously in other developing countries, including the United States in the nineteenth and twentieth centuries and the United Kingdom in the eighteenth and nineteenth centuries. Rapid growth occurred in the railroads during the nineteenth century, following a familiar life-cycle pattern, as illustrated in figure C2.1. One of the key features of many life-cycle processes is overshoot. Shortly after peaking in 1920, U.S. railway mileage began a long inexorable decline, a process to date repeated with all technologies after they mature and when some better technology comes along. China is in the midst of riding what we call the Magic Bullet (Garrison and Levinson 2006). The Magic Bullet (figure C2.2) describes the feedback between economies of scale, service quality, demand, and cost that drives the growth of systems.

Figure C2.1 Life Cycle of U.S. Railroads 450,000 400,000

Maturity

350,000

Route (km)

300,000 Growth

250,000 200,000 150,000 100,000 50,000

Birth 0 1820 1840

1860

1880

1900

1920

Source: Garrison and Levinson (2006).

54

Figure 2C-1

1940

1960

1980

2000

2020

commentary

55

Figure C2.2 The Magic Bullet

Quality +

+ Economies of scale

+

Demand

–

– Cost Source: Garrison and Levinson (2006).

Economies of scale, the property that average cost decreases as throughput (satisfied demand) increases, are found in systems like railways in their growth stage. While some of the economies may be kept as profits, in general, during the growth phase, the economies are reinvested and returned to users as either price Figure 2C.2 reductions or service quality improvements, as investors seek future profits. On a Lincoln_Ingram_Infrastructure passenger rail link, for example, the greater the traffic, the less the cost of movement (due to more frequent services and thus less schedule delay) and the better the service, at least until congestion sets in. While the early railroads in the nineteenth century had to discover this process, China is in a position of not having to invent the railroad, but instead can intelligently emulate it, deploying a well-understood technology across an underdeveloped landscape. This spatial diffusion process should be expected to follow the traditional S-shaped life-cycle curve, as the best links are built first, and links continue to be added as long as the benefits outweigh the costs. By developing later, China has the advantage of being able to deploy better technologies (e.g., high-speed rail), which the United States missed in its first round of deployment and is only now thinking about building. Interestingly, the last decade has seen a lower share of self-financing and more money coming from government budgets and borrowing than in previous years. Self-financing was used primarily for the U.S. interstate highway system (via the motor fuel tax), but borrowing was de rigueur for railways, which at first didn’t have enough revenue to pay for themselves. Later, some investors were

56

Yan Song

paid back (though many others were not, as most U.S. railways went through bankruptcy at one point or another, wiping out investors). The deployment of infrastructure mirrors and reinforces the growth of Chinese cities. Rapid urbanization, enabled by economic expansion and the differential rewards for urban living, is resulting in the transformation of cities and the nearby countryside into modern developments. Clearly, there is some concern about spatial equity in China, as the chapter reports significant investments in rural areas despite the greater growth rates in urban areas. Song documents the fascinating explosion of Chinese cities. We have seen rapid urbanization before. As places in the United States became connected to the national and global system of cities and new areas could be developed, growth was profound. Figures C2.3 to C2.6 show the transformation of growth in Minneapolis, Minnesota, from 1865 to 1891. The Minneapolis-St. Paul metropolitan area peaked at ninth largest in the United States in the 1890 census. (The city of Minneapolis was the 15th largest in the United States at its 1930 peak.) The scale of course differs in China, with Shanghai (at 13.5 million in 2009) much larger than greater Minneapolis (at 305,000 in 1890 and 3.2 million today). Shanghai is building a subway network to serve its core, like large cities before it. Although this chapter corroborates that infrastructure drives development, it is not clear from this analysis whether development also leads infrastructure, though one suspects it is true. Minneapolis and St. Paul saw land growth driven by streetcars in the late nineteenth century (Xie and Levinson 2010). In the case of the Twin Cities, streetcars led land development, but elsewhere, like London, there was mutual causation (Levinson 2008), and in New York, the subway tended to chase population (King 2011). The life-cycle discussion is central in any international comparisons. The United States, Japan, and European nations are mature and well developed, and so they do not demand the same level of investment as fast-growing countries like China, which have proportionally less infrastructure. That China is investing rapidly, and presumably sees returns, does not imply that the United States or European nations should do likewise. China would do well to heed the experiences of the nations that went before and learn from them.

references Garrison, W. L., and D. M. Levinson. 2006. The transportation experience: Policy, planning, and deployment. New York: Oxford University Press. King, D. 2011. Developing densely: Estimating the effect of subway growth on New York City land uses. Journal of Transport and Land Use 4(2):19–32. Levinson, D. 2008. Density and dispersion: The co-development of land use and rail in London. Journal of Economic Geography 8(1):55–77. Xie, F., and D. Levinson. 2010. How streetcars shaped suburbanization: A Granger causality analysis of land use and transit in the Twin Cities. Journal of Economic Geography 10(3):453–470.

57

Source: Library of Congress Geography and Map Division, Washington, DC 20540-4650 dcu, http://hdl.loc.gov/loc.gmd/g4144m.pm003930.

Figure C2.3 Bird’s-Eye View of Minneapolis, 1865

58

Source: Library of Congress Geography and Map Division, Washington, DC 20540-4650 dcu, http://hdl.loc.gov/loc.gmd/g4144m.pm003950.

Figure C2.4 Bird’s-Eye View of Minneapolis, 1879

59

Source: Library of Congress Geography and Map Division, Washington, DC 20540-4650 dcu, www.loc.gov/item/75694644.

Figure C2.5 Bird’s-Eye View of Minneapolis, 1885

60

Source: Library of Congress Geography and Map Division, Washington, DC 20540-4650 dcu, http://hdl.loc.gov/loc.gmd/g4144m.pm003970.

Figure C2.6 Bird’s-Eye View of Minneapolis, 1891

3 Mobile Telephony and Socioeconomic Dynamics in Africa Mirjam de Bruijn The mobile phone revolution in Africa is a fact. —ngo employee1

I

t cannot be denied that mobile telephony has been booming in Africa since 2000. Although there are still regions that are not connected to this type of wireless technology, they are rapidly decreasing in number, and not having direct access to the network does not mean that people do not use the phone. What has this revolution done for Africans and for African societies? Recent studies and articles are quite euphoric about the effect that the mobile phone has had on social relations, economic possibilities, and the political engagement of the “voiceless,” to the point that they have even introduced a new catchphrase in development aid: “mobile interventions.” How well substantiated are these claims?2 Telecommunication infrastructure and development are supposed to be positively related. Roller and Waverman (2001) have analyzed this general relationship in terms of the industry itself and the effect of telecommunication services. Development is thus related to economic growth. In such equations, the concept of development stands for progress, as in moving toward a modern society. The

1. This comment was made by an NGO employee at a meeting about mobile interventions in Africa, held at the African Studies Centre, University of Leiden, April 4, 2012. “Mobile Telephony, Democracy and Development in Africa: Policy and Research Approaches” was organized by the research project Mobile Africa Revisited. 2. In recent publications, this “revolutionary” change as a consequence of mobile telephony is questioned. See Ekyne (2010); Etzo and Collender (2010); and cf. de Bruijn, Nyamnjoh, and Brinkman (2009).

61

62

Mirjam de Bruijn

questions are, Who benefits and when? Is this in fact a positive development? And who judges what is positive? The relationship between infrastructure and development is certainly not linear. It may be better to relate infrastructure to change—social, economic, and political change. The relationship could then be analyzed to determine who experiences the change and who will benefit or lose out. This chapter asks these questions regarding ordinary people in various regions in Africa and focuses in particular on those areas that are less well endowed with infrastructure in general, as it might be expected that this would be where telecommunication infrastructure could have a significant impact and make a difference. Telecommunication infrastructure is seen here as being primarily mobile telephone technology. Changes in society are examined from an economic, political, and social angle, and the chapter is based on research carried out as a part of the Mobile Africa Revisited research program, which is currently investigating the social dynamics related to mobile telephony in Africa.3 The research methodology is basically qualitative and does not pretend to decipher the revolution statistically, although it does attempt to understand changes in dynamics and processes as a consequence of this new era of communication and information technology. The research is based on the notion that present-day dynamics can be understood in relation to past dynamics and that today’s dynamics are always related to specific contexts, which makes it difficult to generalize from these specific data. The chapter also compares different regions in Africa to highlight some commonalities and differences that in turn allow us to draw some general conclusions. This research program started in 2006, which thus introduces a certain temporal depth to the observations. Comparison with the period before mobile telephony also helps us understand social processes. The analysis here departs from the interrelationship between society and technology as it was developed in the Actor Network Theory (Latour 2005). Society shapes technology as technology shapes society. The mobile phone is, therefore, as much a part of society as society is a part of the mobile phone (cf. de Bruijn, Nyamnjoh, and Brinkman 2009). The changes described here are the result of this interaction. The chapter starts by reviewing the statistics on mobile telephony. These general data are related to stories of the construction of the infrastructure as it developed in the project’s different field sites. The question “What has the mobile telephone done for people in Africa?” is examined in terms of the economic infrastructure of communication. The chapter focuses on craftsmanship; social relating and distance; trust and betrayal; and phone cultures. It will become clear that the mobile phone revolution is developing in diverse directions and is widely dependent on the context in which it has evolved.

3. The program is sponsored by WOTRO, The Netherlands (W 01.67.2007.014). For more information on the Mobile Africa Revisited program, see www.mobileafricarevisited .wordpress.com.

mobile telephony and socioeconomic dynamics in africa

63

The Mobile Phone Revolution in Africa: Has the Digital Gap Been Closed? International Telecommunication Union (ITU) statistics show that Africa is being increasingly covered by mobile telephony. The development of mobile telephony expanded considerably at the beginning of the twenty-first century (cf. Castells et al. 2007), and coverage today shows a mean of 50 percent. This is an enormous boom for Africa, but the continent still seems to be lagging behind the developed world and also the developing world, as shown in figures 3.1 and 3.2. However, these global statistics hide a lot of dynamics. Within Africa, the differences between countries are huge, but growth has been seen everywhere. Countries like South Africa, Gabon, and Egypt have coverage of more than 100 percent, while others, such as Nigeria and Cameroon, are said to have 40 to 50 percent coverage. Yet others, like Mozambique, Chad, and the Central African Republic, do not exceed 30 percent. Within these countries, some regions are doing better than others, and some people are better integrated in this form of communication than others. Urban areas are generally much better connected than rural areas, and a difference in connectivity between social categories is inevitable within these areas. These differences relate to the socioeconomic and political contexts in which mobile telephony develops, and the way it is embedded in society affects the Figure 3.1 Estimated Mobile-Cellular Subscriptions Worldwide, 2001–2011

Number of subscriptions (per 100 inhabitants)

140 120

Developed World Developing

100 80 60 40 20 0 2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

Note: The developed/developing country classifications are based on the UN M49; see www.itu.int/ITU-D/ict/definitions/regions/index.html. Source: ITU World Telecommunication/ICT Indicators database.

64

Mirjam de Bruijn

Number of subscriptions (per 100 inhabitants)

Figure 3.2 Estimated Mobile-Cellular Subscriptions in Select Regions, 2011 160 140

143.0 119.5

120

103.3

100

96.7

86.7

80

73.9

60

53.0

40 20 0

Commonwealth of Independent States

Europe

The Americas

Arab states

World

Asia and Pacific

Africa

Note: Regions are based on the ITU BDT Regions; see www.itu.int/ITU-D/ict/definitions/regions/index.html. Source: ITU World Telecommunication/ICT Indicators database.

FigureThe 3.2 appropriation of technology is a way people appropriate the technology. Lincoln_Ingram_Infrastructure two-way process in which technology and society influence each other. Closing the gap will be a gradual process in which both state policies and the policies of companies and local users will play a crucial role.

Technology and Infrastructure: Changing the Landscape Wireless technology consists of magnetic fields produced by mobile phone masts that have a certain reach and send out signals for the phone user (figure 3.3). To have a communications network in an urban space, a mast is needed every few hundred meters. In a rural area, masts can be set farther apart, but providers need to negotiate access to the land on which the masts are erected, by either leasing the land or buying it. The masts are often found next to houses or on the top of multistory buildings. They are highly visible in the urban landscape and are found on the tops of mountains in the rural areas or on flat areas where their reach can be much greater. Mobile phone masts have become recognizable markers on the African landscape. Many anecdotes used to surround the way people in rural areas in Africa tried to access a signal for their phones, with people in remote areas climbing trees or walking long distances to find a signal. In some places known to have a signal, a phone would hang on a cord or in a tree so people could receive calls. These spots of connectivity became social places, and the entrepreneur who was

mobile telephony and socioeconomic dynamics in africa

65

Figure 3.3 Mobile Phone Mast in Nomadic Lands, Mali, 2005

Photo: Mirjam de Bruijn.

able to appropriate them became the village messenger (de Bruijn, Nyamnjoh, and Angwafo 2010; Seli 2012; cf. van Beek 2009). With advances in the network and the growing numbers of masts in remote areas of Africa today, these social spots are disappearing.4 Mobile telephony is a technology that does not work without electricity. To be able to set up masts, the phone companies have had to bring in their own power generators, as rural areas often had no prior electrical supply. The generator itself then provided the sole source of electricity for people to recharge their phones. Mobile Telephone Network (MTN)5 has not, however, been allowed to provide electricity in villages in Cameroon, which has to be done exclusively by 4. For a recent account of these practices in Burkina Faso, see www.metropolistv.nl/nl/themas /de-telefoon/bellen-in-de-boom-in-burkina; this short film reminds us not to draw conclusions about disappearance too rapidly. After all, the whole world will never be connected to wireless; there will always be regions without connections. The film reminds us of the notion of the “fifth world” as the world where connectivity is absent (cf. Castells et al. 2007). 5. MTN is a South African–based mobile telephone company headquartered in Johannesburg.

66

Mirjam de Bruijn

the state company, Société Nationale d’Electricité du Cameroun (SONEL). This restriction has led to innovative solutions by villagers who created new businesses for charging phones with the help of car batteries or even radio batteries. In a few instances, people have even resorted to using bicycles to generate power. The recharging of phones often takes place at shops that have a generator running during the day, as this provides a (small) extra income for the shop owner. Increasingly, generators have also become a part of Africa’s rural landscape. The problems surrounding electricity provision by state companies are enormous, and the companies often default due to corruption and surplus demand. The growth of cities and the increasing demand for electricity for today’s modern lifestyle are the obvious causes.6 AN INTERMEZZO

In 2005, after an absence of a few years, I visited central Mali and came across mobile telephony for the first time in a remote area. I had worked with nomadic people in central Mali between 1987 and 2005 for Ph.D., postdoctoral, and other research projects, and until the beginning of the twenty-first century there had been no electricity, only one tarmac road (built in 1985), and hardly any roads at all into the interior where the nomads lived. In short, travel was difficult. The main means of communication then was by camel or on foot. Land lines were only available at state offices and were not accessible to ordinary people. Fast forward to 2005: The first wireless masts had been established near the area’s small towns, although they still did not have regular power supplies everywhere. I was surprised to meet my nomad “brother” in his camp with a phone. The flat landscape allowed the signal of the mast 35 kilometers (22 miles) to the north to reach the camp, but only in one place. My brother had managed to connect to the elected deputies in the region, and because of this, he found support among his people to become their leader. He has now entered politics, and as he himself admits, this was due to the contacts he was able to maintain through mobile communication (Sangaré 2010). Early mobile telephony in Africa has created many such stories, and the discovery of connection spots has forever changed the communication landscape for many. These spots become hubs of communication and illustrate local enthusiasm for phoning. Individual benefits, as in the extreme case of my nomad brother, enhance expectations. And thus, a change in lifestyle is unavoidable. The lives of those who have accessed mobile telephony are changed forever.

6. The electrification of sub-Saharan Africa is about 30.55 percent, compared with 99 percent in North Africa. There are huge differences within sub-Saharan Africa: 14 percent of the rural areas are covered, compared with about 40 percent of the urban areas. See www.iea.org (World Energy Outlook 2012).

mobile telephony and socioeconomic dynamics in africa

67

The Delivery of Mobile Phone Service Today, mobile telephony in Africa is provided by different companies that all entered the competitive market in 1998. Foreign companies competed with national companies, which has resulted in the encroachment of international phone companies on the continent. One of the first foreign companies to operate in Africa was Celtel, a Dutch company that has since started operations in 14 African countries. Their colorful advertisements were covered with slogans relating to development rhetoric. They were very visible in Chad’s capital, N’Djamena, when I visited in 2007. Their red and yellow advertisements literally colored the town, and their slogans shouted out at people traveling along the main roads (figure 3.4). In 2009, Celtel was sold to Zain, a company from the Middle East that uses a completely different set of colors—black, pink, and pale green—in their advertising campaigns. The message in the advertisements has changed, too. The

Figure 3.4 Celtel Advertisement in N’Djamena, Chad, 2007

Photo: Mirjam de Bruijn.

68

Mirjam de Bruijn

women on the billboards now wear veils, and the images presented are of a happy Muslim family, whereas Celtel campaigns had sold visions of a happy life more generally. MTN is a South African–based company that painted the landscape yellow and introduced words like “airtime” (on a yellow flyer) and a new greeting: “Y’ellow.” They advertise their mobile phone services on huge boards with slogans and images of a beautiful middle-class family life. Their most impressive advertisement to date talks of the “Right to Communicate” (figure 3.5). The French company Orange is also an influential player in the African telecom market, with its orange advertisements and similar campaigns. As part of its campaign, Orange distributes free paint in the company’s color for shops and houses to remind people of the company’s advertising message. Orange and the other mobile phone companies have effectively turned the urban African landscape into a competitive communication landscape. Mobile telephony not only provides possibilities for communication, but also brings new messages and new ideas for society—and, in the end, an inevitable change in people’s livelihoods.

Figure 3.5 MTN and the “Right to Communicate,” Bamenda, Cameroon, 2009

Photo: Mirjam de Bruijn.

mobile telephony and socioeconomic dynamics in africa

69

The Economic Infrastructure of Communication The mobile phone boom has affected different levels of the economy, changing the national and regional economies and certainly local economic dynamics as well. The focus here is on local dynamics and how the mobile phone companies have created a new economic space. The economy that is created in such a field fluctuates with the context, and what seems to be a very firm basis can collapse at any moment. The communication economy has its own specificities, and these are also discussed in this section. Waverman, Meschi, and Fuss (2005) have indicated the possible influence of mobile telephony on trade relations and commerce on a micro scale and on the economy brought into countries with the arrival of the new commodity. The communication economy produces services, goods, and gadgets with their own markets, and mobile phone companies have generally tried to control the market for their services as far as possible. AIRTIME

Most phone use in Africa is with prepaid systems. Subscriptions are a new concept and only accessible to a few relatively rich people. Most people therefore have a SIM card in their phone (often more than one so they can juggle use according to the different prices offered by the companies), and they put credit (“airtime”) on their phone as they need it or can afford it. The main service that the mobile phone companies provide is airtime, which is their biggest moneymaker. Sales of airtime have gone through different stages. The first system worked in a trapped hierarchy: the hierarchy of sale. It worked as follows: The company had a huge bulk of airtime to divide among its customers but could not handle the sales directly, especially in the parts of Africa where roads and transportation are difficult. Most companies thus worked with middlemen, who were often big businessmen who could afford to buy large quantities of airtime directly from the company. These middlemen would then divide up the airtime and sell it to middlemen lower down the chain, who were also relatively rich. This second group bought quite a lot of airtime, which they in turn sold to regional middlemen, who sold it to individuals, who sold it from tables, kiosks, phone boxes, or bikes directly to customers. To make the system attractive, companies introduced a bonus system (cf. Nkwi 2009). AYABA: A STUDENT IN BUEA, CAMEROON

When we first met Ayaba, a university student reading history, in Buea, Cameroon, in March 2012, she had just finished her exams. Her involvement with MTN started in her hometown, Bamenda. Her mother died when she was at secondary school, forcing her to take care of herself and her household. She was eager to continue her education, so she began selling airtime and built up a clientele in Bamenda, which helped pay her school fees and allowed her to contribute to her family’s income. When she left school and headed for the university in

70

Mirjam de Bruijn

Buea, she continued her business, although her new customers were very different from those she had in her hometown, something she felt was related to the university environment. Ayaba’s call box is opposite the main entrance to the university, where she is now well known. Students are not always easy, she says, as they sometimes try to avoid paying for services or create problems at her stand. Nevertheless, the business has allowed her to go to the university. Ayaba works hard, but her sister joined her for a few months to help when she had to study. SELLING AIRTIME IN N’DJAMENA

The young men pictured in figure 3.6 were on the streets of N’Djamena just in front of the Celtel office in 2007 trying to sell phone cards. At that time, the

Figure 3.6 Young Men Selling Airtime for Celtel in N’Djamena, 2007

Photo: Mirjam de Bruijn.

mobile telephony and socioeconomic dynamics in africa

71

Figure 3.7 Young Men Selling Airtime for Zain in N’Djamena, 2009

Photo: Mirjam de Bruijn.

transfer of airtime had not yet been introduced in the city, and it was common to buy a scratch card to put credit on one’s phone. These young men claimed that the business of selling airtime was profitable for them and that they were able to contribute to their family’s income. A few years later, after Celtel was taken over by Zain, young men had Zain bikes, which made them more mobile, increasing the area in which they could sell airtime (figure 3.7). A RECENT ECONOMIC APPLICATION: MOBILE MONEY

The technology of mobile telephony not only has opened a new market related to the phone, its accessories, and its service, but also has become a tool in facilitating economic transactions. In regions with rudimentary economic infrastructure, wireless technology opened possibilities for safe and quick economic transfers. M-banking (mobile banking or mobile money) is becoming increasingly popular in Africa. In regions where banking and the bank system have never been established, mobile technology seems to have found a new niche with a high potential for economic development (Batchelor, Kashorda, and Sylla 2009). The technology is based on the message functions in Global System for Mobile Communications (GSM) networks to which most of the mobile users

72

Mirjam de Bruijn

in Africa are connected. Users whose phone company collaborates with a local bank can create their own accounts, similar to a bank account, and transfer money by mobile messaging to whomever they want. This is a sophisticated adaptation of airtime use for mobile money transfer, a technique that arrived after mobile telephony and was immediately adopted by Africans. Airtime was sent from point A to B, and subsequently in B it was paid out in its monetary form. In such a way people would send small amounts of money from urban to rural areas, often from richer or working family members to poorer ones. Airtime was monetized, and the mobile phone became a safe tool for the transfer of small amounts of money. This first form of mobile money later turned into the more sophisticated system of mobile banking. The first experiment with mobile banking was developed by Safaricom, the first mobile phone operator in Kenya, where the concept M-Pesa (Swahili for mobile money) was introduced in 2007. It turned out to be very successful. Nevertheless, a survey on the use of the M-Pesa in Kenya showed that the system is mainly used by the relatively wealthy, usually in urban rather than rural areas (Jack and Suri 2011). This finding is confirmed by the extensive distribution of the technology in South Africa, which has one of the wealthiest economies of the continent (Batchelor, Kashorda, and Sylla 2009). It is therefore interesting that in May 2012, when we tried to find out if and how mobile banking worked in anglophone Cameroon, more specifically in Bamenda, people admitted that they did not trust the system. At that time, Orange and MTN had just started their campaigns to promote mobile banking. Most phone users said that they never used mobile banking but admitted that they did use the transfer of airtime to send small amounts of money to their parents or other family members in the village. Also important is the fact that in this part of Cameroon, a credit union had been introduced before independence in 1960. People use the services of the credit union to deposit money or to send money home when they travel (cf. Nkwi 2011), which reduced the need for mobile banking. Other regions where we conducted research—Chad, the Casamance region in Senegal, and southwest Angola—were also not yet particularly advanced in this new mobile technology. Nevertheless, the development potential of this new form of money transfer is high. As the case of Kenya has shown, users appreciate it for its safety and the directness of money transfers, for the fact that they no longer have to carry money along while traveling, and because it facilitates commercial activities (cf. Jack and Suri 2011). The newest trend related to M-banking and mobile money is the mobile payment of water and utility bills. Mobile service deliveries have only just begun in remote and underserved areas, but the unequal access to mobile banking, which was one of the major conclusions in the M-Pesa survey study, threatens a possible increase in social and economic inequalities (Jack and Suri 2011). However, the technology’s high rate of adoption and quick spread among ordinary African citizens also reveal an eagerness to appropriate this innovative use of mobile technology in daily life.

mobile telephony and socioeconomic dynamics in africa

73

Communication Craftsmanship Mobile phone maintenance is to link the line. —chinese proverb Mobile phone technology was not originally African but was developed in offices in the West. But technology is not static, and it adapts to the environment it finds itself in and molds itself according to the wishes of the host country’s customers, users, and leaders. The variety of call boxes that now include mobile banking services is an example of how mobile phone technology has been adapted. Other examples are how providers adapt and are developing technology for the “poor” market that they consider Africa to be, as discussed in the previous section. The craftsmanship around the mobile phone as an object has become increasingly adapted to the African communication environment, and this in turn has transformed the very communication landscape itself. I met Appolinaire in Buea in January 2012 when my colleagues and I did an extensive qualitative survey on phone repair shops for London’s Science Museum, which was preparing an exhibition on phone culture in Africa. The most remarkable figure in this research is the phone repairer. He (and it is always a “he”) has become a central figure for people and their phones, replacing Ericsson, Vodafone, or whatever big company one might be attached to in the West. The technological know-how of these men is phenomenal, and it is adapted to the local technological environment and economy. Appolinaire’s story is, in a way, representative of how young men’s lives are developing in Cameroon, Uganda, Mali, and elsewhere in Africa.7 His parents thought their children should go to the university, and they worked hard to give them this chance. Appolinaire earned a bachelor’s degree in accounting from Buea University and, after graduation, went to Limbe, where his parents live. He worked for a while for a Chinese firm and got to know a Ghanaian man who repaired phones. This caught his interest as he had always liked physics, and his knowledge in this field was now useful. In the end, he turned out to be a better repairer than his teacher, and he opened his own workshop. He worked in Limbe for a few years and then went to Buea, where he expected a better market. His shop traveled with him, as it is located inside a mobile kiosk. Inside he built a waiting room and his repair area, which was just wide enough for him to sit in, with room for his table and computer (figure 3.8). In Buea, Appolinaire now repairs the Chinese smartphones that are very common in Cameroon and often require software repairs. Older models need hardware repairs, and much of what looks like garbage in his work area is actually parts of old phones that he has saved over the years and that now make up 7. I have met young men with similar stories in Yumbe, Uganda; in Khartoum and Juba, Sudan; and in Bamako, Mali; and students have reported other cases in Chad and Senegal.

74

Mirjam de Bruijn

Figure 3.8 Appolinaire in His Shop, 2012

Photo: Mirjam de Bruijn.

his tool kit. He collaborates with a group in Buea and is good friends with one of the men whom he got to know through a Chinese phone repairer who was the first to open up such a business in the city. His shop is still there, but the Chinese man has moved on to another business. Appolinaire explains how his encounter with this Chinese phone repairer, who happened to rent space for his shop in his father’s house, introduced him to this business. He started repairing phones with the Chinese man, who hardly spoke any English but was able to teach him the craft. Today, he and Appolinaire teach others, and the craft of phone repair is seen as a type of apprenticeship. Appolinaire and his friends collaborate with repair groups on the Internet and are in almost daily contact with the GSM forum where (phone) repairers from all over the world explain their problems and help one another. The forum works on the basis of knowledge exchange, and if one has nothing to contribute, reception is difficult. A body of repair knowledge is thus being created. To access this knowledge, you have to add to the knowledge in a way that advances improved phone technology. A young man in Bamenda told me that for a long time he exchanged ideas with a friend in India every day. The Cameroonians will not claim ownership of this knowledge, but it will certainly end up improving

mobile telephony and socioeconomic dynamics in africa

75

technological applications somewhere in a Chinese, Korean, or European business. This structure raises interesting questions about the development of phones with two or three SIM cards. Having two or three SIM cards is only useful as long as the various phone companies do not collaborate and make phoning from one to the other more expensive. This is certainly the case in the African market, where competition is intense and people who live in poverty cannot afford to pay more than the bare minimum. Another development is the cheap smartphone that is depicted as being of secondary quality but still fills a clear role in the African economy. These forums appear to be research consortia for the big phone companies. Today, repairing phones is big business in African countries, which often receive secondhand phones from Europe that require repair or cheap phones that need frequent attention. This kind of craftsmanship has allowed many young men to earn a good income.

Social Relating and Distance: Mobile Margins Enlarged With easier communication, social relations over a distance are expected to be transformed (cf. Ling 2008). This has certainly happened in Western Europe and in the United States, where the mobile phone has enhanced regular contact between family members. The mobile phone connection has become a new umbilical cord. It has been suggested that the mobile phone has reinforced strong ties and loosened weaker ones. Following the early social network theory, this would not seem to enhance society in terms of extending networks. On the contrary, it may well reinforce boundaries between socially well-connected groups and thus create more social boundaries and potentially more conflict, too (cf. Granovetter 1973; Horst and Miller 2006; Ling 2008). It would seem obvious that social ties will change, but how and for what reason? Does this social change also influence people’s daily choices and thus change livelihoods and ideas about life? How does this interfere with the changes described in the economy and knowledge sector and people’s expectations? We should also note that voice communication, which so far has been the most important use of the mobile phone in Africa, is very different from communication via social media, which would appear to be on the rise in Africa’s remote areas. This section presents a case illustrating the social use of both forms of communication. Habsatu, a Cameroonian friend, lives in Bamenda, where she has a tailor’s workshop. Her family lives in the north and keeps cattle on semi-sedentary farms. When Habsatu was a child, the family lived a more nomadic life than they do today. Nevertheless, they still consider themselves nomads and respect their own norms and values, which distinguish them from the so-called sedentary farmers. Contact and exchange between groups is frequent, however, and has deep historical roots. Habsatu has a nice white phone (a 2010 model) in the small bag that she carries everywhere. She opens the bag regularly as she is called by her children who

76

Mirjam de Bruijn

are at school in Bamenda, by her customers who want to check on her progress with their dresses, and by her cloth provider who lives in the north and is waiting for money to be transferred through the credit union. She also receives calls from her uncle from Nigeria who is ill, from her younger sister from Nigeria who takes care of her eldest daughter (she sent her daughter to this sister, who has no children of her own), from the cousin from the farm who tells her that her uncle has died and asks if she can come for the funeral, and from her mother who lives in a village 30 minutes’ drive from Bamenda. The phone contains Habsatu’s life in a nutshell, connecting the important relations in her life socially, economically, and emotionally. She explains that the phone has brought two significant changes to her workshop and to her family ties. First, she no longer needs a place in the market where people can find her; she now has her workshop in her house up the hill. Customers reach her by phone. Second, the phone allows her to reestablish contact with family members who live far away. She now has regular contact with her daughter; she receives news from her every week and, if she wants it, every day. This new connection has led her to appreciate even more her relations with the Fulani, the nomadic community, and she emphasizes her relationship with the people she knows on the farms. Habsatu goes there regularly and receives news now and then. The mobile phone has helped her reconnect. Reconnection is increasingly an urgent need, and it has drawn her back to her Fulani heritage. The contact established with her uncle in the United States is probably the best example. He left the family 25 years ago to join a nongovernmental organization (NGO) in the United States, having been one of the few in his family who received a good education. But after he left, the family hardly ever spoke to him, as they could only reach him through the land line at the post office in Bamenda. This was an expensive endeavor, but with the arrival of the mobile phone, they were able to connect regularly, and he has even visited them in Cameroon. The story of Habsatu and her connections can be viewed in a wider picture of Fulani culture in the region. Though not yet completely clear, what can be seen in the region is a unification of the Fulani in associations that have a strong basis in the new communication technologies and social media. The Fulani have united in the Tabital Pulaaku association, which promotes Fulbe culture, organizes festivities related to Fulani identity that reinforce the work of local organizations like Mboscuda, with which Habsatu is involved, and fights for the rights of the Fulani. This new connection at the community level runs parallel to the developments and changes Habsatu is experiencing in her own life. Another interesting development is Facebook and the encounters that are organized between young people in Africa and are linked to the diaspora. These activities are increasingly part of the communication scape of Africa. In some cases, Internet activities are an extension of social contact, enabling distant friends and family to share their lives not only with voice, but also with pictures and other means. In an observation of such a group of Chadians (D. Seli, pers. comm.), it appeared that the Facebook encounters also had political content and were per-

mobile telephony and socioeconomic dynamics in africa

77

haps related to frustrations and political identities. A similar finding was reported by a researcher who was trying to understand Facebook interactions related to the political elections in Cameroon in November 2011 (Meester 2011). In this case, Facebook users tried to organize people to vote and protest against the regime; however, their efforts barely materialized on the ground.

Discussion: Balancing the Good and the Evil of Mobile Phone Communication NEW DEPENDENCE RELATIONSHIPS

A recent development in MTN policy reveals the unpredictability of the mobile telephone market. MTN is increasingly trying to encourage customers to buy airtime directly through their MTN account. This will bypass the middleman system in airtime sales and result in the loss of a significant number of jobs in the sector. For phone repairers, the arrival of Chinese models on the market also means a new development in their careers but will not lead to more repairs. These phones have little hardware and can only be repaired if the right software is available. Not all repairers can afford this software, and the repair business will soon be run by only the cleverest and wealthiest young men, like Appolinaire. The lower echelons of repairers will once again have to search for new forms of employment. In these new dependence relationships, international companies and strong non-African-based businesses will ultimately be the winners, both by obtaining access to knowledge that is developed by others and by creating dependence relationships in an international market that fluctuates with demands and competition that have nothing to do with the local market in Africa. At the same time, the local market and the local dealers and repairers will increasingly be dependent on a global network for their knowledge, spare parts, and clientele. SOCIAL CHANGE AND COMMUNICATION: BETWEEN TRUST AND BETRAYAL

The mobile phone companies’ services have been developed in such a way that they can serve different African economic sectors. Although the African economy is growing rapidly (in some countries at a rate of more than 6 percent a year), most people live in poverty. Poverty has even become part of company policies that strive to reach out to the poorest. In Sudan and Chad, for example, companies were eager to roll out in war regions where they were expecting big returns (cf. Brinkman, de Bruijn, and Bilal 2009). Recent developments in the poverty market mean “business for the poor,” such as the appearance of different organizations and NGOs that develop communication strategies for the poor. The poor are proving to be a profitable market (see, for instance, www.cgap.org). The idea behind the extension of services to the poor is that mobile communication and derived services like mobile banking should be accessible to

78

Mirjam de Bruijn

everybody. Especially in economies where poverty limits communication, mobile telephony offers opportunities to earn an income both for the companies, as they exploit the poverty economy, and for the individual users. Most individual users will answer the question “What does the phone do for you?” with “It helps.” It helps to link people who are far away to keep connections going, and it helps those in need obtain money. Mobile phones have also led to mistrust and even anger among users. For instance, the technique of calling without spending any money is well developed in most African countries, where users place a call just to alert the other person that they want to talk and then immediately hang up (cf. Donner 2007). This “beeping” or “flashing” technique has recently been extended with special services like MTN’s “callmeback” service. The recipient of such a call, who is often wealthier, is then expected to call back. An unexpected outcome, however, is that those who are targeted by these calls are becoming annoyed and only return them when they have to. Text messages are increasingly popular, too, as these are free with some services (in South Africa, Mxt or Whatsapp). Texting is of course only useful for people who can read and write, although people who are illiterate can in fact quite easily acquire the skills required for texting (Hahn and Kibora 2008). Despite these free services, most Africans believe that companies inevitably earn a lot of money from this business. There has been increasing dissatisfaction with companies’ attitudes (cf. Obadare 2006), but it has not yet resulted in boycotts or a change in company policy. On the contrary, companies have been able to surround themselves with people who are part and parcel of this “communication family.” Molony (2007), who was one of the first researchers to do in-depth research into new styles of communication and society, concluded that mobile phone communication does not create the same sense of trust as face-to-face relationships do. How these relationships are shaped depends on the circumstances and the political-social context. In Molony’s case, the context was trade relations in Tanzania at the start of the mobile phone era. Another case is Chad, where, after a long war and period of conflict, the phone was not trusted because it could be used to trace people. At the same time, the phone was central in rebuilding relationships there between people who had fled war zones (cf. Seli 2012). Recent research in South Africa has explored gender relations in relation to the mobile phone. The phone shows who one is (status), but it has also become an element in the building of relationships. Dates are confirmed by sending airtime. A girl is keen to have a boyfriend because he will pay for her airtime, and they will use the phone to share secret messages. But the phone can also betray people, revealing secrets to people who should not know them. Anecdotes about husbands who suspect their wives of having affairs or boyfriends who no longer trust their girlfriends are common (cf. Archambault 2011; Vuyani 2012). In Sudan, the mobile phone has created some freedom for Muslim women who cannot leave their own houses but can organize a (hidden) social life with the help of the phone (Brinkman, de Bruijn, and Bilal 2009). These examples demonstrate how the mobile

mobile telephony and socioeconomic dynamics in africa

79

phone is creating both trust and mistrust between people and, as such, is altering the division between private and public spheres. GLOBAL SHADOWS

This chapter does not elaborate on the cases of Habsatu and the others but presents them to show an ongoing development resulting directly from the mobile phone revolution. The new forms of communication have enabled users to strengthen or reestablish connections. These connections may lead to new social fields cross-cutting traditional (national) borders that then become units with newly defined boundaries, demarcated for social, political, or economic purposes. The cases presented relate to situations in which people define themselves as being on the global margins, but these margins are connected over distance and develop in and give form to the “global shadows” of our world (Ferguson 2006).

Conclusions De Bruijn and van Dijk (2012) recently proposed seeing the world today as having entered the postglobal era in which connections are abundant and taken for granted. Many regions of Africa are approaching this situation. Life without a mobile phone is unthinkable for the younger generation, and mobile phone technology is significantly influencing urban and rural economies. This form of communication may indeed be considered a resource. However, we should always question whether this resource is accessible to everybody and if and how people are accessing it (figure 3.9). This chapter has sketched different ways of accessing the new communication technology and has argued that, through this technology, new social positions with economic and political power are being created, social relationships are reinforced, and in some cases these relationships are taking new forms, leading to a redefinition of the social or a reinforcement of existing relationships and thus social boundaries. This new technology is not neutral, and connections construed through it are following old paths or forging new ones, thus molding social and political hierarchies that already existed or have just come into existence. If development is progress and instigated by new infrastructure like the mobile communication network, then development also means the reformulation of social hierarchies that might offer chances to some while limiting possibilities for others. This is what progress and social change always do. This new form of connectivity is also producing winners and losers. If modern communication is the social glue of society (Vertovec 2004), who then is being glued, and what kind of society is being created? The world is increasingly ordered in “global shadows” (Ferguson 2006), which are worlds of their own on the margins of the wider world (Das and Poole 2004). Communication in the shadows is being affected by information and communications technologies, and people are becoming connected over larger distances. Is this

80

Mirjam de Bruijn

Figure 3.9 Mobile Telephony in Reach of Everybody? Women Carrying Water in Darfur, 2007

Photo: Mirjam de Bruijn.

leading to a world where links over distances are more important than links with neighbors? I posed this question in my inaugural lecture at the African Studies Centre in 2008. It is difficult to find a good answer, but in general, current research encourages us to answer in the affirmative.

references Archambault, J. 2011. Breaking up because of the phone and the transformative potential of information in South Mozambique. New Media and Society 13(3):444–456. Batchelor, S., M. Kashorda, and F. S. Sylla. 2009. M-banking: An African financial revolution. Copenhagen: UN Economic Commission for Africa/Centre of African Studies, University of Copenhagen/International Books.

mobile telephony and socioeconomic dynamics in africa

81

Brinkman, I., M. de Bruijn, and H. Bilal. 2009. The mobile phone, “modernity,” and change in Khartoum, Sudan. In Mobile phones: The new talking drums of everyday Africa, ed. M. de Bruijn, F. B. Nyamnjoh, and I. Brinkman. Bamenda, Cameroon: Langaa; Leiden, The Netherlands: African Studies Centre. Castells, M., M. Fernández-Ardèvol, J. Linchuan Qiu, and A. Sey. 2007. Mobile communication and society: A global perspective. Cambridge, MA: MIT Press. Das, V., and D. Poole, eds. 2004. Anthropology in the margins of the state. Oxford, U.K.: James Currey. de Bruijn, M. 2008. “The telephone has grown legs”: Mobile communication and social change in the margins of African society. Inaugural address for the African Studies Centre, Leiden, The Netherlands (5 September). de Bruijn, M., F. B. Nyamnjoh, and T. Angwafo. 2010. Mobile interconnections: Reinterpreting distance and relating in the Cameroonian grassfields. Journal of African Media Studies 2(3):267–285. de Bruijn, M., F. B. Nyamnjoh, and I. Brinkman, eds. 2009. Mobile phones: The new talking drums of everyday Africa. Bamenda, Cameroon: Langaa; Leiden, The Netherlands: African Studies Centre. de Bruijn, M., and R. van Dijk, eds. 2012. The social life of connectivity in Africa. New York: Palgrave Macmillan. Donner, J. 2007. The rules of beeping: Exchanging messages via intentional “missed calls” on mobile phones. Journal of Computer-Mediated Communication 13(1). http://jcmc.indiana.edu/vol13/issue1/donner.html. Ekyne, S. 2010. Introduction. In SMS uprising: Mobile activism in Africa, ed. S. Ekyne. Cape Town, South Africa: Pambazuka Press. Etzo, S., and G. Collender. 2010. The mobile phone “revolution” in Africa: Rhetoric or reality? African Affairs 109(437):659–668. Ferguson, J. 2006. Global shadows: Africa in the neoliberal world order. Durham, NC: Duke University Press. Granovetter, M. 1973. The strength of weak ties. American Journal of Sociology 78(6):1360–1380. Hahn, H. P., and L. Kibora. 2008. The domestication of the mobile phone: Oral society and new ICT in Burkina Faso. Journal of Modern African Studies 46(1):87–109. Horst, H. A., and D. Miller. 2006. The cell phone: An anthropology of communication. London/New York: Berg. International Energy Agency. 2012. World energy outlook. Paris: International Energy Agency. International Telecommunications Union. 2011. World telecommunications development report. Geneva: International Telecommunications Union. Jack, W., and T. Suri. 2011. Mobile money: The economics of M-Pesa. NBER Working Paper No. 16721 (January). Latour, B. 2005. Reassembling the social: An introduction to actor-network theory. Oxford, U.K.: Oxford University Press. Ling, R. 2008. New tech, new ties: How mobile communication is reshaping social cohesion. Cambridge, MA: MIT Press. Meester, L. 2011. Virtual representations, social realities: On online social networks of Cameroonian political activism and their offline systems of reciprocity. Research proposal, Master of Arts African Studies, Leiden University.

82

Mirjam de Bruijn

Molony, T. 2007. “I don’t trust the phone; it always lies”: Trust and information in communication technologies in Tanzanian micro and small enterprises. Information Technologies and International Development 3(4):67–83. Nkwi, W. 2009. From the elitist to the commonality of voice communication: The history of the telephone in Buea, Cameroon. In Mobile phones: The new talking drums of everyday Africa, ed. M. de Bruijn, F. B. Nyamnjoh, and I. Brinkman. Bamenda, Cameroon: Langaa; Leiden, The Netherlands: African Studies Centre. ———. 2011. Kfaang and its technologies: A social history of mobility in Kom (Cameroon), 1928–1998. Ph.D. diss., African Studies Centre, Leiden University. Obadare, E. 2006. Playing politics with the mobile phone in Nigeria: Civil society, big business and the state. Review of African Political Economy 33(107):93–111. Roller, L.-H., and L. Waverman. 2001. Telecommunication infrastructure and development: A simultaneous approach. American Economic Review 91(4):909–923. Sangaré, B. 2010. Peuls et Mobilité dans le cercle de Douentza: l’espace social et la téléphonie mobile en question. Master’s thesis, University of Bamako. Seli, D. 2012. La communication Mobile au Tchad: entre confiance et conflit (working title). Ph.D. diss., African Studies Centre, Leiden University. van Beek, W. E. A. 2009. The healer and his phone: Medicinal dynamics among the Kapsiki/Higi of North Cameroon. In Mobile phones: The new talking drums of everyday Africa, ed. M. de Bruijn, F. B. Nyamnjoh, and I. Brinkman. Bamenda, Cameroon: Langaa; Leiden, The Netherlands: African Studies Centre. Vertovec, S. 2004. Cheap calls: The social glue of migrant transnationalism. Global Networks 4(2):219–224. Vuyani, H. 2012. Mobile phones in South Africa. Research report. SANPAD/UCT. Waverman, L., M. Meschi, and M. Fuss. 2005. The impact of telecoms on economic growth in developing nations. www.buzzinbees.com/docs/Leonard%20Waverman %20-%20mobile%20penetration%20and%20GDP%20growth.pdf.

commentary Anthony M. Townsend Mirjam de Bruijn’s chapter fills an important gap in current research on the impact of mobile phones in Africa, a topic that is receiving increasing attention in many sectors: development, the telecommunications industry, and among innovation and design strategists. However, most of those reports have been limited to either breathless recitations of the growth statistics or technological fetishism over the unfamiliar devices and practices of mobile telephones in Africa. Most reports have been written by people looking in from the outside to see what of value they can extract. Instead, de Bruijn offers us a human, street-level context for this tumultuous change: the view of Africans themselves. Her work is new to me, but I find it reflects a sensibility similar to that of others like Nancy Odendaal in South Africa, who are bringing a rich contextual perspective to the study of mobile communications’ role in African economies and societies. De Bruijn’s approach to thinking about technology—essentially that of a social constructivist—is one with which I sympathize. In particular, I think there is no other way to understand mobile communications. In my work, I argue that urbanization and ubiquity (the end state of deploying ubiquitous computing) are deeply intertwined, and they have been for a half century already. The basic ideas for cellular telephony were developed shortly after World War II, but much like the optical telegraph, invented in Napoleon’s time, there was no demand yet for it. Slowly, increasing mobility has increased demand for mobile communications, and leaps in technology have driven down cost and increased capacity. The technology of mobile phones—from single-tower to cellular, from analog cellular to digital cellular, from 3G to 4G—is driven by the density of demand in cities. However, technological and infrastructural advances in turn unlock growth, by allowing people to organize more complex choreographies of daily life, which accelerates the metabolism of cities. Mobile telephones are not just society, as de Bruijn notes; they are the city. I like to think that the end result is that we are all becoming nomads again, connected together by this digital telepathic infrastructure of mobile communications. I think of Nikola Tesla’s great prognostication almost a century ago, in Collier’s magazine in 1926: “When wireless is perfectly applied the whole earth will be converted into a huge brain, which in fact it is, all things being particles of a real and rhythmic whole.” As a Westerner, I find it fascinating—as I consider what it will be like to become a nomad again—to reflect, as de Bruijn does, on what we can learn from actual nomads in Africa who are becoming digital. This raises a larger point that is woven throughout this chapter: Africans are inventing a completely different Internet and a completely new form of computing than what we’ve experienced in the North. It is based on the mobile phone and not the PC, it is mobile instead of being fixed, and it is based on services 83

84

Anthony M. Townsend

instead of content. Globally, this is how computing is done by most people now. Just look at the statistics. As of 2008 there were more mobile broadband lines than fixed ones, and this transition is being driven by market forces, not development aid. In the last five years, the One Laptop per Child program started by Nick Negroponte of the MIT Media Lab has deployed a couple million laptops to mostly middle-income countries. In the same time, Nokia and its competitors have doubled the world’s installed base of mobiles from 2.5 billion to more than 5 billion. In my work, I call the mobile phone “a computer for the rest of us.” After 20 years online, and 10 years doing technology forecasts, I rarely get excited about technology anymore, but this trend excites me. As de Bruijn has documented, cheap Chinese smartphones are surging into the wealthier markets in Africa. In a decade, or sooner, we’ll have several billion people walking around the megacities of the global south—many of them slum dwellers—with cloudserved supercomputers in their pockets. As we shift to a more mobile web in the North, some people in the technology community are starting to wonder if the flow of technology transfer and business innovation will start to reverse. Will Africa become a hotspot of innovation? There are signs already in Kenya, South Africa, Ghana, and elsewhere that this is on the cusp of happening. There is a lot to learn from Africa’s experience with mobiles about rapid, on-the-cheap, usercentric ways of stretching scarce dollars for infrastructure investment. What I am left wondering from de Bruijn’s chapter is what it will take to fill in the gaps. Africa’s mobile networks are far from ubiquitous. The stories in her chapter about people climbing trees or walking to high ground to obtain a signal from a cell tower dozens of miles away are fascinating. In the poorest and most rural countries, only 30 percent of the territory is served. What is the business case for extending these networks? In fact, in the United States we have the same problem with both wired and wireless broadband: urban areas are always the first to get served, and rural areas lag badly in attracting investment. This really hit home when I read de Bruijn’s stunning observation that mobile companies are eager to enter the conflict zones of Sudan and Chad. The physical upheaval, volatility, and uncertainty of war turn out to be profound drivers of demand for mobile communications. What I wonder, though, is whether the economic pressure of globalization—as it increasingly penetrates Africa (literally invading from the coasts in the form of several new fiber-optic cables ringing the continent)—is as fundamentally destabilizing as armed conflict, especially for the poor. Is the popularity of mobiles merely a desperate attempt by the poor to mitigate the violence that the global economy inflicts on their social and economic lives?

Finance, Regulation, and Taxation

4 Economic Regulation of Utility Infrastructure Janice A. Beecher

P

ublic infrastructure has characteristics of both public and private goods and earns a separate classification as a toll good. Utilities demonstrate a variety of distinct and interrelated technical, economic, and institutional characteristics that relate to market structure and oversight. Except for the water sector, much of the infrastructure providing essential utility services in the United States is privately owned and operated. Private ownership of utility infrastructure necessitates economic regulation to address market failures and prevent abuse of monopoly power, particularly at the distribution level. The United States can uniquely boast more than 100 years of experience in regulation in the public interest through a social compact that balances and protects the interests of investors and ratepayers both. Jurisdiction is shared between independent federal and state commissions that apply established principles through a quasi-judicial process. The commissions continue to rely primarily on the method known as rate base/rate-of-return regulation, by which regulators review the prudence of infrastructure investment, along with prices, profits, and performance. Regulatory theory and practice have adapted to emerging technologies and evolving market conditions. States—and nation-states—have become the experimental laboratories for structuring, restructuring, and regulating infrastructure industries, and alternative methods have been tried, including price-cap and performance regulation in the United Kingdom and elsewhere. Aging infrastructure and sizable capital requirements, in the absence of effective competition, argue for a regulatory role. All forms of regulation, and their implementation, can and should be Review comments from Tim Brennan, Carl Peterson, Ken Costello, David Wagman, and the Lincoln Institute of Land Policy are greatly appreciated.

87

88

Janice A. Beecher

evaluated in terms of incentives for infrastructure investment and operational performance.

The Role of Public Utilities Public utilities are the enterprises that provide vital telecommunications, energy, and water services. Depending on one’s perspective, public utilities supply products, commodities, services, information, common carriage, networks (interconnectivity), and access (to multiple providers); foundations for development and society; and engines for technological advancement and prosperity. Utilities help satisfy physiological needs by providing creature comforts (heat, light, and safe drinking water). Essential utility services thus become matters of humanity and human rights in both the developing and developed worlds. Beyond basic needs and living standards, utility services make modern lifestyles possible to a degree that goes largely unnoticed until the slightest disruption occurs. In the United States, a very high level of service reliability is presumed; outages and the cascading consequences of interdependency are met with alarm and dismay. Extended disruptions jeopardize public health, safety, and welfare, and destabilize economic and social systems. It is no wonder that utility facilities are strategic targets for terrorism. The essential role of utilities in U.S. and global development cannot be overstated, as observed 80 years ago: To us of this day these utilities are necessities. It is hard to realize how people lived without them only fifty years ago. They are at our command so easily and so cheaply that we accept and use them as a matter of course—the commonplace things of daily life—without a thought of how they got here. . . . The story of their beginnings, their growth and their place in the social, industrial, and economic fabric of the nation is romance made reality. It is the romance of every day life, by the realization of which we get to know and understand and appreciate better, the times in which we live. (Robinson 1932, 2)

Despite many dimensions and meanings, public utilities essentially deliver services through networks of built infrastructure that are ubiquitous but largely invisible; even noticeable network elements go largely unnoticed unless they have particularly distinctive and controversial presences (such as nuclear plant cooling towers). The physical imprint of utility infrastructure on the U.S. landscape is impressive, as recent estimates reveal:1 1. These data (verified as feasible) are from periodic online publications of the U.S. Energy Information Administration (2004), the U.S. Environmental Protection Agency (2009a), the North American Electric Reliability Corporation (2011), the Federal Communications Commission (2010c), Silverstein (2011), and industry trade associations.

economic regulation of utility infrastructure

t t t t t t t t t t t t t t t t t t

89

66 nuclear, 580 coal, 1,169 petroleum, and 1,705 gas plants 1,432 hydroelectric and 39 pumped storage facilities 1,356 renewable energy facilities (nonhydro) 395,000 miles of high-voltage (>100 kV) transmission lines 15,700 transmission substations 6 million miles of electricity distribution lines 20,000 miles of gas-gathering pipelines 306,000 miles of interstate and intrastate transmission pipelines 1,400 gas compressor stations 400 underground natural gas storage facilities 2 million miles of gas distribution mains 75,000 water treatment facilities 2 million miles of water distribution mains 14,500 wastewater treatment facilities 600,000 miles of wastewater collection lines 18.7 million equivalent telephone poles 1.7 billion miles of metallic wire 38 million miles of fiber wire

The United States finds itself in the early stages of what is sure to be an intensive and protracted infrastructure replacement cycle that is not exclusive to utilities but for which investment will be significant and complicated. Depending on conditions, the per-mile cost for replacing water pipes can be as much as $500,000; the per-mile cost of electricity transmission lines ranges from $1 to $15 million. The staggering capital requirements on the sectors will be borne largely by utility ratepayers. The “funding gap” has become a popular construct for rationalizing an accelerated spending pace for infrastructure across the sectors. A gap may be apparent when replacement rates and depreciation practices are unrealistic relative to the actual useful life of utility assets, considering both physical deterioration and technical obsolescence. A lagging replacement rate suggests unrealistic expectations about life expectancy and life-extension potential. A 2006 survey found that U.S. water systems had replaced 2.8 percent of their existing pipe in five years (U.S. Environmental Protection Agency 2009b), a pace that many industry experts consider insufficient. A gap may also arise from failure to meet standards. Many analysts suggest that planned and timely renovation is more cost effective than emergency management after breakage. Much of the concern about the nation’s infrastructure is a function of the age of the existing physical stock along with a perception of disrepair and even disintegration. Though most power systems engineers probably disagree, former Energy Secretary Bill Richardson repeatedly characterized the United States as a “superpower with a third-world electrical grid.” In recent years, the number of outages affecting the bulk power grid has increased, but more were due to

90

Janice A. Beecher

weather-related events than to infrastructure failure alone.2 Fortunately, significant natural gas pipeline safety incidents are not trending upward, although events occur regularly and sometimes tragically.3 The American Society of Civil Engineers (ASCE) has been especially vocal about the condition of the nation’s infrastructure, giving it a D grade in its periodic report card (American Society of Civil Engineers 2009) and bringing attention to perceived expenditure deficits as well as the consequences of “failure to act.” Transportation and utilities have been a focus of federal funding under the American Recovery and Reinvestment Act (ARRA) of 2009 as a means of both stimulating the broader economy and accelerating the pace of renovation and replacement. While immediate investment needs are estimated in the billions of dollars, the long-term projections for all of the sectors could total a few trillion. Complicating the challenge are transformational goals related to market structures, technological advancement, and environmental stewardship, namely the reduction of greenhouse gases (especially decarbonization) in the context of climate change. Complicating the economics are the effects of rising costs and prices on discretionary demand for utility services. For the electricity industry, ASCE estimates that by the year 2040, the nation needs to invest $401 billion in generation, $112 billion in transmission, and $219 billion in distribution, for a total of $732 billion (2010 dollars). Other estimates double or even triple that amount (Chupka et al. 2008). ASCE also estimates that failure to meet that need will cost much more to the U.S. economy (American Society of Civil Engineers 2011): $126 billion to businesses, $71 billion to households, $656 billion in personal income, $496 billion in gross domestic product (GDP), $10 billion in exports, and 529,000 jobs. ASCE provides comparable estimates of economic impact for the water sector. Other analysts put the price tag for a transformed electricity industry much higher, at $1.3 trillion (Eggers 2010), when considering the cumulative investment in smart meters ($22 billion), environmental compliance ($120 billion), transmission for renewable energy ($167 billion), nuclear replacement ($200 billion), compliance with carbon regulation ($250 billion), and compliance with renewable portfolio standards ($500 billion). Other estimates put the 20-year need for transmission alone at about $300 billion (Chupka et al. 2008). Infrastructure needs for the natural gas industry are not as daunting but still considerable. Newly discovered shale reserves and concerns about agerelated pipeline safety are investment drivers for gas transmission and distribution systems. The investment profile reflects the market structure of the industry, which separates wellhead production and gathering, interstate transmission, and

2. Detailed annual events analyses and system disturbance reports can be found at the North American Electricity Reliability Corporation (NERC) website (www.nerc.com). 3. Incident data are available at U.S. Department of Transportation, Pipeline & Hazardous Materials Safety Administration (http://primis.phmsa.dot.gov/comm/reports/safety).

economic regulation of utility infrastructure

91

local distribution. The natural gas supply industry estimates “midstream” infrastructure investment needs at $133 to $210 billion, primarily for pipeline assets (INGAA Foundation 2009). A survey by the U.S. Environmental Protection Agency (2009a) put the total national investment need of the drinking water industry at $335 billion, with 60 percent for transmission and distribution (2007 dollars). Wastewater and storm water management needs total another $300 billion (U.S. EPA 2010). More recently, the American Water Works Association (2012) called for $1 trillion of investment in the nation’s drinking water systems over the next 25 years and $1.7 trillion by 2050, with replacement and population growth about evenly responsible for the need. Unlike its sister sectors, the water industry is actually experiencing flat or declining demand (Zeilig 2011), likely due to the combined influence of efficiency standards, prices, and cultural shifts. Compared to energy, water also has few if any new uses. A “new normal” in water demand is cause for reconsidering drinking water infrastructure investment in favor of optimization solutions. Given that technology has set the replacement pace, the telecommunications sector is better positioned. Although still vital, the risks associated with disruptions are generally not as ominous. Nonetheless, federal regulators have referred to broadband as “the great infrastructure challenge of the early 21st century” (Federal Communications Commission 2010b). A gap analysis in this sector calls for a $23.5 billion investment to reach 7 million unserved homes (Federal Communications Commission 2010a). The economic cost of the “digital exclusion” of more than 40 million households due to a lack of access, affordability, or understanding has been estimated at $55 billion (Digital Impact Group 2010). Estimates of current investment, planned investment, and “need” are variable, subjective, and inexact. Range estimates help but may still suggest a false sense of precision. Actual needs will be very system specific, depending on the nature and condition of local infrastructure. Each estimate of need for supply presumes a level of usage, but usage changes as prices and other demand determinants change. As with all forecasts, longer time frames add to modeling uncertainty. Nonetheless, it can be said with some confidence that infrastructure requirements are the core driver of past, present, and future utility service costs, as well as a defining characteristic of traditional public utilities generally.

Characteristics of Public Utilities Public utilities, at least traditionally, shared several salient qualities that reinforce their characterization as monopolies. Utility services are distinguishable from both public and private goods because they share attributes attached to both. Utility infrastructure and network services are rightfully characterized as toll goods, generally defined by economists in terms of lower rivalry and higher excludability. Owing to British common law, and echoed by the U.S. Supreme Court in Munn v. Illinois (94 US 113, 1877), utilities are also understood as

92

Janice A. Beecher

inherently “affected with a public interest,” which sets them apart in terms of not only function but also purpose. The term public is not accidental. Utilities provide essential service to the public. Utilities provide public works that in many respects substitute for public services. Utilities make use of public rights-of-way as they build infrastructure for public use, operating under certificates of public convenience and necessity and assuming an obligation to serve along with entitlement to just compensation. Utility companies issue publicly traded stock, and good utilities strive to be deserving of the public’s trust. Traditionally, public utilities were regarded as monopolistic, often labeled as “natural” monopolies. Even across sectors, commonalities and similarities overwhelmed differences. This characterization has become somewhat antiquated in the context of evolving technologies and markets. Although monopolies once were considered creatures of their technical and economic traits, monopolies are also creatures of institutions—an amalgam of laws, regulations, and policies. The view that utility monopolies are as “artificial” as they are “natural” rattles some long-held assumptions. Today, technological differences among the sectors are increasingly relevant to structural design. Nonetheless, some of the inherent traits of utilities that tend to reinforce their market power are difficult to overcome. The traditional technical, economic, and institutional traits of the monopolies responsible for utility infrastructure are reviewed qualitatively here. Of course, these categories and the indicators within them are nonexclusive and reinforcing. All relate ultimately to the rationale for economic regulation and its form and function. TECHNICAL CHARACTERISTICS

The technical features of utilities can be defined largely in engineering terms, although they include certain production-cost characteristics that have economic implications. The everyday tasks of flicking a switch, adjusting the thermostat, turning on a faucet, or tapping a keypad are made with little regard to underlying technologies and networks. Most people can only marvel at how electricity infrastructure meets supply with demand in real time; at how natural gas infrastructure collects, stores, and distributes a combustible yet relatively clean and efficient source of energy; at how telecommunications infrastructure integrates wired and wireless systems to relay billions of calls daily; and at how water infrastructure delivers monthly to a typical American home more than 20 tons of a product that is safe to ingest.4 Thus, a leading distinction of utilities is that they provide “safe, adequate, and reliable” service “on demand” at an actual moment’s notice. Utilities are expected to be “ready to serve” with available and dispatchable supplies (a source of controversy for intermittent or variable resources, such as wind energy).

4. Based on 5,000 gallons per month and eight pounds per gallon.

economic regulation of utility infrastructure

93

The value of reliability is very high, but mostly unappreciated. Utility systems must balance supply and demand both spatially and temporally. Utilities have distinctive load characteristics marked by daily, weekly, and seasonal cyclicality that is also affected by factors such as weather. Reliability requires utilities to maintain adequate reserve margins for meeting peak loads. Traditional utility monopolies were vertically integrated; that is, they were responsible for any and all necessary functions, including supply, storage, transmission, distribution, and administrative functions. They operated closed networks with limited or no access by alternative providers. A single service provider or monopoly avoided wasteful duplication or redundancy within the network (such as pipes, wires, or towers) and facilitated achievement of scale economies in production (discussed below). Public utilities are sometimes referred to as “fixed utilities” because they require substantial capital investment in long-life, mostly nonfungible fixed assets that, along with technical knowledge, poses a formidable barrier to market entry. Accordingly, utilities exhibit exceptional capital intensity, measured by the ratio of fixed assets (as shown on the balance sheet) to operating revenues (as shown on the income statement) (figure 4.1). Water and wastewater utilities are particularly capital intensive due to long asset lives. New infrastructure investment (e.g., power plants) traditionally comes in very large increments, a trait known as “lumpy capacity.” The scale of facilities makes siting politically sensitive due to local environmental impacts. The long lives of utility assets also raise intergenerational equity issues with regard to financing and financial accounting practices (namely, depreciation). Finally, utilities themselves are technically interdependent. Electricity and gas can be substituted and provided by “converged” utilities. Natural gas and water resources are used in electricity production. Electricity is needed for natural gas and water operations. Water is needed for wastewater service and fire protection. All utilities rely on telephony for operational intelligence and security, including “smart” metering. Many essential social and economic functions depend on the services supplied by utilities, reinforcing the sense of public interest in safe, adequate, and reliable operations. ECONOMIC CHARACTERISTICS

Although the direct value they add to the nation’s gross domestic product is somewhat modest, utilities are significant players in the U.S. economy. Private utilities add about 2 percent in value to the nation’s GDP (figure 4.2).5 The economic characteristics of traditional public utilities are intrinsically related to their underlying technical characteristics, so much so that a subspecialization of “engineering-economics” was once recognized in the field. Although scale econ-

5. Utilities have dedicated stock indexes; once considered “widow and orphan stocks,” their securities remain a stalwart source of dividend income in investment portfolios.

94

Janice A. Beecher

Figure 4.1 Capital Intensity of Major Companies and Public Utilities for 2012 7.0

Net fixed assets/revenues

Total assets/revenues

Ratio of assets to revenues

6.0 5.0 4.0 3.0 2.0 1.0

Am

eric

an Nike Exp Mi ress cro sof Bo t For eing dM o Wa tor l-M Co art caCo la U Sp S S rin tee tN l ext el* Fed Co Ex mc Firs ast* Ex t Sol So xon M ar uth o we bil s AT& t Air* TI Mc nc.* D Wa onal ste d’s Mg m AG Exe t. L R lo e n Bu sourc * rlin es Ni gto * No Sour n SF* rth ce, eas In t c. Aq Utilit * ua ies Am * eric a*

0.0

Companies and utilities *Infrastructure-intensive network or utility company. Source: Based on publicly reported financial data (available at hoovers.com).

Figure 4.1

omies (declining unit costs ofLincoln_Ingram_Infrastructure production) can be realized in most enterprises, they are pronounced for utilities, particularly supply processes. Scale is achieved in both construction and operation. Scope economies may also be realized, including vertical integration of functions (generation, transmission, and distribution) and horizontal integration of complementary operations (electricity and gas). Utilities also benefit from network economies, meaning the primary and secondary benefits of coordination, interconnection, and “connectedness” among infrastructure providers and users. Both infrastructure and commodity costs are significant cost drivers for public utilities, especially for energy. The substantial fixed costs of infrastructure, however, can result in a marginal cost of production that is below average costs. As a consequence, the competitive cost of marginal-cost pricing would cause utilities to underrecover costs. Though scale economies are apparent, they are neither absolute nor unlimited (see Kwoka 2005). Economies in production are offset by eventual diseconomies of transmission and distribution, which is a function of service territory area and density. Although the economics are complex, some alternative technologies for

economic regulation of utility infrastructure

95

Figure 4.2 Value Added by Utility Infrastructure to the U.S. GDP, 1977–2011 5.0

14 4.0

12 10

3.0

8 2.0

6 4

1.0

Value added as a percentage of GDP

U.S. gross domestic product (GDP in $ trillions)

16

2 0 11

09

20

07

20

05

20

03

20

01

20

99

20

97

19

95

GDP ($ trillions) Broadcasting and telecommunications Utilities

19

93

19

91

19

89

19

87

19

85

19

83

19

81

19

79

19

19

19

77

0

Oil and gas extraction Pipeline transportation

Source: Based on data from the Bureau of Economic Analysis, NAICS category 22 (available at bea.gov).

the provision of energy and water services, namely distributed resources, conflict with long-held views about the cost advantages of centralized production. Utilities are distinctive in that product differentiation is very limited due to technological constraints as well as standards that allow little tolerance for deviation. Regardless of size or structure, for example, Figure 4.2 all electric utilities are expected to comply with voltage standards, and all water utilities are expected to comply with federal drinking water Lincoln_Ingram_Infrastructure standards. Traditionally, public utilities were also highly constrained in their ability to control supply or demand. Utility services have few if any practical substitutes, and switching to alternative technologies or providers is constrained. The utility’s core customers are considered captive because they have few if any choices. Again, modern utilities are more likely to engage in more active portfolio design and demand management, and technologies may open service choices to customers. Utilities typically serve a mixture of residential, commercial, and industrial load. Water utilities tend to rely more on revenues from residential water sales. Demand for utilities varies according to a number of drivers, including income and price. Demand is considered income-elastic but relatively price-inelastic

96

Janice A. Beecher

(that is, not highly price-responsive, as characterized by a more vertical demand curve). This is not to say that customers do not respond to price changes, but rather that a change in price will yield less than a proportionate change in quantity demanded. Importantly, however, price elasticities vary for different types of usage. Utilities make for regressive burdens; that is, low-income households pay a much greater share of their income for essential utility services than do highincome households. Rising costs thus have distributional consequences. Finally, the production and consumption of public utility services result in both positive and negative externalities. Though externalities are typically contemplated in terms of deleterious consequences, namely pollution, a fair accounting also considers the positive public health and welfare effects made possible through access to utility services. Individual utilities are also generally ill-equipped to respond effectively to the problems of externalities without raising the specter of monopoly rents. Externalities may be more appropriately managed through comprehensive public policies (such as standards, mandates, or taxes) with which all market participants must comply (Beecher 2011). INSTITUTIONAL CHARACTERISTICS

The institutional dimension reflects how utilities are organized, managed, and regulated. Institutions are also connected to the technical and economic features already explored. Although monopoly tends to connote enormity, utility monopolies may be small or large in size. To the captive customer, the size of the provider matters little. Larger utilities dominate U.S. markets, often operating multiple establishments or systems, even as numerous small providers serve adjacent and rural areas. The water sector is particularly bifurcated in terms of size distribution. Market power is generally concentrated, even in the water sector, where only 9 percent of the nation’s 50,000 systems serve fully 88 percent of the U.S. population served by community systems (U.S. Environmental Protection Agency 2009b). Some areas are served by converged utilities providing electricity and gas, water and wastewater, and even “water and light” municipal legacy systems. Utilities are also distinguishable in terms of their participation in wholesale (bulk) and retail exchange, as buyers, sellers, or both. Ownership is a particularly relevant structural feature of utilities (table 4.1). In terms of market share, the electricity, natural gas, and telecommunications sectors in the United States are dominated by private ownership, while public ownership prevails for water. Rights-of-way and powers of eminent domain are enjoyed by both publicly and privately owned utilities. Utilities must comply with environmental and other social forms of regulation regardless of ownership. Each sector is also relatively well professionalized and subject to a degree of self-regulation in the form of generally accepted standards and practices. Some practices are probably more variable in the public than in the private sector due to the economic regulation of the latter. Public utilities are usually enfranchised with conditions for an exclusive service territory. As a result, and along with technical and economic reasons, they

economic regulation of utility infrastructure

97

Table 4.1 Ownership Structure for Electricity and Water, 2007 Providers

Revenues ($ billions)

Percentage of Revenues

194 168 2,006 874 9

224.3 48.8 53.4 39.9 1.8

60.9 13.3 14.5 10.8 0.5

23,800 1,047 5,407 9,327 9,554

39.5 4.3 2.4 0.1

85.3 9.3 5.2 0.2

Electricity Utilities

Investor owned Power marketers Publicly owned Cooperatives Federal power agencies Water Systemsa

Publicly owned and operated Publicly owned with private partner Private for-profit Nonprofit Ancillary

Many water utilities operate multiple systems. Sources: Data from American Public Power Association (2012); U.S. Environmental Protection Agency (2009b).

a

face little competition or contestability and thus incur lower associated risks. In the absence of competition, privately owned utilities—and some publicly owned utilities—are regulated as sanctioned monopolies.6 As discussed below, the regulatory paradigm centers on a social compact that specifies utility rights and obligations. In the United States, the states and the federal government share jurisdiction for utilities, delineated generally by oversight of wholesale and retail markets, respectively, in accordance with constitutional power related to interstate commerce. Regulation by the Federal Communications Commission (FCC) and the Federal Energy Regulatory Commission (FERC) can be preemptive of the states. There is no federal economic regulatory presence in the water sector, where states have primacy. Most publicly owned or not-for-profit utilities are subject to some form of local oversight by a municipality or an independent governing board.

Structural Change in the Utility Sectors Once virtually indistinguishable, at least for regulatory purposes, each of the public utility sectors has seen considerable structural change. The restructuring

6. The Wisconsin Public Service Commission has comprehensive jurisdiction for municipal utilities.

98

Janice A. Beecher

movement transpired from the confluence of technological and political forces that challenged long-held presumptions about the vertically integrated monopolies, and thus the traditional regulatory paradigm. The underlying logic and intent are common to all of the industries, but restructuring is contingent on the immutable traits of each sector. By the late twentieth century, the major utilities were transforming. To some extent, services once provided by utilities became commodities provided by network industries. Deregulation in the telecommunications sector was largely driven by technological innovation that opened markets, enhanced choices, and made traditional regulation obsolete. Deregulation in the energy sector was more policy driven and focused on the separation of both production and transmission from distribution. Restructuring generally involves both vertical separation of functional responsibilities (in electricity, “gencos, transcos, and discos”) and horizontal competition among providers. Wholesale and retail markets are delineated and opened to entry. Open access to transmission networks and “wheeling” are provided with appropriate compensation. Competitive services are also “unbundled.” Restructuring introduces new functions and providers, such as licensed suppliers, aggregators, brokers, and marketers. Market-based tools (such as auctions, trading, and hedging) are implemented. Public policies are aimed at facilitating consumer choice (e.g., information campaigns and phone number portability). Restructuring may include “divestiture” of some regulated assets and recovery of certain transition costs (e.g., stranded investment). Finally, restructuring is accompanied by policy reforms, including both alternative regulatory models and selective deregulation when competition is sufficiently workable. The persistent need for market rules, however, means that restructured markets actually remain rather structured. Fundamentally, markets and competition require enabling technologies. Innovation can cause disruption or even “creative destruction” (Schumpeter 1942) within economic systems. The evolution of telephony, particularly wireless and broadband communications, provides the obvious example. Advances in computational power and information management that lower transaction costs have enabled markets for network services. Pursuit of economic interests by key stakeholders drove restructuring as well. New market entrants (such as upstarts MCI and Sprint) brought competitive pressure to bear on the AT&T monopoly, leading ultimately to an antitrust investigation by the U.S. Department of Justice. In the electricity sector, large-volume customers (such as ELCON members and water districts) sought to eliminate interclass cross subsidies and exercise purchasing power. Restructuring can partly be explained by rebellious attitudes: anti-incumbency, sympathy for underdogs, and the appeal of change for the sake of change. Restructuring rests on a theory that favors competitive markets over regulatory institutions for promoting social goals, namely efficiency. Proponents include academics, think tanks, and policy entrepreneurs within legislative and

economic regulation of utility infrastructure

99

even regulatory bodies (including the FCC and the FERC).7 International institutions, namely the World Bank, also have promoted market structures that facilitate private involvement in utilities. Deregulation or “liberalization” has an undeniable ideological connection to preferences for limited government (as espoused by both the Thatcher government and Reagan administration). A related and somewhat cynical rationale for deregulation is a perception of regulatory or “nonmarket” failure (see Wolf 1993) and the acceptance of imperfect markets over imperfect regulation. The promises of restructuring were many. Customers would find freedom from the captivity of monopoly providers and, presumably, not just “choice” but good choices promoting economic well-being. Markets would see an influx of new providers, as well as new products and services. The discipline of competition would keep market power in check while promoting innovation and entrepreneurship. Risks would be shifted from customers to investors. Efficiency gains would lower costs, and improved pricing would “de-skew” cost allocation and eliminate subsidies. Restructuring is an ongoing social experiment that finds each of the sectors, and market segments within them, in a different structural and regulatory status (table 4.2). Persistent technical distinctions and path dependence urge caution about transference and call for sector-specific market and regulatory design. Although much of the telecommunications sector has been deregulated, corporate consolidation and market power are ongoing concerns, and a number of critical public policy issues remain, including universal service, broadband deployment, emergency calling systems, critical infrastructure protection, and smart-grid convergence. The vertically segregated natural gas sector combines competitive production at the wellhead with federal oversight of interstate pipelines and state oversight of intrastate transmission and local distribution, as well as pipeline safety. Key issues include the economic and environmental impacts of developing and transporting unconventional gas resources and utilizing gas for electricity generation. The states are about evenly divided in terms of restructuring electricity markets, to mixed effect. Price escalation in the wake of restructuring caused many states to rethink their policies. Reconciling retail customer choice with the utility’s ongoing obligation to serve (and associated investment) remains a thorny issue. The interest in restructuring has been eclipsed somewhat by the imperatives of global climate change, energy security and resilience, and grid modernization, although “smart” technologies are also seen to enable consumer choice. Infrastructure investment, both centralized and distributed, continues to be the major cost driver. Only the water sector remains highly monopolistic and mostly vertically integrated. Water markets do not lend themselves to restructuring because they are

7. The assertion that deregulation is sometimes favored by regulators contradicts the concept of regulatory capture.

100

Janice A. Beecher

Table 4.2 Structural and Regulatory Status of the Public Utility Sectors Structural Status

Unregulated

Regulated

Electricity

Partial restructuring and wholesale competition with mixed results; some retail choice

Independent power generation; most nonprivate utilities

Interstate and unbundled transmission (federal); retail distribution (state); vertically integrated (shared)

Natural gas

Vertical segregation with competitive wholesale markets; some retail choice

Wellhead (commodity) gas production; most nonprivate utilities

Interstate transmission (federal); intrastate transmission and retail distribution (state); pipeline safety (shared)

Telecom

Oligopolistic with workable competition; regulation is limited in scope

Long-distance, wireless, Internet, and cable services; other services and equipment

Small independent providers (state); network access and universal service (shared)

Water

Generally integrated and monopolistic; some wholesale and contract activity

Most nonprivate utilities; most privatization contracts; most wastewater providers

All privately owned utilities and some nonprivate utilities (state only)

Note: Shared jurisdiction may reflect divided responsibility based on market structure as well as state implementation of federal policy.

largely unstructured and unregulated in the first place. In fact, rising costs and prices may call for more oversight. Wholesale water is not exchanged competitively. Structural change is seen mainly in the form of regionalization and consolidation, as well as public-private contestability. Water’s especially essential nature, along with its connection to local economies, human health, and the natural environment, presents a formidable barrier to overreliance on markets. At the turn of the millennium, the road to restructuring had turned a bit bumpy, and the states began to maneuver more cautiously. Restructured markets are likely no closer to perfection than regulated markets. Efficiency gains have been achieved, but not without trade-offs. Each sector continues to face persistent concerns about service quality and reliability, consumer access and affordability, market and utility performance, and long-term infrastructure investment. Not all utility functions can become sufficiently competitive under current technological and economic realities to achieve desired goals. Despite substantial evolution, there remains a significant role for economic regulation, particularly for distribution services.

Economic Regulation Economic regulation of public utilities is an essential form of governance in the context of market failure, as manifested primarily in the form of monopoly.

economic regulation of utility infrastructure

101

Monopoly provision of electricity, natural gas, and water services, at least at the distribution level, is considered economically efficient due to scale economies, capital intensity, and technological constraints that make competition impractical. Regulation is acknowledged as an essential but imperfect substitute, surrogate, or proxy for competition as well as a corrective policy instrument for a variety of other market failures.8 Given the foundational nature of utilities to modern society, regulation “in the public interest” also serves broader policy goals. U.S. economic regulation has its origins in British common law and progressive political movements. Each of the state commissions originated from the railroad commissions of the mid-nineteenth century; their jurisdiction and authority were modernized after the turn of the twentieth century. The federal and state commissions are complex agencies, simultaneously engaged in functions that are quasi-legislative (policy making), quasi-administrative (policy implementation), and quasi-judicial (adjudicative); the regulatory commissioner is at once an expert, a trustee, and a judge (Beecher 2008). A particular emphasis can be placed on the judicial role model because well-functioning regulation follows accepted administrative procedures and rules of conduct that ensure due process for all participants. Regulatory decisions can be appealed, but the courts generally focus on matters of law and defer to regulators as finders of fact with discretion within a broad “zone of reasonableness.” Commissioners may be appointed or elected (in 14 jurisdictions), but commissions are similarly structured to ensure a higher degree of political independence and continuity, assured by staggered terms, partisan balance, and constraints on commissioner removal. An independent professional staff—well qualified and trained in law, economics, accounting, finance, engineering, and policy analysis—is also essential to effective regulation. In practice, economic regulation has three critical dimensions: jurisdiction (who gets regulated), authority (what activities are regulated), and methods (how regulatory oversight is administered). Jurisdiction, authority, and methods combine to create a variety of regulatory models. Ratemaking is a core function, but modern commissions also control market entry and exit and system expansion; ensure safety, adequacy, and reliability; specify standards and terms of service; process and resolve customer complaints; impose systems of accounts; require annual reports and conduct audits; approve capital structures and financial issuances; review and place conditions on mergers, acquisitions, affiliate transactions, and diversification; conduct prudence reviews and management audits; review resource and infrastructure plans; review forecasts for supply and demand; and ensure openness, transparency, due process, and ethical conduct. Regulation seeks a balance between the interests of utility investors, who devote their capital to utility infrastructure, and core or captive ratepayers, who

8. Competitive markets, of course, are also imperfect in reality and result.

102

Janice A. Beecher

depend on utility services but have limited choices. A putative social compact specifies the various rights and obligations of utilities. The utility enjoys an exclusive franchise for a certificated service territory, protection from competition and antitrust, an opportunity to recover costs and earn a reasonable return on prudent investments, rights of eminent domain, and the ability to charge for the cost of service. The utility also accepts an obligation to provide all paying customers with safe, adequate, reliable, and nondiscriminatory service on just and reasonable terms, while assuming certain business risks and subjecting itself to regulatory oversight. Regulators approve rates in accordance with well-established principles and methodologies. Ratemaking principles are grounded in constitutional law as affirmed throughout a rich history of Supreme Court cases. Many core standards of review are aimed at infrastructure investments, which must be found to be both “used and useful” to ratepayers and “prudent” based on knowable conditions. A “certificate of public convenience and necessity” may be used to establish need, but it does not ensure cost recovery. Utilities must operate with “reasonable economies.” Rates of return must be compensatory but commensurate with risk. Both rates charged and returns allowed must be “just and reasonable” and nondiscriminatory. Importantly, profits to utilities are authorized but not guaranteed, and returns cannot place “unjust burdens” on ratepayers. Full-cost accounting is emphasized, as cost-based rates encourage efficiency in production as well as in consumption and the allocation of society’s resources. Full-cost pricing reflects a matching principle whereby “burdens follow benefits” (and vice versa). Costs are assigned or allocated to cost causers according to allocation rules and not knowingly or unknowingly shifted to others (transfers or subsidies). When necessary and justified by policy goals, subsidies should be exposed and the economic implications understood. The traditional form of economic regulation implemented in the United States, known as the rate base/rate-of-return (RB/ROR) method, applies regulatory judgment to approximate an “efficient” (market-based) price while allowing a “fair” return (figure 4.3). The label cost-plus ratemaking (suggesting recovery of cost plus a return) is misleading because it understates the requisite role of the regulator in cost evaluation. Determining and allocating costs is the essence of ratemaking. A “test year” (or “rate year”) is used to establish the cost of service and base rates. Under this method, utilities are strongly motivated to invest and to include all known and measurable costs in the estimation of revenue requirements for the prospective period for which rates are set. Ratemaking involves three key steps and associated goals. First is the determination of the utility’s prudently incurred revenue requirements, or total budget, for a given test year, comporting with the goal of full-cost pricing. Second, cost allocation links system costs to usage, consistent with the principles of cost causation and “due and undue” discrimination. The third step, rate design, involves constructing revenue-neutral tariffs to recover the full cost of service from customers through rates and charges, comporting with the goal of just and reasonable rates.

economic regulation of utility infrastructure

103

Figure 4.3 Compensatory Pricing for Utility Monopolies

Cost and price

Monopoly price

Fair-return price Socially optimal price Marginal revenue

Average total cost Marginal cost

Demand

Quantity

Revenue requirements are determined as follows (and summarized in table 4.3):

Figure 4.3

RR ! r(RB) " O&M "D"T Lincoln_Ingram_Infrastructure where

! annualized revenue requirements ! authorized rate of return ! rate base (original cost of utility plant in service, net of accumulated depreciation and adjustments) O&M ! operation and maintenance expense D ! depreciation expense T ! taxes RR r RB

Regulation also benefits from uniform systems of accounts and established methods of financial analysis. Utility ratepayers must cover the capital and operating expenses of the utility, including operation and maintenance, depreciation, and taxes (income, property, and other). The depreciation expense compensates the utility for “using up” assets. Depreciation rates are ideally matched to service life, and the associated expense provides cash flow that utilities can use for additional infrastructure investment. Reinvestment of cash is not actually required, but the utility is also expected to be a going concern and meet its obligation to serve.

104

Janice A. Beecher

Table 4.3 Allocation of Utility Revenue Requirements Under Economic Regulation Variable operating costs

Revenue requirements

Fixed operating costsa

Labor Energy

Operations

Other inputs and variable costs Taxes, insurance, contracts, and other fixed costs

Capital recovery

Depreciation Cost of capital

Interest on debt Return on equity

Above the line: Ratepayers cover the prudent cost of service Below the line: Ratepayers compensate debt holders and shareholders (net of disallowances)

aIn

a

the short run, many operating costs are obligatory and thus essentially fixed. In the short run, many operating costs are obligatory and thus essentially fixed.

All utilities invest in and manage infrastructure assets for a public purpose. Publicly owned utilities fund infrastructure through debt instruments (bonds); privately owned utilities utilize a combination of debt and equity (typically about evenly divided). The return on equity is, literally, the price paid for shareholder investment in infrastructure. Setting the rate of return can be difficult and controversial. Investors receiveTable a return 4.3 of (depreciation) and return on (profit) their investment. Investors expect returns that are nonconfiscatory and compensatory Lincoln_Ingram_Infrastructure relative to comparable risk, imploring that they must be sufficient to earn positive credit ratings and attract capital. Arguably, the only risk utilities face is regulatory risk, as every regulatory treatment effectively shifts risks between utility investors and ratepayers (with the potential to affect the cost of capital). Upon a finding of imprudence, regulators will impute a reasonable cost and send the disallowed excess “below the line” (deducted from profits). Revenue requirements determine the size of the pie; rate design slices it up. Rate design can involve science, art, and politics, as compensatory rates are easier to fathom than “just” rates. Rate options can be evaluated according to various criteria, including revenue recovery, efficiency, and equity (see Bonbright, Danielsen, and Kamerschen 1988). Alternative rate structures can recover revenue requirements. A combination of fixed and variable charges is used, but they may not match fixed and variable costs. Conventional ratemaking typically involves averaging costs by customer class. Costs are allocated to customer classes

economic regulation of utility infrastructure

105

based on usage patterns. Pricing can discriminate among users only based on costs. Controversy arises from departure from cost of service based on social criteria. “Socializing costs” involves spreading costs widely; “social ratemaking” includes special-purpose rates, such as those designed to promote economic development or affordability. Rate changes and design are consequential in that a change in prices can induce a change in usage (based on price elasticity); dramatic short-term reductions in usage are induced by “rate shock.”

Regulation and Incentives Regulators do not manage public utilities. They lack the expertise and information to do so, and it is not their purpose. Regulators always have and always will concern themselves with three basic matters: standards, accountability, and incentives. First, regulators set basic standards in the form of minimal requirements, limits on certain behaviors, or simply the rules of engagement. Second, regulators hold the regulated entity accountable through various reporting, auditing, and review processes. Third, regulators provide incentives (or remove disincentives) for performance, either directly or by shaping the circumstances and opportunities affecting the utility. The purpose of all forms of regulation is to provide incentives for desired performance. Although the traditional model often is juxtaposed against incentive regulation, all economic regulation is incentive regulation.9 Like any organization, a public utility will respond to sufficient incentives and disincentives.10 Despite modern rhetoric about multiple “bottom lines,” profit remains the predominant motive. Criticism of the RB/ROR model rests squarely on the dual concerns that it provides too much incentive to invest and too little incentive for innovation and efficiency: Traditional cost-of-service rates do not promote innovation and efficiency by regulated firms. Simply stated, cost-of-service rates are based on a “snapshot” of a firm’s total cost of providing service plus a “fair” profit. Once the regulator sets rates, there is no incentive for a company to try and reduce costs or operate more efficiently since in the long run they could not keep any additional profits in excess of the allowed return. In fact, cost-of-service rates can have the perverse effect of providing incentives for a firm to operate less efficiently. For example, since the rate of return is based on the cost of capital, firms could increase revenues by

9. Attributed to economist and regulator Alfred E. Kahn. Economic regulation is more properly characterized as incentive-oriented than “command-and-control” policy. See also Lyon (1994). 10. Energy utility executive John Rowe once remarked, “The rat must smell the cheese” (Edison Electric Institute 1989).

106

Janice A. Beecher

increasing their invested capital. Also, most day-to-day operating costs, such as the cost of gas for a LDC [local distribution company], can be passed straight through to customers, providing no incentive for firms to seek cheaper gas supplies. (Energy Information Administration 1997, 116, emphasis added)

This “cost-plus” perception of regulation reflects a view that regulation is largely ineffective in substituting for market forces, which raises the question of whether the problem rests with the regulatory model or the quality of implementation. Theoretically and empirically, economic regulation is not necessarily antithetical to efficiency and innovation, or even commercialization (see Porter and Stern 2011). Throughout their long history, regulated utilities and their organizations achieved considerable innovation, as evidenced by Bell Laboratory patents and prizes. Competition among firms following restructuring may explain the apparent shift from collaborative to “sponsored” research. Regulation can motivate utilities through three loosely hierarchical but overlapping tools (figure 4.4): regulatory lag (primarily for cost control), prudence reviews (primarily for efficiency), and incentive rates of return (primarily for innovation). Though not well understood in this regard, and thus a source of consternation, regulatory lag motivates utility performance by design and is embedded within the social compact and prevailing ratemaking methodology (Bailey 1974; Pollock 2010). “The primary incentive for utilities to control their operating costs comes from the existence of regulatory lag . . . the setting of a price that is fixed until the next rate case” (McDermott, Peterson, and Hemphill 2006, 19).11 Indeed, “what may be viewed as an inherent defect of the systems turns out to be one of its strengths” (Wein 1968, 63).12 For practical purposes, regulatory lag is the delay between a change in the cost of service (up or down) and a change in authorized rates for service. Alternative conceptions of lag include the time period between when an unregulated firm and a regulated firm could make a defensive price adjustment in response to a cost increase (economic); the time period between rate filing and rate authorization (procedural); the time associated with test-year or cost-adjustment policies (policy); the time associated with decision-making delays (bureaucratic); or the time period between rate-case decisions (systematic). Lag is affected by the timing of a filing, suspension period, statutory deadlines, agency workload and resources, and the quality of the submission, including the robustness of evidence (e.g., cost studies and load forecasts). Lag is directly countered by “automatic” cost-adjustment mechanisms that should be limited to

11. See also Kahn (1971). 12. Wein (1968) further notes that nonregulated firms also use lagging responsiveness to competitive advantage.

economic regulation of utility infrastructure

107

Figure 4.4 Regulatory Tools and Incentives

Incentive returns

Prudence reviews

Regulatory lag

Innovation

Efficiency

Cost control

costs that are substantial, recurring, volatile, and uncontrollable and still subject to regulatory scrutiny and reconciliation. These mechanisms, once confined to variable operating costs, have in recent years been extended to capital costs in the form of system improvement surcharges (rationalized on the basis of safety and reliability). Their use in this regard justifies regulatory review of capital asset planning and management as well as net effects on revenue requirements. Regulatory lag is strongly associated Figure 4.4with regulatory risk. Importantly, lag cuts both ways. AlthoughLincoln_Ingram_Infrastructure typically portrayed pejoratively in terms of downside risk for utilities, lag also presents upside opportunity. Moreover, while uncertain and protracted cost recovery shifts risk to investors, certain and expedient cost recovery shifts risks to customers. The effects of lag are related to both utility performance and cost inflation, and the range of possibilities widens with time (figure 4.5). Incentives tend to be weak when costs and prices are stable (favoring the status quo). When costs are rising, however, incentives for cost control are stronger. High-performing utilities are more likely than low-performing utilities to realize authorized returns (see Wein 1968). Authorized rates will be ratcheted upward if costs rise, and reset downward if costs fall. Utilities lament, of course, that regulators seize the rewards of performance gains made between cases, but the point is to sustain incentives. In sum, regulatory lag provides potentially powerful (albeit inexact and somewhat clunky) motivation for both efficiency and innovation. It is not suggested here that regulation should be intentionally process inefficient, but rather that lag be recognized for incentive effects. Indeed, regulators might even seek “optimal timing” to encourage innovation and share rewards between utilities and ratepayers (Bailey 1974). Prudence reviews are a more proactive regulatory tool. The well-studied Averch-Johnson, or “AJ,” effect finds that regulated utilities are motivated to invest in the regulated rate base: “For each additional unit of capital input, the

108

Janice A. Beecher

Figure 4.5 Regulatory Lag and Incentives An increasingly inefficient utility might not achieve its allowed return (higher cost, lower return); it will file again. Loss

Cost at rate case 1

Profit

Cost at rate case 2

An increasingly efficient utility might exceed its allowed return (lower cost, higher return); it will stay out.

firm is permitted to earn a profit (equal to the difference between the market cost of capital and rate of return allowed by the regulatory agency) that it would otherwise have to forego” (1962, 1953).13 Researchers have estimated the level of overcapitalization at 20 to 30 percent for electricity generation (Lavado and Figure 4.5 for, restructuring. As other apHua 2004) prior to, and providing an impetus proved expenses are generallyLincoln_Ingram_Infrastructure passed along to customers, profit is made through capital investment combined with favorable returns. The value of the rate base, and thus the potential for profit, is a function of the scale of investment and the pace of replacement. Left to their devices, utilities will favor capital-intensive investments and accelerated depreciation. To the extent that operating expenses are essentially passed along to ratepayers, they have weak incentives for cost control as well as technical innovation. Absent a mandate or incentive mechanism, they also are unlikely to spend on demand management that would depress sales and thwart long-term investment opportunities. Regulators must counter these tendencies and encourage not just any investment but economically efficient and prudent investment. Regulation may induce more or less spending, depending on needs and circumstances. Prudence reviews and audits help guard against preventable excess, waste, and cost inflation. However, they may also encourage investment in conventional technologies over “promising but risky” innovation (Lyon 1995). The cumulative cost and technical complexity associated with infrastructure replacement argues for renewed attention to ensuring prudence. Somewhat ironically, given interest in alternative models, RB/ROR may prove to be an attractive policy instrument precisely because of its potential to harness and manage competing incentives.

13. Publicly owned utilities, by comparison, may be more prone to deferral and underinvestment, in order to avoid rate increases to constituent-ratepayers.

economic regulation of utility infrastructure

109

The authorized rate of return is a third critical regulatory tool. Returns on equity relate directly to perceptions about regulatory climate and regulatory uncertainty, factoring into ratings of both companies and regulators. In the regulatory process, returns are authorized as an effective earnings cap, ceiling, or band. The utility is entitled to the opportunity to earn a fair return, but returns are not guaranteed. In between rate cases, returns can be higher than authorized. In actuality, the authorized return may be very difficult to achieve and very often is unrealized. Returns work similarly to and in tandem with lag to force utilities to reach for profit, much as competitive firms must do. Regulators may consider performance incentives—premiums (carrots) and penalties (sticks)—in setting returns. As long as they are just (neither excessive nor confiscatory), regulators can deliberately use rates of return to reward or punish utilities based on performance. Disallowances and penalties lower effective returns and send a signal about performance expectations to all regulated utilities. A utility that fails to meet performance requirements might also suffer a lower allowed return on equity. Though perhaps less common, incentive returns can be used to reward extraordinary efficiency and innovation. Like other subsidies, returns above the cost of capital are redistributive; they should be well justified by social goals, used sparingly, and limited in scope and duration. A utility, for instance, might be rewarded for investing in research and development that leads to a transformative technological or process improvement.14 Capitalization with returns has been considered for goal-oriented operating expenses (such as those associated with efficiency programs), but enhanced or bonus returns are typically conceived as a means to induce infrastructure spending. A contemporary example is the comparatively generous incentive returns, or “adders,” provided by the FERC for private sector expansion of the transmission network, pursuant to the 2005 Energy Policy Act. Electricity market restructuring resulted both in the loss of vertical economies and in inadequate transmission investment (Brennan 2006). Transmission investment continues to escalate (Pfeifenberger and Hou 2011), but it is unclear whether incentive returns are necessary or efficient (Lyon 2007). Any unchecked investment incentive runs the risk of overinvestment. Another frequent complaint is that profit premiums are not explicitly tied to technical innovation.

Alternative Regulatory Models Regulatory scholars long have been cognizant of the incentives and disincentives attached to the traditional regulatory model (Trebing 1963). A coherent system of standards, accountability, and incentives (as compared to micromanagement

14. Conventionally, research and development costs are typically borne by shareholders, who benefit from performance gains due to regulatory lag.

110

Janice A. Beecher

Table 4.4 Select Alternatives to Ratebase/Rate-of-Return Regulation Incentive-based methods

Revenue-assurance methods

Structural methods

Price-cap regulation (PCR) Performance-based regulation (PBR) Profit sharing Cost indexing Revenue decoupling Formula rate plans Contract-based regulation Structured competition Deregulation

of utilities) is the key to effective regulation. Under the traditional compact, regulators are responsible for ensuring performance. Incentive-based and revenueassurance methods supplant RB/ROR, while structural methods relinquish considerable responsibility to markets and other institutions (table 4.4). The interest in regulatory alternatives is motivated by concerns about regulatory efficacy as well as the cost of implementation. Certain methods are designed specifically to reduce regulatory caseload and expense. In many jurisdictions, utilities must support the cost of regulation through assessments and fees.15 At the federal level, expenditures on economic regulation are much less than expenditures for social regulation (Dudley and Warren 2012). Demonstrating benefits relative to costs, however, is a unique challenge for economic regulation given its quasi-judicial nature. As noted, incentive regulation focuses on performance results versus the particular processes by which they are achieved (that is, ends over means). Over the years, the basic regulatory model has been adapted to address a broad spectrum of social goals. Forward-looking test years, treatment of construction costs, normalization methods, cost-adjustment mechanisms or trackers, special-purpose surcharges, and rate-design innovations are examples. Many of these methods hasten compensation and forestall full rate cases but are suggestive of problematic “single-issue” or piecemeal ratemaking that emphasizes cost recovery over cost minimization (Pollock 2010). While adaptive techniques seek to modify and improve the traditional process, some alternatives seek more radical change. The several variations on the theme of incentive regulation are designed to promote efficiency and innovation by utilities and in the regulatory process itself.

15. In 2011, the American Water Works Company (2012), which serves 3.1 million customers in 16 states, incurred regulatory expenses of $7.4 million, amounting to about $2.40 per customer annually.

economic regulation of utility infrastructure

111

U.S. regulators have tried selective alternatives in connection with restructuring, but other nation-states have become relevant experimental laboratories for broader market structure reforms (see Florio 2007; Jamison 2009). The methods available are not necessarily mutually exclusive. PRICE-CAP REGULATION

Price-cap regulation (PCR) is the leading methodological alternative to RB/ROR (see King 1998). PCR and its variants (such as revenue caps) are referred to as incentive ratemaking because they are designed and advocated specifically to address the perceived incentive deficiencies of RB/ROR. Price-cap regulation requires utilities to price basic services at prices no higher than specified levels for a specified period of time, thus motivating utilities to reduce costs in order to realize higher and thus more “flexible” returns. From a theoretical standpoint, price caps provide incentives for both cost control and price efficiency (Vogelsang 2002) and “can positively affect a utility’s long-term performance” (Costello 2011). Pragmatically, multiyear price caps formalize regulatory lag and can reduce the frequency and cost of rate cases. According to the method, as originally applied, annual price caps are set for each regulated company based on the retail price index, plus an additional “K” factor representing productivity expectations, as follows (Office of Water Services 1994): PC ! Price level " RPI " K where RPI is inflation and K is a composite of anticipated gains in efficiency (–X) and expenditures for service quality improvement (#Q). The K factor, which can be reset during periodic reviews (typically every five years), takes into account the investments needed to meet applicable quality standards as well as offsets for anticipated productivity savings. British regulators use price caps in conjunction with performance benchmarking (discussed below) as well as licensure and competition policies. Price caps were the preferred method adopted by U.K. regulators following privatization of their utility industries. In the United States beginning in the 1980s, price caps were applied in the telecommunications sector in the transition to competition. Price freezes, a form of capping, were used in electricity restructuring. Although the theory behind price-cap regulation is sound, it is not without implementation challenges. Analytical and monitoring effort can be significant. Regulators still must determine the initial base for rates and select an appropriate cost index. Given the chance for error, PCR may be “overpowered” for some newly privatized industries (Wolak n.d.). PCR requires periodic regulatory reviews to realign prices with the cost of service and to ensure that price competition is effective (Loube 1995). Price-cap regulation may be prone to accounting manipulation that would be revealed only through auditing. In terms

112

Janice A. Beecher

of incentives, the chief concern is that profit motive will result in cost avoidance and degradation of service quality, which is why PCR is often paired with performance-based regulation. The potential for excess earnings is another concern, but it can be ameliorated somewhat by profit sharing. Researchers have shown potential benefits of price-cap regulation in terms of price levels (Mathios and Rogers 1989), cost control (Clemenz 1991), and productivity (Seo and Shin 2011). However, PCR also tends to raise the cost of capital and shift risk from consumers to investors (Alexander and Irwin 1996), as well as invite significant regulatory risk in terms of the probability of very high or negative profits (Wolak n.d.). On the whole, the empirical evidence on how PCR affects investment, efficiency, and innovation is conditional or inconclusive (Goel 2000; Grobman and Carey 2001; Roques and Savva 2009). Like other methodologies, it is difficult to distinguish a firm’s efficiency improvements from broader influences on performance, including market forces. RB/ROR and PCR appear to require comparable regulatory effort and resources. When practiced well, these methods seem to work fairly similarly and will likely yield similar results, although both tend to presume natural monopoly (Liston 1993). PCR seems to work well when costs are declining (Sappington and Weisman 2010), as technological advances might facilitate. Given very different sector profiles, the experience with PCR may not be easily transferable: While it is clear that price regulation is superior for telecoms where it may only be needed in the transition to deregulation, it is less clear that permanent price regulation with periodic reviews is superior for core network monopolies like water, gas and electric distribution, balancing the better incentives of price regulation against the lower perceived investor risk and cost of capital of rate-of-return regulation. (Newbery 1997, 8–9)

In 2010, regulators in the United Kingdom introduced a new performancebased regulatory model (RIIO) that includes eight-year price-control periods and incentive returns aimed at attracting infrastructure investment (Office of the Gas and Electricity Markets 2010).16 PERFORMANCE-BASED REGULATION

The various renditions of performance-based regulation (PBR) shift attention from costs and inputs to performance and outcomes. PBR is used to varying degrees in all regulatory regimes. Performance assessment is used in conjunction with RB/ROR, PCR, and regulatory methods that rely on indexing to ensure that profit incentives do not have deleterious effects. Under the traditional model, prudence reviews inform cost disallowances, but not necessarily as part of an overall system of performance regulation. As part of an alternative scheme, PBR

16. RIIO: revenues = incentives + innovation + outputs.

economic regulation of utility infrastructure

113

can also take the form of flexible returns or even incentive compensation for achievement of specified performance targets. PBR is also known as yardstick or benchmark regulation or comparative competition. Utility companies use financial and nonfinancial benchmarks to track performance internally and comparatively. Performance metrics can be developed across various aspects of utility operations, including productivity, service quality, and reliability; customer service and satisfaction; worker safety; employee compensation; load management; losses and loss management (wires and pipes); recovery from outages; and rates charged for services. In some conceptions of PBR, utilities operating within the range of specified benchmarks might be exempt from certain forms of oversight (also known as a “safe harbor” approach). PBR can be used as a deliberate incentive tool. Performance comparison introduces a surrogate form of competition intended to motivate utilities to control costs, make improvements, and adopt innovation. Publicizing performance ratings can be used to educate consumers and pressure utilities to address deficits (Kingdom and Jagannathan 2001). Regulators also can attach incentives to measurable performance standards or goals. Examples include construction-cost or demand-management targets that trigger cost recovery or bonus returns. Like PCR, PBR is intrinsically related to restructuring (Biewald et al. 1997). The usefulness of performance benchmarking depends entirely on the development of valid and reliable measures of industry-specific and generally accepted indicators, so that comparisons can be meaningful. Comparison over time addresses methodological limitations by focusing on trends rather than snapshots. Although the theory of performance regulation emphasizes an orientation toward results (versus means), a relevant risk is the potential for micromanaging utilities, which runs contrary to traditional regulatory principles and precedents. Moreover, the use of incentive returns may be inefficient and largely unnecessary. Kihm (1991) suggested that rewarding managers and employees for performance might be as effective but less costly to ratepayers. Executive compensation and its relationship to performance is generally left to shareholders, although the subject occasionally piques the interest of regulators. By around 2000, the states seemed to be turning from targeted incentive plans to broad-based PBR (Sappington et al. 2001) and to other regulatory alternatives. PROFIT SHARING

A system for sharing profits or earnings encourages efficiency and innovation by allowing the utility to apportion both risks and rewards with customers. Traditional methods are used to establish baseline revenue requirements and allowed returns (either a level or earnings band). Profits may be shared on a 50-50 basis or another formula, including a sliding scale linked to performance criteria or policy goals. Profit sharing helps address the potential disincentive to control costs under RB/ROR regulation. Profit sharing can also help overcome disincentives for

114

Janice A. Beecher

innovation associated with prudence reviews (Lyon 1995). A profit-sharing approach can increase flexibility in terms of expanding utility service offerings. Performance assessment can help ensure that less profitable core functions do not suffer from a shift in focus to more profitable pursuits. Profit sharing can present some cost accounting challenges, although it preserves opportunity for regulatory review. A salient policy issue is whether ratepayers should bear a symmetrical risk of losses when utility ventures fail. COST INDEXING

Cost or rate indexing is a method for adjusting rates according to changes in a standard inflation index, such as consumer or producer prices. A key problem with general indexing is that price inflation becomes self-fulfilling. Indexing also can be used to adjust rates for particular categories of costs (such as energy costs). Once a baseline revenue requirement is established, indexing vastly simplifies the process of adjusting rates. Indexing can be used when setting rates for multiple years to guard against overearning or underearning by the utility due to fluctuations in the overall economy. Indexing motivates efficiency because utilities that are able to keep actual costs below indexed costs are allowed to retain the savings; conversely, utilities will absorb cost overruns. The purpose of rate indexing may differ by the size of the system. For small systems, indexing ensures that utility revenues will keep pace with inflation. However, prudent costs actually may exceed the rate of inflation. Indexing does not promote additional investments and expenditures that may be necessary for the proper maintenance of the system. For larger systems, cost or rate indexing with an inflation adjustment also can be used in conjunction with performance measures. REVENUE DECOUPLING

Though rationales vary historically and by utility sector, certain revenue assurance mechanisms—namely revenue caps and revenue decoupling—seek to address a particular set of incentive issues associated with lost revenues due to declining demand. Like price caps, revenue caps rely on an established index but are more appropriately used when costs do not vary appreciably with sales (Jamison n.d.). Decoupling is a contemporary form that detaches sales from revenues and profits to address perceived problems with the utility’s presumed “incentive to sell.” Aligning fixed and variable costs with fixed and variable prices is a basic form of decoupling; it provides revenue stability but raises concerns about affordability. Full decoupling establishes a revenue cap and can be suggestive of fixed-fee pricing. Reductions in sales are offset by price increase to maintain overall or per-customer revenues (revenue neutrality); revenue stability for the utility is achieved at the expense of rate stability for customers. Although decoupling compensates utilities and neutralizes sales incentives, it does not actually provide positive incentives to reduce sales through conserva-

economic regulation of utility infrastructure

115

tion programs. Like other methods, decoupling may be used with performance incentives. Decoupling presents a number of issues. Decoupling was originally offered as a means of compensating utilities for purposive or mandated demand repression, presuming both that special incentives are needed and that reductions in demand can be attributed to utility programs. In reality, utilities “enjoy” higher sales but can do little to actualize them except underprice. Lost investment opportunity (versus lost sales) is the more intractable dilemma for investors. Decoupling contrasts with how the marketplace is supposed to reflect utility conduct and consumer value (Brennan 2010; Costello 1996). In particular, decoupling sends weak price signals about long-run capacity costs. By disconnecting costs and prices, decoupling thus undermines the principle of consumer sovereignty. Decoupling also conflicts with preferences for variable pricing, incentives for utility performance, and conventions for risk allocation under the social compact. Many of the concerns used to rationalize decoupling actually can be addressed through traditional ratemaking tools. FORMULA RATE PLANS

Also promoted as a “rate stabilization” method, a formula rate plan (FRP) is an alternative ratemaking method “in which the utility adjusts its base rate outside of a general rate case, usually annually, based on an actual or projected rate of return (ROR) on rate base or equity that falls outside some commission-defined band” (Costello 2010, 8). Utilities favor FRPs in light of rate-case expense, regulatory lag, rising costs, and falling earnings.17 FRPs consider both costs and revenues and thus can be used in place of more conventional cost-adjustment mechanisms, as well as revenue decoupling. Regulators still must establish base rates, as well as cost-allocation policies, and rate design. FRPs should reduce the frequency of comprehensive rate cases, but performance incentives may be weakened, and prudence reviews remain essential. Under some plans, rates are adjusted before regulatory review, and disallowances result in customer refunds. The FERC applies formula rates to interstate pipeline and transmission providers. Although conventional ratemaking is formulaic, FRPs raise several significant regulatory policy concerns (Costello 2010, 2011). An FRP can be rationalized only when traditional methods fail and the plan can be shown to serve the public interest and result in just and reasonable rates. FRPs should be conditioned on meeting performance standards, with penalties for noncompliance. Authorized returns should be adjusted for reduced risk and not guaranteed. The earnings band should be wide enough to provide incentives for cost control, and rate adjustments should keep targeted returns beyond the boundaries of the band in

17. ComEd (Exelon) argues, “Our nearly 100-year-old process for determining rates is out of step with modern realities” because cost recovery is unpredictable and not timely, undermining state competitiveness (n.d.).

116

Janice A. Beecher

order to provide continued incentives and risk sharing. In any case, cost recovery should not be automated or implied as such; rate adjustments should still be subject to prudence reviews, and general rate cases should be conducted periodically to review earnings bands, cost allocation, and rate design. CONTRACT-BASED REGULATION

When utilities are publicly owned, contract-based regulation becomes an option. With variations in scope, privatization arrangements are used widely in the water sector for capital projects and operations, domestically and globally. Advocates contend that public-private partnerships through various available contractual vehicles offer a number of taxation, financing, efficiency, and other advantages. Long-term contractual agreements are comparable to concessions and charters utilized in Europe. In the United States, the contract model preserves municipal ownership and circumvents economic regulation because of the limited state jurisdiction in this area, which some view as an advantage of contracting. Local governments that engage contractors are responsible for the economic regulatory function, including cost reviews and rate setting. Municipal contracts are used to establish or maintain governmental control while promoting a degree of competition among alternative vendors. However, privatization cannot be equated with competition. The contest for contracts tends to be oligopolistic and short-lived, particularly for larger systems; long terms of engagement and no-bid renewals are monopolistic. The structural monopoly of the utility also remains intact. Thus, incentives for efficiency and innovation are weak and arguably weaker than those provided through regulation. Contracting inevitably raises principal-agency issues as well. Separating ownership from operation can lead to suboptimal performance and conflict over investment and expenditure decisions. Effective local regulation requires a sound contract vehicle with performance incentives, significant local oversight capacity, dispute resolution processes, appropriate risk allocation, and meaningful monitoring and enforcement mechanisms to ensure service quality and prevent abuses (see Marques and Berg 2010). STRUCTURED COMPETITION

Although more conservative than deregulation, structured competition still seeks to exploit competitive forces within imperfect markets. Structured markets depend in large part on the concept of contestability or competitive threat to motivate performance. Some markets or market segments may be contestable if technical, economic, and institutional barriers to entry are low. A degree of contestability can also be found among utilities of different ownership forms; a notable example is the ongoing consideration of public versus private ownership of water utilities. Artificial markets are structured through alternative regulatory methods that attempt to create and maintain a level playing field for competition. Potential regulatory tools include certification or licensure of providers and competitive

economic regulation of utility infrastructure

117

bidding or auctions to allocate market shares. Regulators also play new roles in monitoring markets and resolving disputes among providers. Structured competition has been used most extensively in wholesale electricity markets, with a significant level of institutional complexity. DEREGULATION

The economic regulation model is premised on the idea that economic regulation is necessary when markets fail, as is the case with traditional utility monopolies. The flip side implies that markets offer superior social controls and performance incentives. Deregulation places a high value and great reliance on customer choice and individual incentives as means of forcing economic discipline. Deregulation constitutes nonincremental or radical policy change. Over time, the United States essentially deregulated banking, transportation (trucking and airlines), cable television, and much of the telecommunications sector. The Interstate Commerce Commission and the Civil Aeronautics Board were terminated. Prior experience with deregulation has motivated interest in deregulating energy and, to a much lesser extent, water. Deregulation cannot be ideological or a matter of faith and is feasible and appropriate only when regulation was a mistake in the first place (Peltzman 2004) or when technological and other forces facilitate the emergence of markets. Competition must be workable and sufficiently robust to relax or eliminate safeguards; residual imperfections must be trivial or tolerable. Utility restructuring in the United States has seen mixed results. In some instances, it has included asset divestiture, which is largely irreversible. It also has included partial deregulation of market segments or “de-tariffing” only. A paradox of deregulation is the need for more regulatory capacity related to market standards, analytics, and oversight. Deregulation defers to reactive policy tools for checking market power, namely fair trade, consumer protection, and antitrust enforcement. Deregulation should achieve net benefits to society with acceptable costs, taking a full range of criteria into account. The biggest concern about deregulation is the denial of underlying market failures, including those related to social equity.

Evaluation and Conclusions Evaluating regulatory alternatives requires consideration of multiple criteria. The efficacy of any model rests on certain assumptions. Care must be taken to recognize how well the model fits with sector-specific characteristics, circumstances, performance goals, and, most important, the public interest. The influence of regulation on utility performance must also be considered within the context of other endogenous and exogenous factors, including managerial competencies, shareholder expectations, economic forces, credit and equity markets, environmental regulations, and so on. All forms of regulation and their implementation should be evaluated in terms of implications for performance incentives and risk allocation.

118

Janice A. Beecher

The advantages of traditional regulation are several, and some come from sheer experience. In the past as well as today, regulation encourages long-term infrastructure investment of scale and provides reasonable, if imperfect, performance incentives associated with efficiency and other social goals.18 Even advocates of regulation, however, recognize potential disadvantages. In particular, the traditional methodology can provide too much incentive for cost maximization and overinvestment (“gold-plating”) and too little incentive for cost control and innovation (“clawing back the savings”). Like all modes of social control, regulation may have reasonable theoretical foundations but is only as good as its stewards. The various alternatives also have advantages and disadvantages, depending on design and implementation. Performance metrics and expectations can be clarified. Incentives can be targeted to a variety of social purposes. Flexibility can allow utilities to respond more effectively to market forces. Risks and rewards can be more efficiently allocated. The administrative cost of regulation to utilities and the state can be reduced. Critics worry, however, that many regulatory alternatives are unproved in terms of long-term results and that unintended consequences are likely. Alternative methods can shift risks and introduce considerable uncertainty to utilities, ratepayers, and regulators. Many techniques require expanded regulatory capacities and resources and thus add to regulatory expense. Some models can complicate, diminish, or sacrifice oversight. Incentive schemes can lead to micromanagement by regulators, who also may be tempted to use rewards to favor a solution or technology, regardless of efficiency or impact. Driven by profit motives, utilities may take on excess risks or reap excess earnings. Incentives can become too much of a good thing. Perhaps the biggest risk of incentive-oriented regulation is that it will overcompensate utilities for doing what they are supposed to do—and what competitive markets would force them to do—in the first place. Regulatory lag should be recognized more explicitly for its role in imposing cost control, essentially by exploiting short-run profits that “provide the essential driving force for progress” (Bailey 1974, 286, drawing on Schumpeter 1934). The prudence standard suggests fair compensation for efficient performance, not extraordinary rewards. Regulation may be ripe for a “new prudence” centered on promoting optimal utility performance relative to rigorous but achievable standards. Given complex goals, trade-offs, and uncertainties, prudence today may call for flexible approaches to infrastructure investment, including incremental, modular, and decentralized technologies. While regulation can identify and motivate efficiency, innovation is a more elusive goal under traditional and alternative regulatory models. By its very nature, innovation cannot be either “standardized” or forced. Incentive returns might be used more deliberately toward this end, though ideally reserved for innovation above and beyond the norms of prudence.

18. Kerin (2012) makes the case that regulation should focus on efficiency as a means of serving the public interest and should leave other social objectives to other institutions.

economic regulation of utility infrastructure

119

Calls for a “new” regulatory paradigm are rampant today (Clifton, Lanthier, and Schroter 2011; Fox-Penner 2010; York and Kushler 2011). New business models for public utilities are presumed to require new models for their oversight. The original regulatory paradigm was premised on concerns about market failure, infrastructure investment needs, rising costs, social goals, distributional equity, and risk allocation; today is not so different (see Clifton, Lanthier, and Schroter 2011). Given immense challenges, more, not less, regulation arguably is needed. Incentives under traditional regulation are imperfect but in many respects more clear and consistent than those provided by alternative means. Surely, if monopoly structures remain and infrastructure investment is the prevailing social goal, a refined RB/ROR method can be a reasonable choice for ensuring that the public interest is well served. In the end, regulatory certainty may prove more important than regulatory perfection.

references Alexander, I., and T. Irwin. 1996. Price caps, rate-of-return regulation, and the cost of capital. Public Policy for the Private Sector. Note no. 87 (September). American Public Power Association. 2012. 2012–2013 annual directory and statistical report. Washington, DC: American Public Power Association. American Society of Civil Engineers. 2009. Infrastructure report card. Reston, VA: American Society of Civil Engineers. ———. 2011. Failure to act: The economic impact of current investment trends in electricity infrastructure. Reston, VA: American Society of Civil Engineers. American Water Works Association. 2012. Buried no longer: Confronting America’s water infrastructure challenge. Washington, DC: American Water Works Association. American Water Works Company. 2012. 2011 annual report. Voorhees, NJ. Averch, H., and L. Johnson. 1962. Behavior of the firm under regulatory constraint. American Economic Review 52:1052–1069. Bailey, E. E. 1974. Innovation and regulation. Journal of Public Economics 3(3):285–295. Beecher, J. A. 2008. The prudent regulator: Politics, independence, ethics, and the public interest. Energy Law Journal 29:577–614. ———. 2011. Why public utilities should ignore externalities. U.S. Association for Energy Economics Dialogue 19(1). Biewald, B., T. Wolf, P. Bradford, P. Chernick, S. Geller, and J. Oppenheim. 1997. Performance-based regulation in a restructured electric industry. Cambridge, MA: Synapse Energy Economics. Bonbright, J. C., A. L. Danielsen, and D. R. Kamerschen. 1988. Principles of public utility rates. Reston, VA: Public Utility Reports. Brennan, T. J. 2006. Alleged transmission inadequacy: Is restructuring the cure or the cause? Electricity Journal 19(4):42–51. ———. 2010. Decoupling in electric utilities. Journal of Regulatory Economics 38(1):49–69. Chupka, M. W., R. Earle, P. Fox-Penner, and R. Hledik. 2008. Transforming America’s power industry: The investment challenge, 2010–2030. Washington, DC: Edison Foundation.

120

Janice A. Beecher

Clemenz, G. 1991. Optimal price-cap regulation. Journal of Industrial Economics 39(4):391–408. Clifton, J., P. Lanthier, and H. Schroter. 2011. Regulating and deregulating the public utilities, 1830–2010. Business History 53(5):659–672. ComEd (Exelon). N.d. Formula rates: A new approach. Brochure. Costello, K. 1996. Revenue caps or price caps? Robust competition later means healthy choices now. Public Utilities Fortnightly (1 May). ———. 2010. Formula rate plans: Do they promote the public interest? National Regulatory Research Institute. ———. 2011. Some advice to regulators on formula rate plans. Electricity Journal 24(2):44–54. Digital Impact Group. 2010. The economic impact of digital exclusion. Philadelphia, PA: Econsult Corp. Dudley, S., and M. Warren. 2012. Growth in regulators’ budget slowed by fiscal stalemate: An analysis of the U.S. budget for fiscal years 2012 and 2013. St. Louis, MO: Weidenbaum Center at Washington University; Washington, DC: George Washington University Regulatory Studies Center. Edison Electric Institute. 1989. Washington letter. Washington, DC (15 September). Eggers, D. 2010. Impediments to achieving the vision. Paper presented at the Aspen Institute Energy Policy Forum, Aspen, CO (3 July). Energy Information Administration. 1997. Natural gas, 1996: Issues and trends. Washington, DC: Energy Information Administration. Federal Communications Commission. 2010a. The broadband availability gap. OBI Technical Paper No. 1. Washington, DC: Federal Communications Commission. ———. 2010b. National broadband plan: Connecting America. Washington, DC: Federal Communications Commission. ———. 2010c. Statistics of communications common carriers 2006/2007 edition. Washington, DC. Florio, M. 2007. Electricity prices as signals for the evaluation of reforms: An empirical analysis of four European countries. International Review of Applied Economics 21(1):1–27. Fox-Penner, P. S. 2010. Smart power. Washington, DC: Island Press. Goel, R. K. 2000. Price-cap regulation and uncertain technical change. Applied Economic Letters 7(11):739–742. Grobman, J. H., and J. M. Carey. 2001. Price caps and investment: Long-run effects in the electric generation industry. Energy Policy 29(7):545–552. INGAA Foundation. 2009. Natural gas pipeline and storage infrastructure projections through 2030. F-2009-04. Washington, DC: Interstate Natural Gas Association of America. Jamison, M. A. 2009. Towards new regulatory regimes in globalized infrastructure. In Internationalization of infrastructures, ed. J.-F. Auger, J. J. Bouma, and R. Kunneke, 257–273. Delft, The Netherlands: Delft University of Technology. ———. N.d. Regulation: Price cap and revenue cap. Gainesville, FL: Public Utility Research Center, University of Florida. Kahn, A. 1971. The economics of regulation: Principles and institutions. Cambridge, MA: MIT Press. Kerin, P. 2012. In whose interest? Network (publication of the Australian Competition and Consumer Commission) 43(March):1–7.

economic regulation of utility infrastructure

121

Kihm, S. 1991. Why utility stockholders don’t need financial incentives to support demand-side management. Electricity Journal 4(5):28–35. King, S. P. 1998. Principles of price cap regulation. In Infrastructure regulation and market reform: Principles and practice, ed. M. Arblaster and M. Jamison. Melbourne, Australia: Competition and Consumer Commission. Kingdom, W., and V. Jagannathan. 2001. Utility benchmarking: Public reporting of service performance. Washington, DC: World Bank. Kwoka, J. E. 2005. Electric power distribution: Economies of scale, mergers, and restructuring. Applied Economics 37(20):2373–2386. Lavado, R., and C. Hua. 2004. An empirical analysis of the Averch-Johnson Effect in electricity generation plants. Working Paper No. 7. Honolulu, HI: East-West Center. Liston, C. 1993. Price-cap versus rate-of-return regulation. Journal of Regulatory Economics 5:25–48. Loube, R. 1995. Price cap regulation: Problems and solutions. Land Economics 71(3):286–298. Lyon, T. P. 1994. Incentive regulation in theory and practice. Topics in Regulatory Economics and Policy Series 18:1–26. ———. 1995. Regulatory hindsight review and innovation by electric utilities. Journal of Regulatory Economics 7(3):233–254. ———. 2007. Why rate-of-return adders are unlikely to increase transmission investment. Electricity Journal 5:48–55. Marques, R. C., and S. Berg. 2010. Revisiting the strengths and limitations of regulatory contracts in infrastructure industries. Journal of Infrastructure Systems 16(4):334–342. Mathios, A. D., and R. P. Rogers. 1989. The impact of alternative forms of state regulation of AT&T on direct-dial, long-distance telephone rates. RAND Journal of Economics 20(3):437–453. McDermott, K. A., C. R. Peterson, and R. C. Hemphill. 2006. Critical issues in the regulation of electric utilities in Wisconsin. Thiensville: Wisconsin Policy Research Institute. Newbery, D. M. 1997. Rate-of-return regulation versus price regulation for public utilities. Department of Applied Economics, Cambridge University, U.K. (14 April). North American Electric Reliability Corporation. 2011. 2011 long-term reliability assessment. Atlanta, GA: NERC. Office of the Gas and Electricity Markets. 2010. RIIO: A new way to regulate energy networks. Factsheet 93. London. Office of Water Services. 1994. Future charges for water and sewerage services. Birmingham, U.K. Peltzman, S. 2004. Regulation and the natural progress of opulence. Washington, DC: AEI-Brookings Joint Center for Regulatory Studies. Pfeifenberger, J. P., and D. Hou. 2011. Employment and economic benefits of transmission infrastructure investment in the U.S. and Canada. Washington, DC: Working Group for Investment in Reliable and Economic Electric Systems. Pollock, J. 2010. Streamlined ratemaking: Recognizing challenges for consumers. Electricity Journal 23(9):7–12. Porter, M. E., and S. Stern. 2011. National innovative capacity. In Global competitiveness report, 2001–2002. Oxford, U.K.: Oxford University Press. Robinson, H. M. 1932. Public utilities and the people. Dallas, TX: B. Upshaw.

122

Janice A. Beecher

Roques, F. A., and N. Savva. 2009. Investment under uncertainty with price ceilings in oligopolies. Journal of Economic Dynamics and Control 33(2):507–524. Sappington, D. E. M., J. P. Pfeifenberger, P. Hanser, and G. N. Basheda. 2001. The state of performance-based regulation in the U.S. electric utility industry. Electricity Journal 14(8):71–79. Sappington, D. E. M., and D. L. Weisman. 2010. Price cap regulation: What have we learned from 25 years of experience in the telecommunications industry? Journal of Regulatory Economics 38(3):227–257. Schumpeter, J. A. 1934. The theory of economic development. Cambridge, MA: Harvard University Press. ———. 1942. Capitalism, socialism and democracy. New York: Harper. Seo, D., and J. Shin. 2011. The impact of incentive regulation on productivity in the US telecommunications industry: A stochastic frontier approach. Information Economics and Policy 23(1):3–11. Silverstein, A. 2011. Transmission 101. NCEP Transmission Technologies Workshop, Denver, CO (20–21 April). Trebing, H. M. 1963. Toward an incentive system of regulation. Public Utilities Fortnightly 72:22–39. U.S. Energy Information Administration. 2004. The basics of underground natural gas storage. Washington, DC: U.S. Department of Energy. U.S. Environmental Protection Agency. 2009a. EPA’s 2007 drinking water infrastructure needs survey and assessment. EPA 816-F-09-003. Washington, DC: Environmental Protection Agency, Office of Water. ———. 2009b. 2006 community water system survey. EPA 815-R-09-001. Washington, DC: Environmental Protection Agency, Office of Water. ———. 2010. Clean watersheds needs survey, 2008: Report to Congress. EPA-832-R10-002. Washington, DC: Environmental Protection Agency. Vogelsang, I. 2002. Incentive regulation and competition in public utility markets: A 20-year perspective. Journal of Regulatory Economics 22(1):5–27. Wein, H. 1968. Fair rate of return and incentives: Some general considerations. In Performance under regulation, ed. H. Trebing. East Lansing, MI: MSU Public Utility Studies. Wolak, F. A. N.d. Price-cap regulation and its use in newly privatized industries. www .stanford.edu/group/fwolak/cgi-bin/sites/default/files/files/Price-Cap%20Regulation %20and%20Its%20Use%20in%20Newly%20Privatized%20Industries_Wolak.pdf. Wolf, C. 1993. Markets or governments: Choosing between imperfect alternatives. Santa Monica, CA: Rand. York, D., and M. Kushler. 2011. The old model isn’t working: Creating the energy utility for the 21st century. Washington, DC: American Council for an Energy-Efficient Economy. Zeilig, N. 2011. Declining demand likely to continue beyond recession. AWWA Streamlines 3(20).

commentary Timothy J. Brennan In the true multidisciplinary spirit of the Lincoln Institute of Land Policy, Janice A. Beecher, a political scientist, has provided a superb review of the basic economics of infrastructure regulation. This leaves me, an economist, to add in some of the noneconomic factors that both motivate and influence the paths that infrastructure regulation can take. Beecher’s contribution serves well as a primer for the rationales, history, and methods of regulating the price of infrastructure services. The standard justification is that many infrastructure services are likely to be provided by monopolies because the high fixed costs associated with their installation—think water mains or local electricity distribution grids—will discourage more than one firm from entering the market. Because the public typically regards these services as essential—water and electricity again being excellent examples—such monopolists would be able to charge extremely high prices. Regulation can protect customers from exorbitant bills and prevent the reduction in purchases that comes about from such high prices. The traditional method for controlling prices is to set them at the average cost of service, just enough to allow the infrastructure provider to earn a reasonable return on its investments. As Beecher notes, because this “cost-of-service” ratemaking stifles incentives to cut costs, regulators have adopted price caps and other methods that break the link between prices and costs, protecting consumers while allowing firms to profit from more efficient operations. Moreover, regulators are also concerned with protecting incentives to invest. Once fixed infrastructure costs are sunk, a regulator could in principle allow the firm to charge just enough to cover operating costs. Investors who see this coming would refuse to provide the infrastructure. To ensure against this, U.S. law strongly guarantees a “fair opportunity” to earn a “just and reasonable return.”1 One way to add to Beecher’s survey is to examine the boundaries of regulation. Not all infrastructure services are regulated; cable television, broadband Internet, and computer operating systems come to mind. For television and broadband, we have some competition because two originally incompatible wire networks (cable and telephone) fortuitously evolved to become competitors. With computer software, costs and capabilities change far too fast for regulation to make useful price determinations. Another boundary issue is the regulation/competition line within infrastructure sectors. Leading examples include the breakup of AT&T into regulated local

1. A colleague from Australia, Darryl Biggar, observes that regulation similarly prevents the firm from taking advantage of investments that consumers might make, such as charging high electricity prices to those who have invested in electric heat.

123

124

Timothy J. Brennan

service and competitive equipment and long-distance markets, deregulation of wellhead natural gas prices, and restructuring of the electricity sector with competing generators delivering power through regulated transmission and distribution lines. The rationale was to reap the fruits of competition wherever possible, but doing so presents a number of complex governance problems. To preserve competition in the open markets, regulators have to limit or proscribe the regulated firm from operating in those markets, creating difficult trade-offs between the benefits of competition and the cost savings and improved coordination from continued integration. In addition, rather than setting relatively simple end-user prices, regulators have to determine access charges paid by competing buyers. When buyers are competitors, they care more about ensuring that no one gets a discount than about whether overall prices are low, leading regulators to care as much or more about preventing discrimination than about keeping prices low overall.2 Moreover, particularly in telecommunications, regulation can facilitate competition through mandatory, cost-based standardization and interconnection to ensure that all competitors can reap the benefits of being on the same network. As the process of introducing competition within sectors unfolded, sector regulators found their responsibilities overlapping with antitrust authorities. For example, electric utility mergers are reviewed by state regulators, the Federal Energy Regulatory Commission, and the Department of Justice’s Antitrust Division. This creates an institutional tension between regulators, who are inclined to find substitutes for markets, and antitrust enforcers, who are dedicated to letting markets work. In the United States, recent Supreme Court decisions have largely precluded antitrust enforcement in regulated sectors. With regard to regulatory governance more broadly, regulatory authority in the United States is divided between the national government and the states. For example, the federal government has authority over interstate transmission and bulk power markets, while state utility commissions determine rates for distributing electricity and decide whether and how to regulate retail electricity rates. Another governance issue is whether the government will regulate a private infrastructure provider or will provide the service itself. Generally, but not universally, the former is typical for telephone and electricity service, while the latter holds for water, mail, roads, and local mass transit. A simple explanation for this complex distinction is that for some infrastructure service, concerns beyond control of monopoly, such as universal service or free access, may warrant public provision. Regulation’s justifications go beyond market power and potential expropriation to include the admittedly open-ended “public interest.” One such consid-

2. Notably, the 1887 Interstate Commerce Act, the founding statute of U.S. federal regulation, was focused more on ensuring that each customer paid the same price than on the price itself.

commentary

125

eration involves promoting access to infrastructure services, including universal service obligations imposed on mail and telecommunications carriers, and setting limits on cutting off delinquent accounts for electricity, heat, and water. To maintain social equality in access, regulation includes assorted subsidies, such as equalizing mail rates for all distances, telephone rates in more expensive sparsely populated areas, and provisions to assist low-income households with energy purchases. A leading telecommunications concern is whether the public should have access rights to broadband and wireless services beyond those provided by the marketplace. But in recognizing the public interest, one also has to recognize political and economic arguments that the regulatory outcomes will tend to be biased toward the interests of the regulated firms and away from those of dispersed consumers, each of whom has little ability or interest in taking part in the process. A final set of issues facing infrastructure regulators has significant effects on land use. Electricity regulators are increasingly charged with addressing environmental concerns, particularly through policies designed to promote renewable generation, foster energy efficiency, and, counter to their historical mandate, discourage electricity use. Along with effects on the atmosphere, particularly from greenhouse gas emissions, these policies can have significant effects on land use. The list includes increased hydrofracturing to produce relatively clean natural gas, the growth of both onshore and offshore wind farms, and centralized solar generation mirror farms. To get energy from these sources to the public, extensive transmission and delivery investments, taking up more land, will be needed. On the flip side, reduced use of coal will reduce the need to take up land through mining. In either direction, infrastructure regulation affects land policy specifically, as well as the larger economy and society for which an efficient and equitable infrastructure is crucial.

5 The Unit Approach to the Taxation of Railroad and Public Utility Property Gary C. Cornia, David J. Crapo, and Lawrence C. Walters The Issue and the Context Administering a property tax system presents ongoing challenges under the best of circumstances. Whether it involves keeping taxpayer rolls up to date, pursuing tax evaders, or updating taxable property values appropriately, property tax administration requires substantial diligence and expertise. This chapter focuses on one particular area of property tax administration in the United States: the taxation of railroads, public utilities, and other multijurisdiction properties. The number of such taxpayers in a state is generally fairly small compared to the total number of households and businesses paying property taxes. However, due to their size, these companies often incur the largest property tax bills in any given state. Because of both size and complexity, these properties are generally valued by state agencies rather than by local assessors, and the state agencies use valuation methods that differ markedly from the methods employed at the local level. In the sections that follow, the differences in method will be described, along with a brief summary of the history and ongoing controversies surrounding these “centrally assessed” properties. Before turning to this more detailed discussion, however, it is useful to describe the context more fully. The companies involved are railroads and railcar companies, gas and electric utilities, telecommunications firms, pipelines, airlines, and other firms with real and personal property assets in multiple taxing jurisdictions, often including multiple states. Value added from these firms represented 3.8 percent of gross domestic product (GDP) in 2010. In some states the list 126

unit approach to the taxation of railroad and public utility property

127

includes natural resource extraction companies such as oil and gas wells, mining operations, and even forestry companies. The 1992 Census of Governments reports that 4.3 percent of the property that is taxable at the local level was centrally assessed (U.S. Census Bureau 1994). Throughout this chapter, examples will be drawn from the electric utility industry because of data availability, but the patterns are similar in the other industries involved. In terms of both dollars and expense ratios, property taxes are a significant expense factor for centrally assessed companies. Table 5.1 reports property taxes paid by a small sample of electric utility operating companies from around the country. Most of these companies have tangible property assets in multiple states and multiple jurisdictions within each state. The companies shown were selected for their geographic diversity and because they are fairly well known within their regions of operation. They are typical of electric utilities as well as other centrally assessed properties. These firms pay property taxes in the tens and hundreds of millions of dollars each year, and property taxes paid are a very significant share

Table 5.1 Property Taxes Paid by Selected Electric Utility Operating Companies, 2011 Electric Operating Company

Property Taxes Paid (2011)

Gross Operating Revenue (2011)

Alabama Power Co. Arizona Public Service Baltimore Gas & Electric Consolidated Edison Co. of New York Duke Energya Florida Power & Light NSTAR Electric Company Ohio Edison Oklahoma Gas & Electric Pacific Gas & Electric PacifiCorp Public Service Co. of Colorado Public Service Co. of New Hampshire Average

$96,223,127 143,413,037 110,405,130 1,316,787,670 239,835,694 291,208,688 103,447,567 68,057,825 61,996,604 271,956,136 116,433,706 90,096,179 53,441,107

$5,702,250,135 3,274,438,030 2,992,614,087 10,610,651,493 11,862,501,752 10,609,210,465 2,633,057,952 1,395,495,932 2,328,466,158 15,160,335,346 4,553,757,373 4,293,125,992 1,033,054,174

Includes Duke Energy Carolinas, Duke Energy Indiana, Duke Energy Kentucky, and Duke Energy Ohio. Source: FERC 2011 Form-1 reports and calculations by the authors.

a

Property Taxes Paid as Percentage of Gross Operating Revenue 1.70 4.40 3.70 12.40 2.00 2.70 3.90 4.90 2.70 1.80 2.60 2.10 5.20 3.90

128

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

of their respective total operating expenses. As a result, it is common for each company to have its own specialized tax staff, which monitors and manages property taxes, in addition to retaining outside legal staff, specialized appraisers, and other experts during any valuation appeal process. Comparing the percentages in table 5.1 to the property taxes assessed on residential property is potentially misleading. For example, property taxes as a percentage of gross rent can vary between 15 and 35 percent, depending on the market area. However, gross rent and gross operating revenue under the unit approach are not equivalent concepts. Gross rent is the rental income directly attributable to a property. Gross operating revenue is the total revenue received from all sources by an operating company. The equivalent concept for a residential property would be total household income, including imputed rental value and the value of household production. To illustrate, residential property taxes in Utah in 2010 represented 1.4 percent of statewide personal income, but was more than 50 percent of the U.S. Bureau of Economic Analysis estimate of net rental property income (including imputed rents for owner-occupied property) for that year. Centrally assessed property represents an important revenue source for local governments. Based on contribution to GDP, these industries represent about 4 percent of the private U.S. economy. Because they tend to be capital intensive, they often represent a large percentage of property value in many states. The 1992 Census of Governments reported that 38 states centrally assessed at least some properties. Among that group of states, centrally assessed companies represented nearly 5 percent of taxable value in county and municipal jurisdictions, but the variance was substantial. Table 5.2 reports the percentage of total locally taxable value that was centrally assessed for the 15 states with the highest percentages. It is clear from the table that utility, railroad, and other centrally assessed properties constitute a significant share of the property tax base in many states. National data for property assessments were last reported in the 1992 Census of Governments. Although more recent comprehensive valuation data on centrally assessed properties are not available, the importance of these properties for local governments is readily highlighted in table 5.3. The table summarizes the experience of seven states in 2011. The first data column of the table reports public utility and/or centrally assessed property values as a percentage of overall taxable value in the state. Because the statewide experience often masks the importance of these properties in the individual local government tax base, the same ratio was calculated for each county (or school district in the case of South Carolina) in the state. The percentage of total county property value that is centrally assessed in the county with the highest ratio is reported in the second data column, and the last column shows the percentage of counties (or school districts) with more than 20 percent of their property tax base in centrally assessed companies. The table shows clearly that even in states where centrally assessed property is not a large proportion of the overall property tax base, such

unit approach to the taxation of railroad and public utility property

129

Table 5.2 State Assessed Taxable Value as a Percentage of Total Locally Taxable Value, 1992 State

State Assessed Value as a Percentage of Total Locally Taxable Value

Wyoming Alaska Montana Utah Arizona Ohio South Carolina New Mexico Oklahoma Kansas Maryland Louisiana Georgia Alabama Mississippi

71.00 42.40 30.10 24.00 22.10 20.70 18.00 17.40 16.60 16.30 15.10 14.70 14.60 13.50 12.40

Source: U.S. Census (1994) and calculations by the authors.

Table 5.3 Public Utility/Centrally Assessed Property as a Percentage of Total Taxable Value, 2011 State

California Georgia Kansas Mississippi Oregon South Carolina (school districts) Utah

Public Utility/Centrally Assessed Property as Percentage of Total Taxable Property Value

Maximum Local Ratio Within the State (%)

Percentage of Counties with 20% or More of Their Property Tax Base Centrally Assessed

1.80 3.60 11.40 11.20 4.90 7.20 10.90

15.60 65.50 84.30 45.30 70.60 51.20 83.40

0 3.20 35.20 20.70 14.30 8.20 34.50

Source: Calculations by the authors from state tax agency annual reports.

130

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

as California, it is extremely important for selected local governments. And in many counties, these industries constitute a very large proportion of the local base. Because of the importance of these companies in the local tax base, how they are valued is a matter of significant concern to local government officials. Their concern is often expressed in terms of independent appeals of state valuations and active involvement in appeals filed by taxpayers. Should the property owners choose to appeal their value, local governments are faced with a dilemma. Their financial plans and programming are built around an assumed revenue stream tied to the state’s estimation of value. While the appeal is pending, a significant percentage of the revenue stream becomes uncertain and may need to be refunded to the taxpayer at some future date. If local governments proceed with their plans but the state loses the appeal, local governments may be forced to increase taxes on other taxpayers in order to refund taxes and/or fulfill obligations incurred under the assumed increase in tax revenue. Local governments could delay implementation until the appeal is settled, but the uncertainty could last for years. Hence, both the taxpayer and benefiting local governments have a strong and active interest in the valuation process for centrally assessed property. (See box 5.1.) To summarize the context, public utilities, railroads, and other complex industrial properties are frequently valued for property tax purposes by state agencies. The resulting property tax bills constitute a significant expense for these companies. At the same time, these centrally assessed properties make up an important share of the overall property tax base, and in a number of cases the share exceeds 50 percent of the local tax base. As a result, local governments pay close attention to the valuation methods used and the resulting taxable values. State tax administrators are thus pressured by taxpayers to lower values and by local governments to raise values. The courts are certainly no strangers to this tension. As discussed later in this chapter, the methods employed by state agencies charged with valuing centrally assessed properties differ markedly from the methods used by local assessors. The next section describes the logic employed in these assessments. A subsequent section provides a more detailed description of the appraisal methods used.

The Unit Approach In 1890, the Cincinnati, Lafayette and Chicago Railroad operated a branch line that ran from Templeton, Indiana, to Kankakee, Illinois, a distance of some 60 miles. The line crossed the Indiana-Illinois state boundary and passed through several Indiana and Illinois counties. Most of the Indiana track was in Benton County, but a small two-mile section ran through a corner of adjacent Newton

unit approach to the taxation of railroad and public utility property

131

Box 5.1 It may be helpful to consider a specific example. Beaver County, Utah, is a rural county with a population of between 6,000 and 6,500 people and has a little more than 2,000 residential properties. The county is proud to be the only region in Utah with geothermal plants that deliver electricity to the grid. In the 2009 tax year, centrally assessed electric assets constituted 15 percent of the property tax base in Beaver County. In 2010, new geothermal capacity entered the tax base and increased the value of centrally assessed electric utility property by a little more than $305 million, increasing the electric utility share of the tax base to nearly 42 percent. The impact on local government revenue was dramatic. In Beaver County, 69 percent of the property tax goes to the local school district, while 19 percent flows to the county government. The remainder is divided between special service districts (9 to 10 percent) and the cities and towns (3 to 4 percent). Obviously, the school district is the main beneficiary from taxes on the new electric capacity. District tax revenues per household increased by nearly 70 percent between 2009 and 2010. School property tax revenues increased from $4.6 million in the 2009–2010 school year to $7.1 million in the 2011–2012 budget. On the expenditure side, the school district launched a capital program and increased outlays from an annual average of about $430,000 between 2007–2008 and 2009–2010 to $11.5 million in the 2011–2012 budget. At the same time, state aid dropped from $7.6 million to $6.7 million. The impact on the taxpayers was equally large. The $305 million in additional taxable value resulted in $2.87 million in annual property taxes for the property owners. Under the methods employed by the Utah State Tax Commission in valuing electric utility property, the $305 million in value represents 100 percent of market value and is arrived at using a combination of the methods outlined below. Based on the Utah capitalization rate study for utilities in 2010 (Property Tax Division 2010) and assuming a 10 percent cost of debt, the annual Net Operating Income (NOI) necessary to arrive at a final value of $305 million was on the order of $26 million, and the resulting annual tax burden represented over 10.8 percent of NOI. Even if the cost of debt is widely different from that assumed here, the property tax burden in all likelihood exceeded 8 percent of NOI. The point is not to ask whether the Utah State Tax Commission was correct in their analysis. Our point is simply that the $2.87 million in property taxes paid by the owners of new electric capacity in Beaver County represents a substantial charge against NOI, and those owners are likely to consider carefully whether they agree that the tax is appropriate.

132

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

County before the line crossed into Illinois. There were no stations or other facilities in Newton County. As in other similar cases, property tax administrators in Indiana confronted the question of how to value the rail line. The perspective adopted by many states in answering this question relies on a crucial assumption about the nature of the property being valued. A classic report published by the National Association of Tax Administrators points to the example of a parcel of city land that includes an older but serviceable house. The report argues that if the land and house were sold separately, they would be worth much less than if sold as a single integrated unit. The authors go on to argue that the best indicator of value for the land and house is the sales price for the unit, without reference to how that value is divided between land and improvements. With regard to this example, the report concludes: A unit appraisal is superior to a summation appraisal in this case not only because it produces a result that is closer to the true value of the property as a whole but because it produces that result by resort to more reliable and more readily available evidences of value than those that would be used for a summation appraisal. (Chapman et al. 1954, 14)

It is worth noting that this example is still cited by some states in justifying a unit approach (e.g., California State Board of Equalization 2003). The point being made is that virtually all appraisals of real property involve the identification of a tangible bundle of land and improvements, which is defined as the unit to be valued. While this is true, it is also critical to recognize that the unit generally valued by local assessors differs fundamentally from the unit as defined by states in valuing railroads and public utilities. The point here is that both local assessors and state tax agencies define a unit to be valued, and for both the object is to value the unit that is most likely to be traded in the marketplace. Unit valuations of railroads and other public utilities attempt to value the combination of properties that is likely to sell in the market as a single operating unit. Local assessors most commonly consider the value of a parcel of land and associated permanent improvements without reference to the overall business activities of the occupants. State assessors, in contrast, begin with the business enterprise as the unit to be valued. Traditionally, professional appraisers have used three approaches to estimate market value, and all are grounded in attempts to replicate the reasoning of potential actors in the real estate market. This is the “willing buyer, willing seller” concept, which assumes two reasonably well-informed parties who wish to engage in a transaction but are not required to do so. Hence, one approach to the railroad valuation question would be to estimate the cost of rights-of-way, rails, railroad ties, and other permanent improvements. But it can be argued that this cost approach greatly understates the value of the railroad. Put another way, no seller would be willing to accept a price calculated in this manner, and knowledgeable buyers would be willing to pay considerably more. Both buyers

unit approach to the taxation of railroad and public utility property

133

and sellers would look at the income-generating potential of the line in negotiating a mutually agreeable price. The problem arises in considering the two miles of track in Newton County. Without terminals, stations, switching yards, or other facilities, the income potential of the track was zero. As a railroad, it was worthless. But at the same time, the 60-mile line from Templeton to Kankakee was also worth substantially less without the two miles of track in Newton County. Tax administrators argued that in order to value each section of track, the overall value of the railroad as a going concern first had to be estimated. This value was then distributed across the entire line, based on the argument that each rail section was essential to realizing the potential of the overall enterprise. Without the Newton County track, the remaining 58 miles of track and related facilities had very little value. By the same argument, each mile of track contributed equally to company value. This method of estimating market value came to be known as the unit (or unitary) approach (or rule). It is based on the argument that assembled and operating tangible assets are worth more in real-world markets than the sum of the individual unassembled parts. That claim lies at the root of the valuation methods employed by state agencies in valuing centrally assessed property. It is also a claim that taxpayers have resisted for well over 100 years. The next section summarizes some of that history.

History and Development of the Unit Approach to Value As suggested by the example above, the unit approach was first applied to railroads. The specific case mentioned was resolved by the U.S. Supreme Court in 1894 (Cleveland, Cincinnati, Chicago & St. Louis Railway Co. v. Backus, 154 U.S. 439 [1894]). State supreme courts had heard earlier cases in 1868 in Kentucky and 1871 in Kansas, and the U.S. Supreme Court had ruled on the validity of the unit approach as applied to railroads in 1875 (Taylor v. Secor, 92 U.S. 575 [1875]). Two points are at issue in these early cases. First was the validity of using the going concern value as the basis for valuing railroads. Second was how the portion of overall value should be determined for each taxing jurisdiction. Three statements taken from these early court cases summarize the judicial conclusion: Applegate v. Ernst, 66 Ky. 648, 650 (1868): “The law treats a railroad and all of its appurtenances as one entire thing.” Missouri River, Fort Scott & Gulf Railroad v. Morris, 7 Kan. 210, 222–23 (Kan. 1871): “A railroad is an entire thing and should be assessed as a whole. It would be almost as easy and as reasonable to divide a house or a locomotive into portions and assess each portion separately, as to divide a railroad into portions and assess each portion separately.” Taylor v. Secor, 92 U.S. 575, 608 (1875): “It may well be doubted whether any better mode of determining the value of that portion of the track

134

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

within any one county has been devised than to ascertain the value of the whole road, and apportion the value within the county by its relative length to the whole.” The earliest railroad cases considered the valuation of railroads within a state, but as noted in the Indiana-Illinois example, railroads soon expanded and crossed state lines. Given the constitutional prohibition on taxing interstate commerce, it was certain that the courts would be asked to rule on whether the unit approach violated the commerce clause. In 1894, the U.S. Supreme Court upheld the application of the unit approach in valuing interstate railroads, arguing that the unit approach was used to estimate value, and track mileage was then used to measure value in each jurisdiction, but the tax was only on the value within a given jurisdiction (Cleveland, Cincinnati, Chicago & St. Louis Railway Co. v. Backus). By the turn of the century, other multijurisdiction companies were emerging, and tax administrators began applying the unit approach to those entities as well. In 1896, the Supreme Court upheld the application of the unit approach to telegraph lines (Western Union Telegraph Co. v. Taggart, 163 U.S. 1, 18 [1896]) and, the following year, to express companies (Adams Express Co. v. Ohio State Auditor, 165 U.S. 194 [1897]). In the twentieth century, the unit approach was also extended to electric utilities, telecommunication, pipeline, and other capitalintensive firms that employ a system of interconnected assets. Today, 40 out of 50 states centrally assess at least some portion of real property, and 39 states employ the unit approach (Department of Taxation and Finance 2005). Because of the size of the taxpayers involved and the magnitude of the resulting property tax bills, appeals have been common, and an extensive body of case law has developed. Many of the fundamental concerns raised by taxpayers over the years have proved difficult to resolve completely. The Advisory Commission on Intergovernmental Relations (ACIR) published an assessment of property tax practices in 1962 and reissued the report in 1975. That chapter on central assessment practices begins with this statement: “Among the problems of property taxation one of the most controversial and perplexing is that of administering the ad valorem tax on railroads and other public utilities” (Advisory Commission on Intergovernmental Relations 1963, 147). To understand why controversy persists after more than 150 years of scrutiny and debate, we highlight four nagging issues that routinely resurface in the debate. DEFINING THE UNIT

The unitary valuation approach is typically applied to properties that operate across county and state boundaries and whose “value depends on the interrelation and operation of [all of the properties] as a unit. Many of the separate assets would be practically valueless without the rest of the system. Ten miles of telephone wire or one specially designed turbine would have a questionable value, other than as scrap, without the benefit of the rest of the system as a whole” (ITT

unit approach to the taxation of railroad and public utility property

135

World Communications, Inc. v. San Francisco, 37 Cal.3d 859, 210 Cal. Rptr. 226, 693 P.2d 811 [1985]). As technologies and business organizations evolve, questions frequently arise as to whether a particular type of business asset or group of operating properties should be subject to unitary assessment. For example, in the past, telephone companies were required to be assessed as a unit, but the properties of cable companies were not. Technologies have changed, so cable companies now provide many of the same telecommunications services that are provided by telephone companies, and jurisdictions have grappled with whether the properties of cable companies are now subject to unitary assessment: Comcast Corp. v. Department of Revenue, No. TC 4909, 2011 WL 3505148 (Or. T.C. Aug. 10, 2011), holding that a cable company may not be subject to central assessment even though it transmits certain data. Qwest Corp. v. Colorado Div. of Property Taxation, No. 10CA1320, 2011 WL 3332876 (Colo. Ct. App. Aug. 4, 2011), holding that a telecommunications company was not able to be treated like a cable company, whose intangible property is exempt and overall value is capped; also that there was no constitutional violation for treating a telecommunications company and a cable company differently, even though they perform many of the same services. In addition to cable companies, other types of business organizations have been identified as possible candidates for central assessment because they operate across county or state lines. The resolution of these matters will largely depend on the statutory scheme adopted by the particular state. For example, in Beaver County v. WilTel, Inc. (2000 UT 29 ¶19, 995 P.2d 6002), a provider of longdistance telecommunications services argued that it should not be subjected to central assessment when other “classes of enterprises” such as banks, retail furniture chains, cable television companies, and Internet service providers operate across county or state lines and are locally assessed. The court attempted to resolve this dispute by analyzing the level of physical, economic, and functional integration of the operating properties. The court ultimately ruled that the properties WilTel used to provide interstate long-distance communication services were completely integrated and thus were appropriately subject to central assessment as a single operating unit. The court then indicated, in dicta, that it did not appear that the properties associated with the branches of a bank or retail outlet were as functionally integrated because the branches “in some cases could operate independently.” DEFINING WHAT SHOULD BE INCLUDED IN THE UNIT

Almost from the earliest cases there has been controversy regarding which properties should be included in the unit. Should the unit include all property owned by the firm or only the property that is actually used to conduct utility

136

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

operations? Should exempt properties be included in the unit? Should property leased from others be included? Over time, it has generally become accepted that properties owned by a firm that are not required for the operation of the utility business (i.e., “nonoperating properties”) should not be included in the unit (Chapman et al. 1954, 18). Nonoperating properties may include such assets as those associated with other lines of business owned by the firm or unrelated investments. The properties used to conduct the utility business are generally referred to as “operating properties.” Operating properties include all properties necessary to conduct the unitary business, regardless of whether they are owned, leased, or otherwise exempt from taxation in the particular jurisdiction. Over the years, many controversies have arisen about how to properly value the leased operating properties used within the unit and how to properly remove exempt operating properties (i.e., intangible property) from the unit assessments. Utilities and railroads frequently lease properties that are used in their unitary operations. These entities can lease such properties through either a capital lease or an operating lease. Capital leases do not pose much of an issue in unitary assessment because the property is treated as owned on the records of the company. Property used under an operating lease, however, is not recorded on the books of the company, and for financial accounting purposes the company recognizes a rental expense but no depreciation expense. Appraisers recognize that this accounting treatment affects the cash flows capitalized in the income approach described below and that an adjustment typically needs to be made to the income approach to account for the use of operating leased property. The most frequently used method for making this adjustment is to treat the leased property as if it were owned by (1) disallowing the rent expense; (2) allowing a depreciation expense; (3) recalculating the tax obligation associated for the prior adjustments; and (4) making an appropriate allowance for the capital expenditures that would be required to maintain and replace the subject unitary property. Several governmental appraisers have started to suggest that different methods should be used to account for the value of the leased operating property, and litigation has ensued: Delta Airlines, Inc. v. Department of Revenue, 328 Or. 596, 984 P.2d 836 (1999), holding that in the context of a limited life model, the leased operating property could be valued by capitalizing the annual lease payments and adding a residual value of the leased aircraft. Union Pacific Railroad v. Utah State Tax Commission, Case No. 090700830 AA (2nd Dist. Ct. Utah 2012) (decision is pending), dealing with whether the proper way to value operating leased property is simply treating it as though it is owned, or capitalizing the income stream to the lessor.

unit approach to the taxation of railroad and public utility property

137

EXCLUDING INTANGIBLE VALUE

A common dispute in unitary matters is whether unitary assessed property owners are receiving equal protection under the tax laws in relation to locally assessed property owners. This dispute most frequently arises in the context of whether the unit approach is capturing and taxing the value of nontaxable intangible business assets (i.e., goodwill, workforce, software) and nonproperty assets (i.e., business growth expectations) when the locally assessed approaches are not capturing and taxing such values. More than a century ago, the U.S. Supreme Court ruled that states may impose property tax on both tangible and intangible properties: A distinction must be noticed between the construction of a state law and the power of a state. If a statute, properly construed, contemplates only the taxation of horses and wagons, then those belonging to an express company can be taxed at no higher value than those belonging to a farmer. But, if the state comprehends all property in its scheme of taxation, then the goodwill of an organization and established industry must be recognized as a thing of value. (Adams Express Co. v. Ohio State Auditor, 166 U.S. 185, 221 17 S.Ct. 604 [1897])

It has also long been recognized that it is difficult to separate the tangible from the intangible in property values. As one oft-quoted 1906 opinion put it, “One might as well try to value the life-blood of a horse, or his capacity to breathe, as try to place a value upon the visible part of a railroad property separate from its rights, franchises and privileges” (Chicago and North Western R.R. Co. v. State, 108 N.W. 557, 573 [Wis. 1906]). Numerous states have chosen1 to exempt intangible properties from their ad valorem property tax schemes and thus are required to identify and remove the value of intangible properties that may be captured in their unit valuations. The identification and removal of intangible property values from unit assessments has been an ongoing area of litigation. Recent court decisions include the following: T-Mobile USA, Inc. v. Utah State Tax Comm’n, 2011 UT 28, ¶51, 254 P.3d 752, holding that goodwill constitutes intangible property under the generally accepted definition of intangible property, and therefore the value attributable to goodwill is not subject to Utah property tax and must be removed from the unitary assessment.

1. Some states have decided to exempt intangible properties from property tax because they impose an income tax on the earnings from the intangible properties. Such states often perceive that imposing both a property tax and an income tax on such intangible properties could result in a form of double taxation. See generally, T-Mobile USA, Inc. v. Utah State Tax Comm’n, 2011 UT 28, ¶29, 254 P.3d 752, 762.

138

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

Elk Hills Power, LLC v. Board of Equalization, 123 Cal.Rptr.3d 906 (2011) (appealed to Calif. Supreme Court), holding that the value of emission reduction credits (ERCs) could be included in the value of the electric utility property even though the ERCs are intangible assets. Union Pacific Railroad v. Utah State Tax Commission, Case No. 090700830 AA (2nd Dist. Ct. Utah 2012) (decision is pending), dealing with whether custom computer software and a trained and assembled workforce constitute intangible property exempt from property taxation. Beaver County v. Property Tax Division of the Utah State Tax Comm’n, No. 080905451 (3rd Dist. Ct. Utah, Feb. 15, 2012), holding that the stock and debt valuation model should not be used to value unitary property unless the intangible property captured in the model is removed. THE IMPACT OF REGULATION ON UNITARY VALUATIONS

One of the arguments made for central assessment and the application of the unit approach is that the firms in question have historically been regulated by state or federal agencies (Chapman et al. 1954; Janata 1993). These properties are often subject to regulatory schemes that restrict the rates that may be charged by the utility or require compliance with certain operating expenditures. Questions continue to arise regarding the valuation impacts associated with regulation: PacifiCorp v. Idaho State Tax Comm’n, No. CV OC 08 18158 (4th Dist. Ct. Idaho, Sept. 16, 2010), holding that the Idaho State Tax Commission’s assessed value of PacifiCorp was erroneously high, that book depreciation does not account for all forms of obsolescence, and that a rate-regulated utility suffers obsolescence as a result of regulation. Boston Gas Co. v. Board of Assessors, 941 N.E.2d 595, 607–08 (Mass. 2011), holding that the fair market value of a pipeline was greater than the net book value even though it was a regulated utility and only able to earn a return on its rate base. PacifiCorp v. State of Montana, 253 P.3d 847 (Mont. 2011), holding that there was no additional obsolescence to deduct from the original cost less depreciation. Beaver County v. Property Tax Division of the Utah State Tax Comm’n, No. 080905451 (3rd Dist. Ct. Utah, Feb. 15, 2012), holding that a rateregulated utility suffers from additional obsolescence due to its being regulated. Jones v. Southern Natural Gas Co., 63 So.3d 1080 (La. Ct. App. 2011), holding that there was insufficient evidence to allow additional obsolescence as a result of a gas pipeline being a regulated utility.

unit approach to the taxation of railroad and public utility property

139

By the late 1960s and early 1970s, the railroad industry in the United States was in serious financial difficulty. A number of railroad companies were insolvent, and Congress became concerned that the U.S. rail system would collapse. It responded with several reforms that led to a substantial restructuring of the industry. As part of that restructuring, Congress passed the Railroad Revitalization and Regulatory Reform Act of 1976, known since as the 4R Act (Pub. L. 94–210, 90 Stat. 31, 49 U.S.C. § 11501). One of the arguments made by the railroad industry was that states were too slow to grant valuation relief in the face of the industry’s financial troubles. As part of the 4R Act, states were prohibited from unreasonable or unjust discrimination against or an unreasonable burden on interstate commerce. Most important, the act prohibited states from assessing railroads at a higher ratio to market value than that used for other property in the state or at a higher tax rate. The act also gave companies the right to access the federal courts directly in seeking relief from state assessments (Adams 1977). The result was a fundamental change in the application of the unit approach to railroads. The 4R Act was followed by the Motor Carrier Act of 1980 (Pub. L. 96–296, 94 Stat. 792) and section 1371 of the Tax Equity and Fiscal Responsibility Act of 1982 (Pub. L. 97–248, 96 Stat. 324). The net result was to provide railroads, motor carriers, and airlines with effective federal remedies against what were perceived to be discriminatory state tax practices (Janata 1993). A more detailed description of the current approaches, methods, and practices employed by central assessment offices follows.

Overview of the Valuation Methods Employed in the Unit Approach Three questions must be answered before a property tax can be imposed: (1) who will conduct the valuation; (2) what should be valued; and (3) how will the appraiser conduct the valuation? In the context of this chapter, the first question is concerned with whether local or central officials will do the appraisal. As noted in the introduction, state revenue departments usually appraise the complex firms that are the subject here. There are two basic approaches to the question of what should be valued. Assessors may value each separate parcel, improvement, or piece of equipment owned by a firm. When this approach is followed, each taxable property is valued using some measure of cost, such as the historical cost or the current cost of replacement. The final value is the summation of the values of the properties in the individual state. This method is followed in several states (e.g., Virginia and New York) and by most local assessors. The other approach is the unit approach described earlier. Under this approach, the value of the entire enterprise is estimated, net of nontaxable components. The unit method takes advantage of the company-wide financial and operational data made available to regulatory bodies and the financial data

140

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

provided to equity owners. As an example, an appraiser valuing Delta Airlines would estimate the entire value of the company. After determining the total value, the appraiser would allocate a specific “share” of the total value to the state doing the appraisal. The geographic and financial size of the companies that state tax agencies must deal with can be intimidating. Transportation companies, for example, may provide service in every state and literally thousands of local taxing jurisdictions. Electric utilities usually operate in fewer but still multiple states. Furthermore, electric utilities are often the largest industrial and commercial operation in a state or region. Telecommunication companies not only operate in the entire United States, but they are also becoming worldwide in their dimensions. Regardless of the geographic size of such firms, even more intimidating are their total assets. It is not uncommon to value companies whose assets are in excess of several billion dollars. The third question, concerning how public utilities and transportation companies are valued, has significant policy implications. The valuation process for all taxed properties must meet the requirement that the process achieve a reasonable estimate of fair market value or a valuation result that represents what informed buyers and sellers would accept. Appraisal professionals may employ three approaches to estimate the value of properties and firms: the cost approach, the income approach, and the sales approach. Some aspect of each of these approaches is typically used when valuing complex properties using the unit approach. COST APPROACH

The logic implicit in using a cost approach to value property is that no buyer would be willing to pay more for an asset than the cost of purchasing and assembling the various components of the property. Several variations of the cost approach can be used. The four most widely accepted cost models are historical cost less depreciation, regulatory rate base, reproduction cost less depreciation, and replacement cost new less depreciation.2 Historical Cost Less Depreciation (HCLD) This widely used cost model attempts to determine value by taking the historical costs for the property reported on the firm’s books minus the accounting depreciation allowed by regulatory or financial accounting rules. Book depreciation is often based on a straight-line convention and may not be consistent with the actual depreciation of the property within its respective market. Consequently, the HCLD model should be adjusted for additional forms of market depreciation (i.e., obsolescence) if an accurate valuation estimate is to be achieved using this cost model. If the firm being valued is regulated, the appraiser will also need to consider adjusting the HCLD

2. For a good discussion of the cost approach applied to utilities, transportation companies, and telecommunication companies, see Janata (1993).

unit approach to the taxation of railroad and public utility property

141

model for costs on which the firm is not allowed to earn a return. Theoretically, purchasers would not pay more for a property than its existing earnings base unless they were willing to accept a lower rate of return than is currently being earned. HCLD must be adjusted for obsolescence (i.e., loss in potential earning power) and to recognize some assets that do not contribute to earnings. The goal is to determine the net investment of all taxable assets adjusted for changes in earning power of those assets that do not contribute to earnings in the normal rate-of-return-times-earnings-base calculation. For a description of how obsolescence is calculated for railroads, see Adolphson, Cornia, and Walters (1989). Rate Base This model is similar to HCLD, but may include additional adjustments for items on which the firm is not allowed to earn a return. This model is based on the observation that the earnings of regulated companies in the United States are primarily determined by rate of return times the rate base that is allowed by the regulatory body. Because rate base is the base on which earnings are determined, it follows that this base may be a primary indicator of value.3 This model uses the actual rate base of the company, which is established by a regulatory body, instead of using HCLD as a surrogate. The basic formula employed in the determination of rate base is given by this equation: Rate Base ! (C " D)r # $ # % # m " CWP " DFIT " DFITC where

C ! original cost of the tangible assets D ! accumulated depreciation r ! allowed rate of return $ ! cash % ! working capital m ! materials and supplies CWP ! construction work in progress DFIT ! deferred federal income taxes DFITC ! deferred federal investment tax credits

As for the HCLD model, this approach is further adjusted for obsolescence to account for differences between the earned rate of return and the allowed rate of return due to changing external conditions. Reproduction Cost Less Depreciation This approach inflates HCLD by a price index to bring it up to current costs. It is an estimate of the current cost to replicate the existing system. This model is much less widely used because of the complications of the indexing procedures as well as the necessity of adjusting

3. For a discussion of the establishment of rate base and the regulatory process, see Phillips (1993).

142

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

for the functional obsolescence of existing plant. Because reproduction cost less depreciation has no relationship to the existing rate base, additional economic and functional obsolescence must be accounted for. Replacement Cost New Less Depreciation The replacement cost model estimates the current costs to construct a replacement property of equivalent utility using current technology and design standards. This model tends to eliminate much of the functional obsolescence that might be present in a reproduction or HCLD model. Replacement cost models are best suited for industries that experience rapid technological advancements, such as the wireless telecommunications industry. Once the current replacement cost estimate is derived, the appraiser then has to remove depreciation from the estimate to account for the age and condition of the property being appraised. One of the major challenges associated with implementing the cost approach has to do with the frequent disconnect between historical cost and the earning potential of assets. It is often the case that assets maintain their functional efficiency, but external forces such as changing technology render the assets obsolete. Yet it is sometimes the case that buyers may be willing to pay a premium for a particular asset because market conditions suggest extraordinary earning potential for that asset. In general, nonregulated markets pay much less attention to historical or reproduction cost than they do to potential future earnings. This leads to the use of projected income as another indicator of value (Adolphson, Cornia, and Walters 1992). INCOME APPROACH

The logic behind the income approach to value is that a knowledgeable buyer would not pay more (and no seller would take less) for an asset than the timeadjusted value of the net income stream generated by the asset. Two basic income approaches are utilized in utility valuation. Both are variations of the yield capitalization methodology (Damodaran 2002; Koller, Goedhart, and Wessels 2010). Traditional Perpetuity Capitalization This approach is well suited to public utility valuation. It capitalizes a stable, level annual income by assuming that annual depreciation charges will be reinvested annually. This produces a level rate base and, thus, a level income. This net operating income (NOI) is then simply divided by the market capitalization rate (r). The same mathematical formula (V ! NOI/r) is also appropriate if the intent is to only value the assets in existence on the lien date. In such a situation, it is assumed that depreciation is equal to the amount of replacement capital expenditures necessary to maintain the existing assets into perpetuity. Thus, the cash flow (CF) to be capitalized is deemed to be equal to NOI. This formula can also be expressed as V ! CF/r " g where the growth (g) is equal to 0 percent when it is expected that the cash flows will remain constant into perpetuity.

unit approach to the taxation of railroad and public utility property

143

Discounted Cash Flow Analysis This method attempts to quantify all future cash flow (net operating income plus depreciation plus deferred taxes less capital replacement) for some period into the future and then assumes perpetuity thereafter. It, too, is well suited to utility valuation because future income is somewhat predictable because of regulatory oversight. It is also more sensitive than perpetuity capitalization because it attempts to quantify all future cash flows. The basic formula for a discounted cash flow is as follows: T

NOIt t (1 t
0  r)

Value
å where

NOI ! net operating income in time period t r ! rate of return T ! the life of the asset in years

One of the major challenges of the income approach is the determination of r, the rate of return. States have detailed processes to collect market data and determine reasonable estimates of market rates of return for various industries (e.g., Property Tax Division 2010). The process often involves industry input and comment, but it can nevertheless be controversial. Expert opinion may concur on a range for r, but differences at the second or third decimal point can result in substantial differences in the ultimate tax bill when overall values are in the billions of dollars. SALES COMPARISON APPROACH

The third approach to value is the sales comparison approach, which involves examining the market and observing what actual buyers are paying (and sellers accepting) for similar assets. Because public utilities seldom sell on the open market, the typical sales approach cannot be used. Two alternative approaches have been applied: the stock-and-debt method and the direct capitalization method. Stock-and-Debt Method This approach follows the accounting principle that assets equal liabilities plus capital. Although it is not possible to directly determine the market value of all items on the asset side of the balance sheet, it is feasible to determine the market value of all liabilities and equity (longterm debt and common stock). This is then assumed to equal the value of all of the firm’s assets: tangible, intangible, operating, and nonoperating. This method is now used infrequently. Several difficulties arise in this approach because of problems in determining certain liabilities (e.g., accumulated deferred income taxes), allocating common stock value to subsidiary companies, and figuring the deduction for nontaxable intangible and nonoperating assets.

144

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

Direct Capitalization Direct capitalization is a form of the comparable sales approach. The most popular version is the use of capitalization rates based on income-to-market ratios (e.g., earnings/price ratios), although other nonincome ratios can also be utilized (e.g., book value/price ratios). The reliability of the method is directly proportional to the comparability of those companies chosen for the derivation of the ratios. To the extent that such ratios are not derived from truly comparable companies, this approach can produce unreliable results. From this discussion, it should be clear that the major challenge in implementing the comparable sales approach to value is the difficulty in finding truly comparable properties that are actively traded. Surrogates based on market transactions for entire firms or individual stocks require the appraiser to make adjustments for intangible and nontaxable components embedded in observed prices. This has proved to be controversial. CORRELATION

The multiple appraisal approaches are listed in figure 5.1. After the various appraisal approaches, or some selected subset, are completed, the appraiser must Figure 5.1 Appraisal Approaches and Techniques Used to Value Utilities and Transportation Companies

Income indicators # Perpetuity capitalization # Discounted cash flow

Cost indicators # HCLD # ReproCLD # Rate base # ReplCLD

Correlation

Allocate value to the state

Determine legal assessment level

Apportion value to local taxing jurisdictions

Comparable sales indicators # Stock and debt # Direct capitalization

unit approach to the taxation of railroad and public utility property

145

then determine the appropriate estimate of the unit value. Determining the final value is required because the estimates of value from the various appraisal methods are rarely equal. The appraiser must weigh the evidence and determine the final value for taxation. Correlation, the process of combining various approaches to value and determining one value, is a common practice in the appraisal industry. The appraiser employs his or her best judgment in considering the quality of underlying data and the results of the various approaches, and then makes a final estimate of unit value. In this process, all the knowledge and experience of the appraiser come into play to form a final opinion of value.

Administrative Issues The process of moving from the unit value to taxable value in each jurisdiction is also illustrated in figure 5.1. Allocation of the correlated unit value to a specific state is usually determined by comparing relative investment or usage in a specific taxing jurisdiction. Quantitative measures include such things as the relative number of units (miles of pipe, generating stations, seat miles, barrels of oil, etc.), the cost of facilities (either depreciated or undepreciated), or revenue factors such as gross or net revenue per state. If, for example, a state generated 50 percent of total ton miles shipped by a railroad, it could be allocated 50 percent of the total unit value of the railroad. It is important to understand that allocation is an assignment of value rather than a determination of value. The next step, as shown in figure 5.1, is to adjust the allocated property for any exempt property value and for the legal assessment level. It is often the case that property not subject to property tax is included in the unit value. For example, motor vehicles may be separately taxed, but their aggregate value is included in the unit value. As noted earlier, intangible property is often also excluded from the property tax base. Unless the intangible property has previously been removed from the individual valuation indicators used to derive the correlated value, the state agencies must identify and subtract any property (tangible or intangible) in their state that is included in the correlated unit value but is not subject to the property tax. The assessment level is then adjusted to the legally established ratio between the market value of a property and the taxable assessed value. Currently, in the United States, assessment ratios range from less than 10 percent to 100 percent of value. The final step is to determine how much assessed or taxable value to assign to each specific taxing jurisdiction within the state. This process is known as apportionment. It is similar to allocation but is practiced on an intrastate basis. The apportionment is usually accomplished by dividing the total taxable value of a firm by a distribution base. The distribution base varies by the class of utility or railroad. For example, the distribution base for transportation companies may be the total miles of track operated by a company in a state. The resulting factor is then multiplied by the track miles in a specific taxing jurisdiction. A taxing

146

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

jurisdiction with 10 percent of the total track miles in the state would have 10 percent of the taxable value of the railroad apportioned to it.

Valuation Appeals Companies are given informal opportunities to provide input on issues of fact and on the methods used before a final valuation is determined. After a final value has been determined, disagreements are appealed to a central or state board of equalization or its equivalent. Matters before the state board of equalization are usually conducted as a formal hearing and result in a written decision from the board. Issues about value can often be raised by both companies and other interested parties, such as local governments. Issues from the board of equalization can often be appealed directly to the trial or appellate court. It is not uncommon to have numerous annual valuations appealed in each state. The resolution of valuation appeals can take years, and these appeals can cause substantial revenue instability for local governments. Unitary valuation methods are subject to the same standards as nonunitary valuation methods in that the methods must be accurate and reasonably designed to achieve a fair market value estimate for the unitary property that is taxable. Consequently, there are continued controversies in the arena of unitary valuation regarding (1) the use of models like the stock-and-debt and direct capitalization models, which tend to include significant amounts of nontaxable, intangible properties that are difficult to remove from the assessment; and (2) proper valuation techniques used with particular models, such as proper capitalization techniques: Utah Ass’n of Counties v. Tax Comm’n, 895 P.2d 825, 828 (1995), holding that the evidence before the state tax commission supported the “Commission’s decision to disregard the stock and debt method.” PacifiCorp v. State of Montana, 253 P.3d 847 (Mont. 2011), holding that it was appropriate to use an earnings-to-price ratio to determine a discount rate to be used in a direct capitalization income method. Airtouch Communications, Inc. v. Department of Revenue, State of Wyoming, 76 P.3d 342, 360 (Wyo. 2003), holding that there should have been a flotation cost adjustment to the capitalization rate. PacifiCorp v. Idaho State Tax Comm’n, No. CV OC 08 18158, ¶¶ 44, 47, 51 (4th Dist. Ct. Idaho, Sept. 16, 2010), accepting the valuation proposed by PacifiCorp and acknowledging that it contained a flotation cost adjustment to the cost of capital. Colorado Interstate Gas Co. v. Property Tax Administrator Mary Huddleston, 28 P.3d 958, 962 (Colo. Ct. App. 2000), holding that when assessing property as a unit it is not required to make a flotation cost adjustment to the capitalization rate.

unit approach to the taxation of railroad and public utility property

147

PacifiCorp v. Property Tax Division of the Utah State Tax Commission, Utah State Tax Commission Appeal No. 06–0767, ¶¶ 152–168 (Feb. 28, 2008), holding that the appropriate cost of capital should not include a flotation cost adjustment. Global GT LP v. Golden Telecom, Inc., 993 A.2d 497 (Del. Ch. 2010), in which the court used the lower Ibbotson supply-side equity risk premium as opposed to the higher Ibbotson historical equity risk premium when valuing a business in an appraisal action brought by dissenting shareholders of a merger. Assessor of Roger Mills County v. Unit Drilling Co., 247 P.3d 1170 (Okla. 2011), holding that a statute mandating the use of a certain publication to determine the value of property violated the fair market value constitutional mandate. Beaver County v. Property Tax Division of the Utah State Tax Comm’n, No. 080905451 (3rd Dist. Ct. Utah, Feb. 15, 2012), holding that an administrative rule requiring certain applications when valuing property did not violate the constitutional fair market standard.

Political Issues In practice, the division of functions between valuing property and using tax revenues from the property creates organizational and political problems. A central department of revenue faces a number of external publics that it must try to please. The two most important of these are the companies that are appraised by the department and the local government that benefits from the resulting tax base. These two groups are often at odds, as noted earlier. The companies understandably want to minimize the appraised value of their properties and lower their tax obligations. Local governments want the appraised value to be as high as possible. Pleasing one group often means offending the other. Displeasing the companies will likely lead to a series of appeals before a hearing board and, eventually, cases briefed and argued before the courts. Property tax litigation can be an expensive and uncertain process for resolving differences between taxpayers and departments of revenue. Such cases are often complex and require careful analysis and a substantial degree of understanding on the part of the parties and the judge in order to obtain a judgment that can resolve the dispute and provide a workable framework for future valuations. The administrative hearing and court processes can also be time-consuming. It is not uncommon for a centrally assessed value dispute to take four to eight years to complete. The lengthy pendency of these actions places a heavy burden on both the taxpayer and the taxing authority. Neither may be certain as to the outcome of the dispute. Thus, the taxpayer may be required to either pay or accrue significant amounts of tax that may ultimately be determined to not be owed. Similarly, the taxing

148

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

authority may have to adjust its operating budgets because it cannot collect the disputed taxes or because it is required to escrow the disputed taxes in order to be able to pay a potential future refund. Displeasing local governments can be equally uncomfortable. While many local governments do not have the resources to carry a fight over valuations to the courts, they have other avenues at their disposal. One of the most effective techniques local governments use is to raise questions with elected state political officials about the ability and motivation of the revenue department. When valuation issues are raised in a political context, the outcome is perhaps even more uncertain than when these issues are resolved by the courts. Thus, a central revenue department must maintain balance on a difficult tightrope, with the ultimate balance point being the fair market value. In the Beaver County example described earlier, centrally assessed property was responsible for a significant change in overall revenues. However, it may be the case that the valuation process and timing within a state result in significant shifts in the tax burden. The normal practice is for states to update the value of centrally assessed property every year. But few local governments follow a similar practice for locally assessed property. One recent survey of state tax policies and practice found that while 35 percent of states legally require that property be reappraised every year, less than 26 percent of states actually revalue locally assessed property annually. Nearly half the states surveyed indicated that the common practice was to reappraise locally assessed property on a four-year cycle or longer (Dornfest et al. 2010). This disparity in appraisal cycles can create unintended shifts in the property tax burden and significant political problems for local officials. For example, assume that both locally and centrally assessed properties are increasing at a rate of 2 percent each year. Assume further that centrally assessed property is revalued every year and the revaluation accurately captures the 2 percent increase in value, but locally assessed property is only revalued every fifth year. Under this scenario, each year a greater proportion of the tax collected will be paid by centrally assessed property owners. In the fifth year, when locally assessed property is revalued, there will be a large increase in taxable value. Given the assumed 2 percent annual increase, the taxable value of locally assessed property will increase by 12.6 percent. Local elected officials will then face a major challenge. If they leave the tax rate unchanged, local property owners will incur a 12.6 percent increase in their property taxes. Faced with the likely political fallout from such an increase, local officials may lower the tax rate to achieve a revenue-neutral outcome. Even if neutrality is the objective, local property owners will face a substantial tax increase, while centrally assessed property owners will enjoy a substantial tax reduction. The only way to avoid this dilemma is to match the revaluation cycles of local and centrally assessed property. Some states accomplish annual local revaluations with statistical models and value indexing.

unit approach to the taxation of railroad and public utility property

149

Conclusions This chapter began by describing the context in which unit valuation takes place. It is a high-stakes environment for both taxpayers and local governments. Many of the issues that continue to be litigated today have been debated in one form or another for the last century. But one reason for the continued debate is that industries change and evolve over time, and those changes should be reflected in tax policy and practice. All too often, the policy and practice lag is substantial, resulting in inequitable treatment of one party or another. One may ask, for example, if the tax burden placed on centrally assessed property is commensurate with the benefits received from public services or the costs imposed on local communities. Return for a moment to the Beaver County example. In 2010, electric utilities in the county paid more than $3.8 million in property taxes, while householders paid less than $1.5 million. It seems very unlikely that electric utilities imposed more than 2.5 times the burden on local services as did local residents. In more urbanized jurisdictions, the relative tax burdens may be more commensurate with benefits received, but that does not seem to be the case in many more rural parts of the country. The final issue relates to the impact of regulation on valuation. As noted, one of the arguments made for using unit valuation and centrally assessing these properties is that they have traditionally been regulated by state or federal agencies. But over time, many of these industries have been increasingly deregulated. A recent study by the New York State Department of Taxation and Finance found that even in states where industries such as electric utilities have been largely deregulated, little has changed in the assessment practices of state agencies (Department of Taxation and Finance 2005). This finding may be perfectly reasonable if in fact there are other compelling reasons why some properties should be centrally assessed using unitary valuation methods while others continue to be assessed by local officials using a very different (local) unit and only a subset of the methods employed by state tax officials. In other contexts, we have explored some of the implications of deregulation for tax assessment in the electric industry (Cornia and Walters 2000; Walters and Cornia 1997; Walters and Cornia 2001), but the issue deserves further study. It may well be the case that complex properties with assets in multiple jurisdictions and working as an integrated whole merit a different definition of the unit to be valued and a different set of valuation methods. If such is the case, then it is also reasonable to ask why other industries that meet this criterion are not uniformly assessed by states using the unit approach. Consider again the example of cable TV companies. As cable companies increasingly offer services that have traditionally been provided by telephone companies, it is very hard to see why the two industries should be treated differently by property tax policies. Indeed, some states, but certainly not all, have now begun centrally assessing cable operations.

150

Gary C. Cornia, David J. Crapo, and Lawrence C. Walters

These considerations raise again some of the central questions that have been at the heart of the unit approach since its earliest application in the railroad industry. What tangible, intangible, operating, and nonoperating properties are included in the unit that is centrally assessed, and how does that combination of properties differ from the unit identified and assessed by local tax assessors? Does the difference in the choice of unit result in unfair or inequitable treatment of some taxpayers? Is there a way to reliably measure the fair market value of integrated, multijurisdiction tangible property assets as a unit without capturing the intangible asset values policy makers wish to exempt? These remain important issues that merit significantly more attention than they have received from scholars in recent years.

references Adams, A. 1977. Railroad Revitalization and Regulatory Reform Act of 1976: An interim review. Business Law 32:975–1001. Adolphson, D. L., G. C. Cornia, and L. C. Walters. 1989. Railroad property valuation using Data Envelopment Analysis. Interfaces 19(3):18–32. ———. 1992. Measuring obsolescence in an electric utility. Journal of Property Tax Management 1:134–156. Advisory Commission on Intergovernmental Relations. 1963. The role of the states in strengthening the property tax, vol. 1. Washington, DC: U.S. Government Printing Office. California State Board of Equalization. 2003. State Assessment Manual. Property and Special Taxes Dept. Sacramento: California State Board of Equalization. Chapman, C. M., L. A. Stiles, R. B. Welch, and A. L. Weston. 1954. Appraisal of railroad and other public utility property for ad valorem tax purposes. In Report of the committee on unit valuation of the National Association of Tax Administrators. Chicago, IL: Federation of Tax Administrators. Cornia, G. C., and L. C. Walters. 2000. Electric utility deregulation and the property tax in the United States. In Impacts of electric utility deregulation on property taxation, ed. P. Burling, 43–63. Cambridge, MA: Lincoln Institute of Land Policy. Damodaran, A. 2002. Investment valuation: Tools and techniques for determining the value of any asset. New York: John Wiley. Department of Taxation and Finance. 2005. Survey of railroad and utility taxation practices among the states: 2005 update. Albany: Department of Taxation and Finance, State of New York. Dornfest, A. S., S. Van Sant, R. Anderson, and R. Brown. 2010. State and provincial property tax policies and administrative practices (PTAPP): Compilation and report. Journal of Property Tax Assessment and Administration 7(4):5–111. Janata, J. F. 1993. Property taxation. 2nd ed. Washington, DC: Institute of Property Taxation. Koller, T., M. Goedhart, and D. Wessels. 2010. Valuation: Measuring and managing the value of companies. 5th ed. Hoboken, NJ: John Wiley. Phillips, C. F. 1993. The regulation of public utilities. Arlington, VA: Public Utility Reports.

unit approach to the taxation of railroad and public utility property

151

Property Tax Division. 2010. Capitalization rate study for centrally assessed properties for January 1, 2011. Salt Lake City: Utah State Tax Commission. U.S. Census Bureau. 1994. Assessed valuations for local general property taxation. Washington, DC: U.S. Government Printing Office. Walters, L. C., and G. C. Cornia. 1997. The implications of utility and telecommunications deregulation for local finance. State and Local Government Review 29(3):172–187. ———. 2001. Electric utility deregulation and school finance in the United States. Journal of Education Finance 26(Spring):345–372.

commentary J. Fred Giertz The property tax is the most important tax source for most local governments in the United States. The availability of an independent source of revenue is key to the autonomy of local governments. Unlike most other taxes, where the base is determined by a market transaction, the base of the property tax depends on an assessment process. The assessment function is essential to the effective operation of the property tax. Property tax assessment is often viewed as a routine, mundane part of property tax administration that simply requires the application of well-established algorithms to determine the tax base. This may be true to a certain extent for property such as homogeneous residential tracts in stable jurisdictions with tranquil housing markets. As the property being assessed becomes larger and more unique, the difficulty of the assessment process increases because there are few if any comparable sales and little new construction of similar assets. Property owned by a business whose operations extend beyond the boundaries of the jurisdiction may present even more challenges. Should the local property be assessed as a freestanding asset similar to most other types of property, or should it be valued as part of a larger interconnected operation (i.e., as a component of a unitary operation)? The issues related to unit assessment are not new; the practice has been in place for more than a century. Somewhat surprisingly, a number of unresolved questions regarding the unit approach remain. These go to the heart of the practice, including both when and how it should be applied. The chapter by Gary G. Cornia, David J. Crapo, and Lawrence C. Walters provides a valuable review of the unit approach, including its history and development, current practices, and new issues that arise in a changing legal and technological environment. The authors make a compelling case for the importance of the unit approach as it relates to the broader issue of infrastructure development and its privatization. They bring an impressive array of economic, accounting, and legal expertise to the analysis, along with an equally powerful range of experience, including scholarly writing, teaching, government administration, and private consulting. They are quick to point out that the unit approach is still a developing and sometimes controversial practice, the result of imprecise legal rules and administrative practices that vary markedly from jurisdiction to jurisdiction. Some states tax only real property, while others tax real and personal property, which opens the door to the slippery issue of the inclusion or exclusion of intangibles and how to value them. Many states centrally assess unit operations and then apportion the assessed values among the jurisdictions, while some integrated businesses are assessed locally in some jurisdictions. Differing regulatory environments may also affect assessments, which means that there will never be a universal template for unit assessment that applies in all settings. 152

commentary

153

It should be noted that there is often a degree of intrinsic arbitrariness to tax administration practices. For example, depreciation rules for tax purposes do not precisely measure economic depreciation; transfer pricing rules are only a rough estimate of the value of intrafirm transactions; and the apportionment of corporate income for state tax purposes is often based on a three-factor formula (some combination of sales, employment, and property) that may bear little relationship to the profitability of a firm in a particular state. The unit assessment process is another example. The unit approach must not only determine the value of the unit operation, such as a railroad or other public utility, but must also distribute the assessed value among the component jurisdictions. Because of the indivisibility of the unit operation, this apportionment process cannot precisely reflect the local value of the unit operation. In fact, there may be no definitive way to apportion value geographically. Probably the most important and difficult issue in regard to unit assessment is not how to conduct the analysis, but when. After a century, there is still not a clear rule or procedure for when to use the process. For example, it is generally accepted for railroads and electric utilities. However, new issues arise when structural changes take place. For example, electric utilities are separating the generation and distribution aspects of their operations. How does this affect the unit approach? The appropriate extent of the application of the unit approach also raises some interesting issues. Consider the assessment of two retail establishments in a situation where intangibles are included. If a Walmart and a Kmart of nearly identical construction are located near each other, the usual nonunit assessment approach would value the two properties about the same. However, if unit approaches are used where the overall profitability of the two businesses is included and then apportioned among the various components, the Walmart facility would likely be assessed at a much higher level than the Kmart property. The sum of all the locally assessed Walmart properties would likely fall far short of the overall value of the firm. In 2011, a Forbes article reported that the value of Walmart’s trademark constituted nearly 20 percent of the firm’s capitalized value (Stonefield 2011). It could be argued that the operations of Walmart are no less integrated than those of railroads and electric utilities. Walmart’s success is based on an efficient, centralized system of purchasing, distribution, and promotion. No stand-alone retail establishment can match this. However, for legal reasons and historical precedents, the organization is not treated as a unit entity. It would be useful to establish some clearer guidelines about the rationale for unit assessment. In regard to infrastructure questions, a number of property taxation issues need to be addressed. How should previously government-owned, tax-exempt operations be treated when their functions are privatized, either through an outright sale or through a long-term lease? Privatized parking facilities, parking meters, toll roads, and airports are typically not subject to property taxation when they leave the public sphere. In one sense, the question may be of only minor

154

J. Fred Giertz

significance since the exemption from property taxation would presumably increase the value of the franchise, yielding higher purchase prices or franchise fees and offsetting the lost potential property tax revenue. However, there would likely be a reallocation of revenues among governments, where the granting government would capture the revenue instead of the jurisdictions in which the facilities are located. To the extent that privatized infrastructure is taxed, the issue of unit assessment is certain to arise. In summary, the chapter provides an important contribution to the understanding of the unit assessment approach, including its history, current practice, and unresolved issues. It is not surprising that the analysis does not solve all the issues connected with the unit approach since some of these issues result from unclear legal and regulatory policies. On a more basic level, some issues can never be finally resolved because of the intrinsic geographic indivisibility of certain costs and revenues for highly integrated operations. The chapter does, however, provide the reader with a clearer understanding of these issues.

reference Stonefield, S. 2011. The 10 most valuable trademarks. Forbes (15 June). www.forbes .com/sites/seanstonefield/2011/06/15/the-10-most-valuable-trademarks.

6 The Location Effects of Alternative Road-Pricing Policies Alex Anas

S

ince the early 1970s, urban economists have recognized the importance of a general equilibrium model of the urban economy. Initially, however, they developed such models only for monocentric cities in which all jobs were assumed to stay in a central business district (CBD). Thus, although the analytical solution of the monocentric city model yielded many theoretical insights, it remained empirically inappropriate and difficult to justify in policy application. Early contributions to the general equilibrium model of a monocentric city included Mills (1972), Dixit (1973), and Sullivan (1986), who developed the most complete models all solved numerically. A parallel line of developments that started in the early 1970s led to the National Bureau of Economic Research (NBER) urban simulation model (Ingram, Kain, and Ginn 1972; Kain and Apgar 1985; Kain, Apgar, and Ginn 1976, 1977, 1982) and the Urban Institute Model (Struyk and Turner 1986; Turner and Struyk 1983; Vanski and Ozanne 1978). These two urban simulation models were applied primarily to the study of housing market issues, but they did not The author acknowledges the support of research award RD-83184101-0 from the United States Environmental Protection Agency’s 2004 Science to Achieve Results (STAR) competition as well as the Multi-campus Research Program and Initiative (MRPI) grant from the Office of the President, University of California (award number 142934), of which grant Alex Anas is the scientific director. Views expressed are solely those of the author and not of the financial supporters. Don Pickrell was the discussant of this paper at the 7th Annual Land Policy Conference: Infrastructure and Land Policies, convened by the Lincoln Institute of Land Policy, June 4–5, 2012. His generous and careful comments on a rough earlier draft are highly appreciated. Gregory Ingram read and made helpful comments on an earlier draft. Tomoru Hiramatsu helped with the simulations reported here.

155

156

Alex Anas

treat in sufficient detail and microeconomic depth employment location, the relationships among industries and interindustry trade, the redevelopment of the building stock and the complexity of trip making, or the interaction between labor and housing markets. They also did not treat traffic congestion. The general equilibrium theory of a polycentric city with dispersed employment is more recent. Such models have been developed for linearly shaped hypothetical cities, with jobs endogenously located anywhere in the city. The earliest version of such models, by Anas and Kim (1996), included traffic congestion and agglomeration economies: the weaker the agglomeration economies or the higher the traffic congestion, the larger the number of places to which jobs disperse in equilibrium. The effects of congestion pricing on job and residence location were studied by Anas and Xu (1999), and tolls and the urban growth boundary were compared in Anas and Rhee (2006). Cordon tolls have been studied numerically by Fujishima (2011), who applied the Anas-Xu model to a stylized version of Osaka, and by Anas and Hiramatsu (2012a), who applied the RELU-TRAN2 model to the Chicago metropolitan statistical area (MSA). The RELU-TRAN computable general equilibrium (CGE) model (Anas and Liu 2007) is an empirically applicable model that is well grounded in microeconomics based on the theoretical structure of the Anas-Xu model. It includes real estate development and redevelopment under perfect foresight based on Anas and Arnott (1991, 1993, 1997). RELU-TRAN2 is an extension that includes gasoline consumption and the choice of vehicle type. Using RELU-TRAN2, the effects of the gas price on the urban economy were studied in Anas and Hiramatsu (2012b) and the effect of cordon tolling policies for the Chicago MSA in Anas and Hiramatsu (2012a). The purpose of this chapter is to report an empirical application of RELU-TRAN2 to the analysis of alternative road-pricing policies other than cordon tolling to reduce traffic congestion in the Chicago MSA. The model and its calibration are described and some simulation results are presented from the application of RELU-TRAN2 to assess the impacts of hypothetical road-pricing policies in the Chicago MSA. Tolling the traffic congestion externality has two major effects on location patterns. First, workers are induced to move closer to employment centers to reduce travel distances over which the toll must be paid. Second, employers may decentralize and move closer to employees or customers to avoid paying higher wages to attract workers. Because job and residence locations are interdependent, the net effect is ambiguous. An important question, therefore, is how the net effect varies according to the specifics of a road-pricing policy. In the presence of public transit, these locational responses become relatively milder to the extent that drivers can avoid road charges by switching to transit. This chapter focuses on the quasi-Pigouvian tolling of all major roads only, and of major and local roads only. Against these benchmarks, it compares the effects of a tax on gasoline that is revenue neutral with respect to each quasiPigouvian tolling. These policies are introduced and discussed in more detail, and the results of the simulations are presented.

the location effects of alternative road-pricing policies

157

The main focus is to understand the effect of road-pricing policies on the location of jobs and residences within the Chicago MSA, on urban wages and rents, and on real estate prices and land development. The issue is relevant to the inquiry about the impact of road pricing on central city revitalization and whether pricing centralizes or decentralizes land use, jobs, and population. Almost two decades ago, in 1994, a special report of the National Research Council debated the issue and concluded: Neither theory nor research on the relationship between the cost of transportation and urban development provides compelling evidence to support whether congestion pricing would have a centralizing or decentralizing effect. (Deakin 1994)

The Chicago simulations show that quasi-Pigouvian tolls can both centralize and decentralize the location of jobs and residences but that fuel taxes much more strongly centralize the location of jobs and residences. Under the quasi-Pigouvian tolling of only the major roads, jobs are weakly centralized in the CBD while a much stronger movement of jobs to the outer suburbs is also observed. Average wages, rents, and real estate prices increase under all of these policies. Urban sprawl, measured as the depletion of undeveloped land, increases under all policies, but there are significant differences among the policies. A conclusion that emerges from these results is that the road-pricing policies, and especially the fuel tax, can indeed help significantly concentrate jobs and population in the central city and toward the downtown and thus may help central city revitalization and the reduction of urban sprawl. This conclusion provides at least a first tentative answer to the 20-year-old controversy. The following section explains the structure and calibration of the CGE model with a heavier focus on consumer behavior, including travel. More detailed descriptions of the model can be found in Anas and Liu (2007), and its calibration is covered in Anas and Hiramatsu (2012a, 2012b). The next sections of the chapter describe the road-pricing policies to be tested by simulation and discuss the results of the policy tests.

The RELU-TRAN CGE Model RELU-TRAN is a CGE model, calibrated and tested for the Chicago MSA, described in Anas and Liu (2007).1 In RELU-TRAN2, an extension of RELUTRAN, the travel behavior of the consumer has been enriched by treating the choice of automobile type by fuel economy level and by adding equations that calculate gasoline consumption and carbon dioxide (CO2) emissions from automobile travel (Hiramatsu 2010). In the model, the Chicago area is represented by 1. This section describes the structure of the model and its calibration, and thus borrows heavily from corresponding sections in Anas and Hiramatsu (2012a, 2012b).

158

Alex Anas

a system of 15 zones covering the entire area and by an aggregation of the major road network and of local roads. REPRESENTING THE CHICAGO MSA

Figure 6.1 shows the 15-zone Chicago MSA used in the model. The zones can be grouped into five concentric rings. Ring 1 consists of zone 3, the major employment center in the region, which we will refer to as the CBD. Ring 2 includes zones 1, 2, 4, and 5, which together with the CBD make up the city of Chicago. Ring 3 consists of zones 6 to 10, which include all of the inner suburbs encircling the city. Ring 4 (zones 11 to 14) covers the outer suburbs, and zone 15, a single peripheral zone, represents the exurban areas, which are primarily rural

Figure 6.1 RELU-TRAN Zones for Chicago MSA

15

12

13

McHenry

Lake

6

N.W. Cook

11

Kane

15

Lake Michigan 7

N. Cook

1

5

10

DuPage

8

W. Cook 2

3 4

9

S. Cook

14 Will

15

Source: Anas and Liu (2007).

1 2 3 4 5

Chicago North Chicago West Chicago Center Chicago South O’Hare

the location effects of alternative road-pricing policies

159

in character and include some parts of northwestern Indiana and southeastern Wisconsin. All 15 zones are included as possible locations for consumers, but those who choose either a residence or a job in the peripheral zone 15 are treated as having partially exited the region. Such consumers can still choose a job or residence in one of the 14 zones, but the wages they earn or the rents they pay in zone 15 are taken as exogenous and are not adjusted in the general equilibrium that the model calculates for the 14 nonperipheral zones. In the base simulation reported in this chapter, residents located in peripheral zone 15 are only 5 percent of the total. All trips that originate and terminate within the same zone utilize a local road, which is an abstract aggregation of the underlying system of streets and minor roads. Trips originating in one zone and terminating in another utilize a path over the interzonal road links shown in figure 6.2, a crude aggregation of major roads and highways, but they also use the intrazonal roads to access and egress from the interzonal road network. Figure 6.2 shows the aggregated interzonal road network consisting of 34 two-way (68 one-way) interzonal road links connecting the zone system. In the model, each local road, each one-way interzonal link, and each intrazonal road is represented by a capacity that is crucial in calculating congestion. The model calculates an equilibrium congested travel time for each local road and each one-way interzonal link, as discussed below. MODEL STRUCTURE: CONSUMERS, FIRMS, AND DEVELOPERS

The model is microeconomic in structure and consists of consumers, firms, real estate developers, and an abstract public sector that sets road tolls or other taxes and may or may not redistribute the revenues generated by various policies. Consumers, firms, and developers in the RELU model are treated in submodels that correspond to different markets: the housing market, the labor market, and the markets for industrial output. In all of these markets, consumers and firms are perfectly competitive (price-takers). All consumer decisions involving travel mode and the choice of a travel route on the road network are treated in TRAN, the transportation submodel. RELU and TRAN are linked sequentially but are iterated to a fully simultaneous equilibrium (see Anas and Liu [2007] for the algorithm). Consumers in RELU Consumers in RELU are adults, potentially active in the labor market. Each is either a whole or fractional household. Conclusions about households can be drawn only by pasting together the consumption or other decisions of consumers. Consumers are divided into four groups representing skill levels in the labor market that correspond to quartiles of the income distribution in the calibration of the model. Each consumer makes a set of simultaneously determined utility-maximizing decisions consisting of discrete and continuous choices. Consumers are myopic, spending the income of each period during that period, neither saving nor borrowing.

160

Alex Anas

Figure 6.2 Network of Major Roads in RELU-TRAN2

13

12

6

7

11

1

5

10

8

3

2 4 9

14

Source: Anas and Liu (2007).

The highest-level decision of a consumer is whether to enter the labor market or remain outside the labor market (voluntary unemployment). An unemployed consumer has an exogenous unearned income that is constant, increasing by skill level. The exogenous unearned income of an employed consumer is suppleFigure 6.2 mented by wage income. An unemployed consumer chooses a fictitious job location (zone 0) and incurs noLincoln_Ingram_Infrastructure commuting travel time or cost. Should wages increase (decrease), then consumers are more (less) likely to choose work, rather than unemployment. Both employed and unemployed consumers make shopping trips

the location effects of alternative road-pricing policies

161

to all model zones. The number of such trips depends on the gross-of-transport cost unit price, which attenuates with distance and congestion. Three discrete decisions are common to all consumers: 1. Job-residence location. Consumers choose any two of the MSA’s zones as a place of work and a place of residence. Each zone is an imperfect substitute in the labor and housing markets. Thus, each consumer has an idiosyncratic preference for each one of the job-residence pairs. Wages in each zone are determined by the skill level of the consumer (not by industry of employment). The choice of a residence-job location pair (i, j) by an employed consumer also determines the consumer’s commute, as will be discussed in more detail below. 2. Housing type. There are two housing types representing floor space in single-family housing or in a multiple-family housing structure. Housing choices are treated as renting. 3. Car type. Five discrete car types differ by fuel economy. Fuel-inefficient vehicles are larger, are more comfortable, and have higher acquisition and maintenance costs. The consumer’s utility function has a systematic preference that increases with the comfort, safety, and size of the vehicle and an idiosyncratic component for each car type. Thus, the choice of a car type involves a trade-off between the marginal utility of owning a larger and less-fuel-efficient vehicle and the higher acquisition, maintenance, and fuel costs for such a vehicle. Thus, less-fuel-efficient vehicles are on average owned by higher-skill and higher-income consumers with idiosyncratic variation within each skill-income group. Choice of continuous variables depends on the discrete choices (i, j, k, c), where i ! 1, . . . ,15 are zones of residence, j ! 1, . . . ,15 are zones of job location, k ! 1,2 are the housing types, and c ! 1, . . . ,5 are the car types. Thus, a working consumer faces 2,250 discrete bundles to choose from, whereas a nonworking consumer faces 150 discrete bundles. The conditional choices of the continuous variables depend on the discrete choices as follows: 1. Housing quantity: Given (i, k), how much housing floor space to rent. 2. Labor hours: Given (i, j), how many hours to supply at j. 3. Shopping trips: Given i, the quantity of retailed goods to buy at z ! 1, . . . ,14 and the number of trips required to make those purchases at fixed rates per unit of the good. Goods purchased at alternative locations are imperfect substitutes, and all locations are patronized because the consumer’s utility incorporates a taste for location variety in shopping. An important aspect for the consumer is the trade-off in the utility function between work, leisure, and travel. In the model, leisure is fixed and the remaining time is allocated between work and travel, including commuting (once per

162

Alex Anas

workday) and endogenous shopping trips. Time is valued at the wage rate since it is assumed that an extra hour of travel means that the consumer will have one hour less to earn wages. It is also assumed that commuting time creates some disutility. Thus, the marginal rate of substitution between disposable income and commuting time exceeds the wage. Formally, each consumer of skill/income f maximizes utility in the continuous variables Z ! [Z1, Z2, . . . ,Z14] and b; and the discrete bundles (i, j, k, c), where i is residence location, j is job location, k is housing, and c is car type: (1)

f ì é  f æ ö ï ê f ln êçå 









Z b1f exp 1f Gijcf 2f mc ïMax Uijkc f 






































































L z ijf ( z ) ÷ ijkc f  uijkc f "Z z , b ø ï êè"z ë ï ï sijf gizcf Z z bRik ïï subject to: å p z 























D j dgijcf z " Max í "(i , j, k, c) ï æ ö ï K(mc )  D j wjf ç H 
































 dGijcf å sijf Zz Gizcf÷ Mf ï










è ø "z ï ï ïand H D j dGijcf  å sijf ZzGizcf ³ 0 for j  0. ïî "z

(

(

)

Given are the prices of goods in z, pÂz; the rent of residential floor space, Rik; the wage rate, wjf ; nonwage income, Mf ; mode- and route-composite shopping travel times, Gizcf , and commuting times, Gijcf ; mode- and route-composite monetary costs of commuting and shopping trips, gijcf and gijzf ; the quantity shopped per trip, sijf ; the fuel inefficiencies (gallons per mile) of the available car types, mc; the annual time endowment available for work and travel, H; and the number of days per year, d, for which a commute is required. "ijkc#f are constant effects associated with the discrete choice bundle (i, j, k, c) and uijkc#f are the idiosyncratic tastes. z#ijf are constant effects that reflect the attractiveness of a retail location z to consumers of type f located at residence-job locations i, j. $f in the CES subutility defined over retail locations is related to the elasticity of substitution among the retail locations, and %f is the share of the disposable income spent on purchasing retailed goods, and 1 & %f the share that will be spent on renting housing. '1f is the marginal disutility of commuting time, and '2f the marginal utility of a larger, safer, but less-fuel-efficient car. The right side of the budget constraint is the money income of the consumer who is paid a wage per hour of labor supplied after all travel time (for commuting plus shopping). If the consumer chooses not to work by choosing j ! 0 in the outer stage, then (j ! 0, and the consumer has no wage income. Otherwise, for any j ) 0, (j ! 1. The left side of the budget is the monetary expenditure on retail goods, commuting and housing

ùü úï úï úï ûï ï ïï ý ï ï ï ï ï ï ïþ

)

the location effects of alternative road-pricing policies

163

space, and annual car-ownership costs, K(mc). The prices of the retail goods are the prices at the retail location plus the monetary cost of the travel from home to the retail location. In the inner stage (inside { }), given the discrete choice bundle (i, j, k, c) determined at the outer stage, the consumer chooses the optimal quantities of the retailed composite goods to shop from each retail location z, (vector Z ! [Z1, Z2, . . . ,Z14]); and the residential floor space b to rent. This gives Marshallian demands Z*ijkc"f and b*ijkc"f . At the outer stage, the consumer chooses the most preferred (i, j, k, c), given the indirect utility function U* # uijkc"f from the inner ijkc"f stage. The discrete choice probabilities have the nested-logit structure, where a marginal probability describes the binary choice of entering the labor market versus not participating in the labor market. The conditional multinomial logit probability, P* , describes the distribution of employed consumers of type f i,j>0,kc"f among the bundles (i, j $ 0, k, c). RELU connects with TRAN via the mode- and route-composite trip times and monetary costs, which are the matrices éëGijc f ùû , éë gijc f ùû. RELU-TRAN2 does not treat traffic congestion by time of day, so all who use a road experience the same congestion. The monetary cost, on the other hand, does depend on car type since gasoline consumption depends on traffic speed determined by congestion and since car type is a discrete choice that depends on car acquisition and operating costs and on car preferences, which vary with income. Consumers in TRAN In the TRAN submodel, each consumer chooses the mode of travel for each trip and the routing of that trip over the road network if the mode is car. 1. Mode choice. For each residence-job-car bundle (i, j, c), the consumer of type f chooses a travel mode for each trip (whether for commuting or for shopping) that is determined in RELU. There are three modes of travel: m ! 1 (car), m ! 2 (public transit), and m ! 3 (nonmotorized). The third applies largely to intrazonal trips, especially in the suburbs. When the choice is car, it is assumed that the chosen car type, c, is used. Systematic and idiosyncratic generalized costs are treated in the choice of mode. 2. Route choice. For car trips, the consumer chooses the route from triporigin zone i to trip-destination zone j with the minimum round-trip generalized cost over the road network. As in mode choice, the systematic and idiosyncratic generalized costs of the available routes are considered. The consumer takes as given the speed of travel on each road link on that route since speed is determined by traffic congestion. As congestion increases, traffic slows down. The speed and time on each link is endogenously determined at equilibrium. All car types are assumed to cause the same congestion on one another. The generalized cost of travel on a link is a weighted sum of the monetary cost and the value of travel time.

164

Alex Anas

This value of time is exogenous and increasing by skill-income group. The monetary cost depends on vehicle type (fuel economy) and on the cost of gasoline. Figure 6.3 plots the U-shaped speed versus fuel consumption curves based by smoothing those estimated by Davis and Diegel (2004) for nine actual car models. These relationships were obtained by fitting a polynomial curve to the Geo Prizm and then multiplicatively shifting this polynomial. Consumers determine their monetary expenditure on operating a car by choosing their car type in RELU (as we saw) and by choosing routes that are faster or slower in TRAN. Consumers with lower (higher) values of time are more likely to prefer cheaper (faster) routes, and this, together with their preference for car size and the level of caracquisition costs relative to their income, determines fuel economy and gasoline consumption. The gallons/mile versus miles/hour U-shaped polynomial curve is f(s)mc, where: (2)

f(s) ! 0.12262 " 1.172 # s $ 6.413 # 10–4s2 " 1.8732 # 10–5s3 $ 3.0 # 10–7s4 " 2.472 # 10–9s5 $ 8.233 # 10–12s6

Figure 6.3 Fuel Intensity Versus Speed

Gasoline consumption (gallons/mile)

0.12 0.10 0.08 0.06 0.04 0.02 0 5

10

15

20

25

30

35

40

45

50

55

60

65

70

Speed (mph) m1 = 0.85

m2 = 1

m3 = 1.15

Source: Anas and Hiramatsu (2012a).

Figure 6.3

m4 = 1.3

m5 = 1.45

75

80

the location effects of alternative road-pricing policies

165

pFf(s)mcd is the fuel cost of driving a road distance d at speed s using a car of fuel efficiency level mc when the price of a gallon of fuel is pF. The speed is 1 , where d is the road distance and Time the congested time to travel one s= T ime mile. Time is given by a Bureau of Public Roads (BPR)-type congestion function c2 æ æ Flow ö ö . Flow is the aggregate volume of traffic on the road, Time = c0 ç 1 + c1 ç ÷ è CAP ø ÷ø è and CAP is the road’s capacity (constant all along the road). The generalized ædö cost of traveling a road of length d is gcos t fc = (votf ) ç ÷ + pF f(s)mc d, where èsø vot f is the value of time in route choice that depends on the consumer’s income, indicated by f. Firms RELU includes four industries: (1) agriculture; (2) manufacturing; (3) business services; and (4) retail. Production functions are constant returns, and all firms producing in the same zone and industry are perfectly competitive profit maximizers in input and output markets, charging the same price and paying the same wages and rents. Goods in the same industry produced in different zones are variants of the same good. As explained earlier, consumers buy only the retail good by shopping it in every zone. All location variants of a good are also used as intermediate inputs in the production of the other goods except for the retail good, which is produced by the input of the other goods but is not itself an input in the production of other goods. In addition, each industry uses primary inputs, which are business capital, space in commercial and industrial buildings, and labor from each of the skill groups (income quartiles) of the working consumers. All outputs can be exported to other regions from any zone where they are produced. Developers The treatment of developer behavior in this model is based on Anas and Arnott (1991, 1993, 1997). Developers are agents who incorporate the activities of landlords, who rent out floor space and collect rents on it; investors, who buy and sell real estate; and contractors, who construct or demolish. Unlike the model’s firms and consumers, who are myopic, developers operate with perfect foresight and are risk-neutral profit maximizers. In this chapter, the model is implemented as a stationary-state or long-run equilibrium model, and developers therefore operate with perfect foresight of this stationary state. Time is in discrete periods of five years in duration. There are no transaction costs in buying and selling. In the beginning of each period, a developer is the owner of vacant land or of residential or commercial or industrial buildings. Developers in the same zone who own vacant land face a common cost of construction but are horizontally differentiated by idiosyncratic costs. The idiosyncratic cost draw of each developer for constructing each type of building and for just keeping the land vacant is determined toward the end of each period.

166

Alex Anas

When these costs are determined, the developer decides whether to continue to hold the land vacant or to construct a particular building type, given the construction cost per square foot of floor space. At the beginning of the period, when the uncertainty has not been resolved, the developer values the vacant land asset at the expected maximum profit the land would fetch from the most profitable construction or from doing nothing at the end of the period. Similarly, developers who start the period owning a particular type of building decide whether to demolish it at the end of the period, while in the beginning of the period they value the building asset knowing only the expected value of the profitmaximizing action. Since developers are perfectly competitive, asset prices for vacant land and for each type of building are determined in the beginning of each period. Since the developers’ behavior is assumed to be stationary in the aggregate in each zone and for each type of building and vacant land, the asset prices for building and land make all expected economic profits zero so that developers earn only normal profits, while stocks, rents, and values are stationary because the construction flow of the floor space of each building type equals the demolition flow of the floor space of the same building type. An exogenous change would alter the long-run equilibrium stocks that prevailed but would also change the rates of demolition and construction necessary to maintain the stocks at a stationary level (Anas and Arnott 1993). MODEL STRUCTURE: GENERAL EQUILIBRIUM

The model includes four markets: (1) the labor market for each labor skill level in each zone (56 equations of 14 zones by 4 skill levels); (2) the rental market for each residential building type (single-family and multiple-family) in each zone (28 equations of 14 zones by 2 housing types); (3) the business rental market for commercial and industrial buildings (28 equations of 14 zones by 2 building types); and (4) the goods markets for each industry and zone (56 equations of 14 zones by 4 industries). Solving these equations determines the rental prices per square foot, the hourly wages for each skill level, and the output prices for each industry. Real estate values are then calculated from rents and construction costs, and the stocks of each building type in each zone are adjusted to the new equilibrium. CALIBRATION OF THE MODEL

The model’s calibration is evaluated by key elasticity measures and the marginal rate of substitution between commuting time and disposable income. The values of these relationships are for the year 2000 Chicago MSA data and are shown in table 6.1. It is important to put these numbers in the context of the literature, where the same relationships have been estimated by others. The elasticity of location demand with respect to commuting time was estimated in the 1970s by Charles River Associates (1972), Lerman (1977), Atherton, Suhrbier, and Jessiman (1975), and Train (1976). A survey of the literature,

Table 6.1 Calibrated Elasticities in RELU-TRAN2 (Chicago MSA) Consumers

Income Quartiles 1

MRS (disposable income, commute time), ($/hour/day) Elasticity of location demand with respect to commuting time Elasticity of housing demand with respect to rent Elasticity of labor supply with respect to wage

2

3

4

12.295

21.056

36.204

93.215

−0.619

−0.602

−0.607

−0.544

−1.95 3.83

−1.76 2.93

−1.57 2.1

−1.38 1.32

Developers

Elasticity of floor space supply with respect to rent (short-run) Elasticity of construction flow with respect to asset value Overall City Suburbs Elasticity of demolition flow with respect to asset value Overall City Suburbs Elasticity of floor space stock with respect to asset value Overall City Suburbs Driving Gasoline consumption (CO2 emissions) with respect to fuel price (base fuel price is $1.90) VMT with respect to fuel price MPG with respect to fuel price

Building Type 1 Single family

2 Multifamily

0.0991

0.23

0.268

0.138

0.0521 0.0335 0.0526

0.421 0.0564 0.681

0.420 0.261 0.452

0.0744 0.0396 0.0785

−1.612 −0.0550 −1.719

0.0535 0.00102 0.0672

−0.982 −0.528 −1.375

0.0147 0.0068 0.0218

3 Commercial

−0.176 −0.346 −0.073

0.00542 0.00643 0.00480

4 Industrial

−0.523 −0.667 −0.465

0.00872 0.00786 0.00922

−0.0899 −0.0721 −0.0180

167

168

Alex Anas

which includes their own estimates, is given by Anas and Chu (1984). They reported: The in-vehicle time elasticity ranges from 0.36 to 1.40 for transit and from 0.55 to 1.77 for the drive-alone mode. Out-of-vehicle time elasticities range from 0.23 to 2.7 for transit and are 0.42 in the CSI model. Train and CRA do not report out-of-vehicle time elasticities for the auto mode.

As shown in table 6.1, the workers’ travel time elasticity of location demand in RELU-TRAN2 ranges from 0.544 to 0.619 and is in the range of the above estimates. It is reported in Anas and Arnott (1993) that the average rent elasticity of housing demand, the rent elasticity of white households, and the rent elasticity of nonwhite households in the Chicago MSA for 1970 to 1980 are 0.554, 0.516, and 0.683, respectively. In our model, the rent elasticity of housing demand cannot be larger than 1 because of the functional form of the utility function, and it ranges from 1.38 to 1.95. Our elasticity combines two aspects of the demand for housing: the demand for housing size as floor space, which has elasticity of 1, and the number of consumers who demand housing at a particular location, which has elasticity that ranges from 0.38 to 0.95. Housing demand at a particular location is the product of these two quantities. Thus, our elasticity is higher than that in Anas and Arnott (1993), who estimate a model in which the housing size effect is fixed. Kimmel and Kniesner (1998) studied U.S. household data for the period from 1983 to 1986. Their wage elasticity of labor supply (hours worked) is !0.51. In our model, the consumer makes more nonwork trips when the wage increases (because of the income effect for shopping normal goods), and this reduces the labor supply. In Anas and Arnott (1993), the elasticity of housing floor space supply with respect to rent is !0.1016 and !0.1136 for single-family and multifamily housing, respectively. In our model, the corresponding values are !0.0991 and !0.23. Thus, the elasticity of our single-family housing is similar to theirs, but our multifamily housing supply is more elastic than theirs. This elasticity measures the percentage of existing housing stock that will be put on the market to be rented (rather than being kept vacant) by the landlords. Our !0.23 estimate for multifamily housing is almost the same as that reported by Anas (1982) for the Chicago MSA using 1970 data. DiPasquale and Wheaton (1994) report that the long-run price elasticity of the aggregate housing stock is in the !1.2 to !1.4 range. Blackley (1999) reports that the construction elasticity ranges from !1.0 to !1.2 and that the long-run price elasticity of new housing supply (supply measured in value terms) in the United States for 1950 to 1994 ranges from !1.6 to !3.7. Green, Malpezzi, and Mayo (2005) report a price elasticity of housing supply in the Chicago MSA for

the location effects of alternative road-pricing policies

169

the period from 1979 to 1996 as !2.48, but their estimate is not significantly different from zero. Their housing supply is defined as the number of housing units for which building permits were issued, multiplied by 2.5 (the average household size), divided by the population. Our elasticity of housing construction measures what percentage of the land available for construction will be developed into type k building (housing) if the asset price of type k building rises. This elasticity ranges from !0.03 (for single-family housing in the city) to !0.68 (for multifamily housing in the suburbs). There are a few reasons why our elasticity of construction is so small. First, many of our modeled zones are urbanized and there is not much land left to be developed. The area covered by the Chicago MSA in Green, Malpezzi, and Mayo (2005) is broader than in our modeled zones. Second, by the year 2000, our modeled zones had become more developed than they were during their period, and the available land had decreased significantly. Finally, the definition of our elasticity of construction is different from theirs because they measure how much an increase in asset price would increase building permits multiplied by the population that would use the newly constructed housing, whereas our elasticity measures the percentage by which the developed land would increase. Two additional assumptions could be affecting our elasticity in real estate variables. First, our building structural density (in floor space per unit of land) is constant by building type and zone. However, average structural density in our model zones is not constant and can change over time by demolishing low structural density buildings and constructing higher structural density buildings, for example. If the developer could directly choose the building’s floor space amount, the stock could be more elastic when the building value increases. This would be especially true in the zones where vacant land is scarce. Smith (1976) reports that the price elasticity of density is !5.27, where Smith’s density is the number of dwelling units built on a unit land area, from Chicago MSA cross-section data between 1971 and 1972. The second assumption that could be affecting our low elasticity of stock is the condition that the construction and demolition flow of each building stock in each zone is equalized by the real estate market being in stationary equilibrium. In reality, the construction flow would be larger than demolition and stock in a growing economy. The above discussion suggests that the methodology used in the literature to estimate the supply elasticity of housing is not robust. There are important data-driven or definitional differences between any two studies. Hence, it might be better to evaluate the reasonableness of our housing supply elasticity by actually simulating the model in a comparative statics exercise and observing how the housing stock responds in quantity. In such a comparative statics exercise, Hiramatsu (2010) simulated a simple urban growth scenario in which he increased the total population and the net exports by 10 percent. The vacant land stock decreases in both the city and the suburbs. The single-family housing stock decreases in the city and increases in the suburbs. The multifamily housing stock

170

Alex Anas

increases in both the city and the suburbs, increasing more in the suburbs than in the city. Both single-family and multifamily housing stocks increase by less than the 10 percent population growth, and the average floor space per person decreases. The industrial and commercial buildings also increase in the city and in the suburbs. The rate of increase is more in the city than in the suburbs but is not as high as the rate of increase of the housing stock. In the city, where the available land is limited, some single-family housing is demolished and multifamily housing, industrial buildings, and commercial buildings are constructed. In the suburbs, where there is plenty of land, both single- and multifamily housing is constructed, as well as industrial and commercial buildings. Thus, the building stocks respond reasonably with respect to the increase of the population and net exports. Accordingly, the rents and values of each building type change in a normal way. In the city, the rent of single-family housing increases by more than 10 percent because the supply decreases. The other building rents also increase since demand increases by more than supply does. Both rent and value increase more for those building types and locations where the demand increases more and the supply increases less. In this way, we conclude that the building markets—including stocks, rents, and values—respond reasonably under the calibrated elasticities of the model.

Road-Pricing Policies: Congestion Tolls and Fuel Taxes The model calculates two externalities of traffic congestion. One is the delay caused by the volume of traffic on each road. The other is the excess fuel consumption induced by the traffic: when traffic moves more slowly, vehicles consume more gasoline per mile, as shown in figure 6.3. These two externalities are calculated on each major (interzonal) and local (intrazonal) road, but the model does not distinguish between different times of the day, thus implying that all the travel occurs over a relatively wide rush hour. The policies examined in this chapter directly or indirectly target these two congestion externalities caused by driving. The following alternative policies are considered: t

t

A quasi-Pigouvian congestion toll that varies by type of road and is charged on each road link. There are two versions of this: QP1, under which only the major (interzonal) roads are tolled and local (intrazonal) roads remain untolled, and QP2, under which all roads (interzonal and intrazonal) are tolled. A per-gallon fuel tax, the rate of which is calculated so that the aggregate fuel tax revenues match the revenues of QP1 or QP2.

QUASI-PIGOUVIAN TOLLS

In theory, first-best Pigouvian tolling would perfectly internalize both externalities over the entire network. The first-best Pigouvian tolls measure the excess

the location effects of alternative road-pricing policies

171

time delay plus the excess fuel consumption imposed by each car trip on all other car trips. This chapter refers to tolls as quasi-Pigouvian because they deviate in three ways from first-best Pigouvian tolls, which would be very difficult to implement in reality. First, every mile of road is shared by travelers with different values of time. The first-best Pigouvian toll would be calculated by multiplying the marginal time delay experienced by each traveler on each road by the traveler’s marginal rate of substitution between travel time and disposable income and then adding these up over all travelers on the road. Instead, we assume that the road authorities know only the average value of time of the drivers on each road, which is exogenously given according to the income level of the traveler. The second reason why congestion tolls in RELU are quasi-Pigouvian is that consumers can save fuel not only by traveling faster (see figure 6.3), but also by switching to vehicles with higher fuel economy. The first-best policy might vary the part of the Pigouvian toll aimed to capture the fuel externality, not only according to route, but also according to the car types on the road. We assume that road authorities know only the average car on each road and set a toll that is common to all vehicles. The third and final reason is that RELU-TRAN2 treats heterogeneity among consumers, and when such heterogeneity is present, toll revenue should in general be distributed unequally among the consumers. However, doing so would be difficult in practice since road authorities would need to know how the marginal utility of income varied in the driver population. RELU-TRAN2 assumes that toll revenue is equally distributed among all consumers, including nondrivers. FUEL TAXES

The fuel tax also acts globally over the entire network, but it is a lower-best instrument since it targets only fuel consumption, thus working on the congestion only indirectly. In fact, the fuel tax is, a priori, a crude instrument because it is paid for the fuel consumed on each mile of road regardless of the congestion level on the road. Although figure 6.3 shows that fuel consumption indeed rises with congestion (that is, with lower traffic speed), the fuel tax would be paid even on a road with zero congestion. The fuel tax is very easy to implement since all car traffic pays the same fuel tax per gallon of gasoline. Cars with lower fuel economy consume more gasoline and pay higher fuel taxes. Thus, on the one hand, the fuel tax creates an incentive for using vehicles that have higher fuel efficiency. On the other hand, the fuel tax may do a poor job of internalizing the delay externality of congestion. It affects congestion only indirectly by raising the fuel cost of travel and thus reducing travel volume and improving speed. In contrast, our quasi-Pigouvian toll is directly proportional to the delay caused by congestion and reduces the timedelay externality more efficiently by differentially pricing the externality on each road.

172

Alex Anas

TAX-AVOIDANCE BEHAVIOR UNDER THE POLICIES

In our general equilibrium model, the effects of the policies will differ according to the way the market agents (consumers and firms) exercise tax-avoidance behavior directly or become influenced by changing travel times, prices, rents, and wages indirectly. Since the model entails many margins of adjustment, the overall effects are complex and require netting out the various changes across all margins. The most immediate form of adjustment would be in the choice of route. For example, a commuter who passes through the CBD could be induced to travel around it to avoid the higher tolls that would prevail on the highly congested roads terminating in the CBD. As many travelers do this, roads circumventing the CBD would become more congested, and the roads going into the CBD would become less congested. Another example is that a quasi-Pigouvian congestion toll would increase the monetary cost of travel, inducing consumers with low values of time to choose longer but less congested routes with lower tolls. Commuters with higher time values would prefer to pay higher tolls and travel on the faster routes. These adjustments would not work as well under fuel taxation; in that case, fuel taxes would be more correlated with distance traveled than with congestion. Hence, under fuel taxation, shortening the distance traveled would be a more dominant response. A second margin of adjustment concerns the fuel efficiency of the car. The higher monetary cost of the fuel tax, for example, would induce consumers to switch to more-fuel-efficient cars. This effect, however, is small (Anas and Hiramatsu 2012b). A third margin of adjustment entails switching between car and mass transit. Higher tolls or taxes would induce consumers with lower values of time to switch to the slower but cheaper transit mode. As tolls or fuel taxes reduce congestion and speed up driving, some consumers with high values of time would switch from transit to car. A fourth margin of adjustment would be to change the destination, number, and length of nonwork trips from the locations that involved a high tax or toll layout to other locations that involved less. All of these effects are treated in the model. Changing job or residence locations requires longer-term adjustments. Some examples of residence location changes would be for a worker who commutes into the congested CBD to move his residence into the CBD, reducing housing size at the same time in response to the higher CBD rents. Such a choice would be favored by workers who dislike transit or who reside in suburban areas in which transit is inaccessible. Others may indeed switch to transit, but to do so they may have to move from the suburbs to the city, where transit is more easily accessed. Still others may reject these options and prefer to switch to a suburban job from one in the CBD. Firms, meanwhile, would also respond to tolls or taxes. For example, a firm located inside the CBD that employs many employees who drive into the CBD

the location effects of alternative road-pricing policies

173

but dislike switching to transit or moving their residences into the CBD faces a choice: pay higher wages to induce employees to keep their CBD jobs or relocate outside the CBD to lower the tolls and taxes that employees incur. However, the CBD may attract more firms if enough consumers are willing to locate their residences within it or to switch to transit and if such shifts increased the supply of labor within the CBD enough to lower wages. Such shifts could also induce developers to replace commercial real estate with residential housing. In realistic schemes, only major roads may be proposed for tolling. If the quasi-Pigouvian toll is levied on major roads only, the differences between quasiPigouvian tolling and gasoline taxation are magnified because drivers on local roads (i.e., those traveling intrazonally) would not be charged under quasiPigouvian tolling but would pay the fuel tax. Under such quasi-Pigouvian tolling, interzonal trips and congestion would decrease while intrazonal trips and congestion would increase as consumers and firms relocate to avoid using the major roads and rely more on local roads. The quasi-Pigouvian toll paid will be higher than the fuel tax on highly congested roads, while the fuel tax paid would be higher on the less congested roads. Hence, drivers would feel that the fuel tax is too high on long-distance and slower routes since fuel consumption increases with distance and falls with speed. Because the fuel tax affects all roads, it would not be very helpful for drivers to make detours.

The Impacts of the Policies The salient results of the road-pricing policy simulations are presented in tables 6.2 to 6.4. Table 6.2 shows the effects of the policies on driving-related aggregates such as fuel consumption and CO2, vehicle miles traveled (VMT), gallons per mile (GPM), and total travel time. Table 6.3 juxtaposes the effects of the policies on the distribution of jobs and residences and on land development Table 6.2 Percent Changes in Driving-Related Aggregates Under Road-Pricing Policies

Gasoline and CO2 Vehicle miles traveled (VMT) Gallons per mile (GPM) Total travel time Total travel monetary cost (including tolls or taxes)

Quasi-Pigouvian Toll, QP1, on Major Roads

RevenueNeutral Fuel Tax at 55.5%

Quasi-Pigouvian Toll, QP2, on All Roads

RevenueNeutral Fuel Tax at 287%

−4.64 −3.93 −0.67 −2.45

−2.65 −2.11 −0.56 −1.30

−12.52 −9.93 −2.66 −5.34

−13.35 −11.10 −2.40 −5.45

23.96

24.61

114.65

112.91

174

Alex Anas

Table 6.3 Effects of the Pricing Policies on Jobs, Residences, and Undeveloped Land Changes in:

Location

Jobs

CBD City ex-CBD Inner suburbs Outer suburbs Total

Residences

Undeveloped Land (sq. feet and % changes)

Base Level

QuasiPigouvian Toll, QP1, on Major Roads

RevenueNeutral Fuel Tax at 55.5%

QuasiPigouvian Toll, QP2, on All Roads

RevenueNeutral Fuel Tax at 287%

537,861 793,798 1,720,045 693,578

+747 −4,058 −5,027 +9,475 +1,137

+2,385 +2,314 −2,526 −1,759 +414

+6,215 +2,275 −12,385 +5,064 +1,169

+10,987 +10,713 −15,081 −5,198 +1,621

CBD City ex-CBD Inner suburbs Outer suburbs

39,688 1,413,312 2,157,789 1,080,057

+535 −5,940 −10,768 +16,174

+781 +8,117 −3,713 −5,184

+3,237 +14,272 −27,253 +9,745

+3,755 +40,296 −28,106 −15,944

CBD City ex-CBD Inner suburbs Outer suburbs

79,357,608 664,730,392 6,368,076,040 44,984,268,688

−4.50% −2.68% −7.92% −1.90%

−1.94% −1.24% −3.35% −0.67%

−6.99% −4.23% −10.72% −2.63%

−7.92% −5.06% −12.25% −2.64%

by geographic ring within the MSA: the CBD, the rest of the city of Chicago, the inner suburbs, and the outer suburbs. Table 6.4 shows how consumer utility, revenue from the policy, real estate values, wages, and rents change under the alternative policies. REVENUE AND WELFARE

A first observation from these tables (see table 6.4) is that the revenue raised by the quasi-Pigouvian tolling of all roads (QP2) is 4.6 times the revenue raised from the tolling of the major roads only. The revenue-neutral per-gallon fuel tax rate that corresponds to QP1 is 55.5 percent and that which corresponds to QP2 is 287 percent. The former increases the after-tax gasoline price by about one-half, while the latter would almost quadruple it. Next (also from table 6.4), the total welfare change is positive. The average tax revenue change per consumer is redistributed back equally among the consumers. Hence, the welfare per consumer consists of two parts: (1) the change in compensating variation (CV), which measures how much the average consumer would be willing to pay to accept the policy; plus (2) the average annualized

the location effects of alternative road-pricing policies

175

Table 6.4 Changes in Welfare Components, Wages, and Rents Under Pricing Policies

Revenue/consumer ($/year) (a) CVa per consumer ($/year)

QuasiPigouvian Toll, QP1, on Major Roads

RevenueNeutral Fuel Tax at 55.5%

284

284

1,306

1,306

244 1,149

14 485

264 1,701

97 1,967

499

1,965

2,064

90.41 107.16 136.27 351.02

379.86 377.49 516.02 1,741.03

405.11 427.65 515.81 1,440.95

−273.19 −523.95 −761.93 −1,801.55

−616.18 −1,503.94 −2,327.90 −5,950.33

−900.41 −1,914.12 −2,873.74 −7,056.66

(b) Annualized real estate income per consumer ($/year) Total welfare/consumer ($/year) = (a)+(b) 1,393 CV/consumer of workers by income level ($/year) 349.22 1 410.85 2 550.22 3 1,388.13 4 CV/consumer of nonworkers by income level ($/year) −902.78 1 −1,416.71 2 −1,938.02 3 −4,189.55 4 Change in average wages Change in average rents by building type Single-family homes Apartments Commercial Industrial a

QuasiPigouvian Toll, QP2, on All Roads

RevenueNeutral Fuel Tax at 287%

+8.6%

+3.4%

+11.14%

+13.03%

+4.55% +3.62% +5.97% +5.78%

+1.97% +1.66% +2.49% +2.45%

+6.72% +5.50% +8.49% +8.30%

+7.57% +6.41% +9.67% +9.50%

CV = compensating variation.

change in real estate values per consumer. Real estate developers make zero expected profits, as explained earlier. However, the introduction of a road-pricing policy causes the holders of land and buildings in the pre-tax equilibrium to experience windfall gains and losses in the values of their assets. An annualized income stream is calculated from these aggregated net gains, but this is not redistributed to the consumers. Thus, implicitly, all consumers are treated as renters, and all real estate asset owners are treated as absentee. Welfare gains therefore consist of the two parts above, which are aggregated.

176

Alex Anas

The welfare change numbers of table 6.4 suggest several observations. One is that consumers are better off under Pigouvian tolling than under the equivalent fuel taxation. Tolling removes the negative externality where it is present, but fuel taxation removes it imperfectly while also inefficiently penalizing those who create little or no congestion (and overpenalizing them for the pollution they create).2 EQUITY UNDER THE ALTERNATIVE POLICIES

Table 6.4 also shows how CV gains or losses are distributed among the various consumer groups. In the model’s baseline, four income groups correspond roughly to the quartiles of the 2000 personal income distribution. In each income group, the model divides the population endogenously into consumers who are and are not working. Employed consumers experience a disutility from commuting time and forego wage income when allocating more time to work and nonwork travel. The marginal rate of substitution between commuting time and the disposable income allocated to buying goods and services increases with the income of the employed consumer. Nonworking consumers in the model have a low value of time from traveling for nonwork trips. A property of consumer behavior in the model is that nonwork trips are made to acquire goods and services, which are normal goods. Therefore, richer consumers make more nonwork trips, and this holds true for both working and nonworking consumers. Then, road pricing reduces the CV of a richer nonworking consumer by more since such a consumer makes more nonwork trips but cannot allocate time saved from less congestion to earn more income. Table 6.4 shows that, among consumers who are employed, the CV gain increases with income since time saved is valued more the higher the income/wage of the working consumer. Among those who are not working, the low values of time but the higher monetary cost of travel after road pricing cause the opposite result: CV is negative and becomes more negative with income since the higher the income the greater the number of trips made for shopping normal goods and, therefore, the higher the exposure to the higher monetary cost of travel under tolling or fuel taxes. Note that wages are endogenous in the model and increase under all scenarios (the reasons for which will be discussed later), and this causes an increase in the value of time in RELU, which affects location decisions and trip making. AGGREGATES RELATED TO DRIVING

Not surprisingly, fuel and emissions of CO2, vehicle miles traveled, gallons per mile, and total travel time all decrease as the monetary cost of travel, including

2. Note, however, that the total welfare increase under the gas tax is 2,064 per consumer, which is larger than the corresponding 1,965 under QP2. This is a minor anomaly due to the fact that our realistic congestion tolls are quasi-Pigouvian and not first-best for the reasons explained earlier.

the location effects of alternative road-pricing policies

177

tolls or fuel taxes, increases under each policy (see table 6.2). The 55.5 percent increase in the cost of fuel causes travel monetary cost to increase by 24.61 percent, and the 287 percent increase in the cost of fuel causes travel monetary cost to increase by 112.91 percent. These percentage increases are similar to those that occur under quasi-Pigouvian tolling. Importantly, the percentage increases in monetary cost are less than half of the percentage increases in the cost of fuel, which points to the adjustments consumers and firms undertake to blunt the impact on their budgets of the roadpricing policies. These adjustments may be grouped into two broad categories: switching to transit, which is the biggest effect (transit ridership increases by 13 percent or more), and making fewer and shorter car trips. There are, of course, rebound effects in fuel, CO2, VMT, GPM, and travel time induced by reduced congestion, which in turn is caused by fewer and shorter car trips. CENTRALIZATION OR DECENTRALIZATION OF JOBS AND RESIDENCES

Table 6.3 shows how the spatial distribution of jobs and residences and of undeveloped land changes under each policy. In the model, the total number of consumers is fixed, and they may choose whether to work. Therefore, in addition to changes in job locations, the model also indicates whether a particular policy increases or decreases the number of consumers in the labor force. That is why the positive and negative job changes do not sum to zero. Note, however, that all consumers have housing whether they are in the labor force or not. Therefore, the consumer increases and decreases by residential location do sum to zero (net of rounding). Why does the total number of jobs increase (though very slightly) under each policy? It is directly related to the result in table 6.4 that shows an increase in wages. While the increase in wages is explained below, note for now that the higher wages cause some residents who are initially not in the labor force to choose to enter the labor force (the extensive margin of labor supply). Several additional results are seen from a systematic examination of table 6.3. Under the quasi-Pigouvian tolling of the major roads only (QP1), jobs and residences move similarly, decreasing in the city of Chicago ex-CBD and the inner suburbs and increasing in the outer suburbs and only slightly in the CBD. Since only major roads are taxed, two important toll-avoidance margins are at work. One of these is that some residents who previously drove on major roads now switch to mass transit, and doing so may entail moving their residences to the city, where transit is more available than in the suburbs. The other margin is that some outer suburban residents who commuted downtown from the suburbs by car and who would thus be greatly affected by the tolls on major roads, now want to work in the suburbs, preferably in their zone of residence. Doing so, they avoid driving on the major roads and paying the tolls. This creates an abundance of labor supply in the outer suburbs, and firms from the city are then attracted to relocating to the suburbs.

178

Alex Anas

Next, also from table 6.3, consider the effects of the fuel tax that achieves revenue neutrality with QP1. Because the fuel tax is paid by all car travel, whether on major roads or not, it is much harder to avoid using the second margin of intrazonal location of job and residence that was important under QP1. Under the fuel tax that is 1.5 times higher, the margin of switching to transit becomes much more important, and there is a strong trend toward moving residences from the inner and outer suburbs to the city, including the CBD. Of course, some consumers prefer to continue driving but to shorten the length of their trips, and this means that some of those who worked in the city but resided in the suburbs would move to the city. Now, still from table 6.3, look at QP2, the quasi-Pigouvian tolling of all roads (major and local), and its revenue-neutral fuel tax. Note that comparing QP2 tolling and QP1 tolling, the job and residence changes are qualitatively similar but quantitatively different. Under QP2, as under QP1, jobs leave the inner suburbs and increase rather significantly in the city ex-CBD and CBD but increase less in the outer suburbs than they did under QP1. The suburbanization effect is weaker and the centralization effect is stronger under QP2 because the margin of intrazonal location is less effective under QP2 since intrazonal as well as interzonal roads are tolled. So the margin of central relocation to make better use of transit is relatively stronger. Under the QP2 revenue-neutral fuel tax, the effects are qualitatively identical to those under the QP1 neutral fuel tax but larger in magnitude, and the reason is simply that the fuel tax is a lot higher under the QP2 revenue-neutral scenario. WAGES AND RENTS

As shown in table 6.4, the average Chicago MSA wages and rents increase by significant percentages under each of the policies. Furthermore, wages increase by a higher percentage than rents do. The wage increase results from the need of most firms to entice their workers to continue commuting to the same jobs despite the higher monetary cost of transportation caused by the tolls or the fuel taxes. At the margin, firms and consumers adjust by relocating closer to each other, though infra-marginally most firms and consumers stay put. In the new equilibrium—after the fuel tax or the tolls—firms must pay higher wages. Rents increase primarily because of two effects, one that operates in the floor space for business use and the other in residential demand for housing. In the case of the business, labor and building space are substitutes in the production functions. As labor becomes more expensive (wages increase), its substitute also becomes more expensive. In the residential case, consumers want to live closer to their jobs and to shops at the margin. This intensifies the demand for housing, causing housing rents to increase. Meanwhile, the higher wages also have an income effect that operates in the housing market, by raising the demand for housing, a normal good.

the location effects of alternative road-pricing policies

179

The higher rents cause higher prices for each type of real estate floor space (since real estate prices are the discounted sum of rents plus expected capital gains). These higher floor space prices cause real estate construction that expands each type of developed stock and depletes some of the initially undeveloped land (table 6.3), increasing, at the margin, both infill development in the CBD, city, and inner suburbs and sprawl development in the outer suburbs.

Conclusions and Extensions This chapter showed that road-pricing policies—and especially those applied broadly, such as fuel taxation or the Pigouvian tolling of all roads—would indeed cause the centralization of jobs and residences to the city of Chicago from the suburbs. This result, however, does come with some important qualifications. One is that Pigouvian tolling of only the major roads could very well cause centralization of jobs and residences to the CBD or their decentralization to the outer suburbs. Cordon tolling was not discussed in this chapter but has been examined in greater detail in Anas and Hiramatsu (2012a). That article looked at the location of three cordons for Chicago and calculated the optimal cordon toll level for each cordon location.3 By consulting that article as a companion to this chapter, we learn that London-type cordons that circumscribe the CBD and Stockholm-type cordons that circumscribe a much larger area that includes most or all of the inner city cause jobs and residences to move out of the cordon. This reduces real economic output inside the cordon, increasing it outside the cordon. The combined nominal output, however, increases for the MSA. A counterfactual outer cordon that circumscribes the perimeter of the outer suburbs has the opposite effect. Cordon toll avoidance in this case concentrates residences and jobs within the cordoned area. Real output increases within and decreases outside the cordon, while aggregate nominal output again increases. The cordon policies are not nearly as efficient as fuel taxes or congestion tolls in correcting the externalities of congestion, and the CBD and inner-city cordons captured about 65 percent of the total welfare gains of quasi-Pigouvian tolling. The outer cordon was the least efficient, capturing about half of the efficiency gains of the narrower cordons. But planners interested in the economic revitalization of the central areas would find a meaningful trade-off between the location effects and the efficiency gains of this outer cordon. Similarly, the simulations presented in this chapter suggest that planners interested in the vitalization of the central city should favor the gasoline tax over Pigouvian tolling since the gasoline tax caused greater centralization of jobs and residences.

3. A toll is paid every time a car crosses the cordon in the inbound direction.

180

Alex Anas

references Anas, A. 1982. Residential location market and urban transportation: Economic theory, econometrics and policy analysis with discrete choice models. New York: Academic Press. Anas, A., and R. J. Arnott. 1991. Dynamic housing market equilibrium with taste heterogeneity, idiosyncratic perfect foresight and stock conversions. Journal of Housing Economics 1:2–32. ———. 1993. Technological progress in a model of the housing-land cycle. Journal of Urban Economics 34:186–206. ———. 1997. Taxes and allowances in a dynamic equilibrium model of urban housing with a size-quality hierarchy. Regional Science and Urban Economics 27:547–580. Anas, A., and C. Chu. 1984. Discrete choice models and the housing price and travel to work elasticities of location demand. Journal of Urban Economics 15:107–123. Anas, A., and T. Hiramatsu. 2012a. The economics of cordon tolling: General equilibrium and welfare analysis. Economics of Transportation. www.sciencedirect .com/science/article/pii/S221201221200007X. ———. 2012b. The effect of the price of gasoline on the urban economy: From route choice to general equilibrium. Transportation Research Part A 46:855–873. Anas, A., and I. Kim. 1996. General equilibrium models of polycentric urban land use with endogenous congestion and job agglomeration. Journal of Urban Economics 40:232–256. Anas, A., and Y. Liu. 2007. A regional economy, land use, and transportation model (RELU-TRAN©): Formulation, algorithm design, and testing. Journal of Regional Science 47(3):415–455. Anas, A., and H.-J. Rhee. 2006. Curbing urban sprawl with congestion tolls and urban boundaries. Regional Science and Urban Economics 36:510–541. Anas, A., and R. Xu. 1999. Congestion, land use and job dispersion: A general equilibrium analysis. Journal of Urban Economics 45(3):451–473. Atherton, T. J., J. H. Suhrbier, and W. A. Jessiman. 1975. The use of disaggregate travel demand models to analyze carpooling incentives. Report. Cambridge, MA: Cambridge Systematics. Blackley, D. M. 1999. The long-run elasticity of new housing supply in the United States: Empirical evidence for 1950 to 1994. Journal of Real Estate Finance and Economics 18(1):25–42. Charles River Associates, Inc. 1972. A disaggregated behavioral model of urban travel demand. Report to the U.S. Department of Transportation, Federal Highway Administration. Davis, S. C., and S. W. Diegel. 2004. Transportation energy data book: Edition 24. Oak Ridge National Laboratory. Deakin, E. 1994. Urban transportation congestion pricing: Effects on urban form. In Curbing gridlock: Peak-period fees to relieve traffic congestion. Vol. 2, Commissioned Papers. Transportation Research Board Special Report 242. National Research Council, Committee for Study on Urban Transportation Congestion Pricing, National Academy Press. DiPasquale, D., and W. C. Wheaton. 1994. Housing market dynamics and the future of housing prices. Journal of Urban Economics 35:1–27.

the location effects of alternative road-pricing policies

181

Dixit, A. 1973. The optimum factory town. Bell Journal of Economics and Management Science 4:637–654. Fujishima, S. 2011. The welfare effects of cordon pricing and area pricing: Simulation with a multi-regional general equilibrium model. Journal of Transport Economics and Policy 45:481–504. Green, R. K., S. Malpezzi, and S. K. Mayo. 2005. Metropolitan-specific estimates of the price elasticity of supply of housing, and their sources. American Economic Review 95(2):334–339. Hiramatsu, T. 2010. The impact of anti-congestion policies on fuel consumption, CO2 emissions and urban sprawl: Application of RELU-TRAN2, a CGE model. Unpublished Ph.D. diss., State University of New York at Buffalo. Ingram, G. K., J. F. Kain, and J. R. Ginn. 1972. The Detroit prototype of the NBER urban simulation model. New York: National Bureau of Economic Research. Kain, J. F., and W. C. Apgar Jr. 1985. Housing and neighborhood dynamics. Cambridge, MA: Harvard University Press. Kain, J. F., W. C. Apgar Jr., and J. R. Ginn. 1976. Simulation of the market effects of housing allowances. Vol. 1, Description of the NBER urban simulation model. Cambridge, MA: Harvard University Press. ———. 1977. Simulation of the market effects of housing allowances. Vol. 2, Baseline and policy simulations for Pittsburgh and Chicago. Cambridge, MA: Harvard University Press. ———. 1982. Revitalizing central city neighborhoods: An evaluation of concentrated housing and neighborhood improvement strategies. Cambridge, MA: Harvard University Press. Kimmel, J., and T. J. Kniesner. 1998. New evidence on labor supply: Employment versus hours elasticities by sex and marital status. Journal of Monetary Economics 42:289–301. Lerman, S. R. 1977. Location, housing and automobile ownership and mode choice to work: A joint choice model. Transportation Research Record 610:6–11. Mills, E. S. 1972. Markets and efficient resource allocation in urban areas. Swedish Journal of Economics 74(1):100–113. Smith, B. A. 1976. The supply of urban housing. Quarterly Journal of Economics 90(3):389–405. Struyk, R. J., and M. A. Turner. 1986. Exploring the effects of racial preferences on urban housing markets. Journal of Urban Economics 19:131–147. Sullivan, A. M. 1986. A general equilibrium model with agglomerative economies and decentralized employment. Journal of Urban Economics 20:55–74. Train, K. E. 1976. A post-BART model of mode choice: Some specification tests. Working Paper No. 7620. Urban Travel Demand Forecasting Project, University of California at Berkeley. Turner, M. A., and R. J. Struyk. 1983. Urban housing in the 1980s: Markets and policies. Washington, DC: Urban Institute. Vanski, J., and L. Ozanne. 1978. Simulating the housing allowance program in Green Bay and South Bend: A comparison of the Urban Institute housing model and the supply experiment. Washington, DC: Urban Institute.

commentary Don Pickrell Alex Anas’s chapter is succinct, logically organized, and clearly written. The model it uses to analyze road pricing is comprehensive yet not unnecessarily complicated, and the author’s description of its logic and mechanics is easy for readers—including nonexperts in land use models—to comprehend. Urban land use models have become considerably more complex since I last surveyed them, and much of this progress has resulted from Anas’s unstinting efforts to make them more informative and realistic tools for policy analysis. The chapter analyzes various forms of road pricing, including comprehensive congestion charges, fuel taxes, peak-hour tolls on major routes, and central business district (CBD) cordon charges. The model’s comprehensiveness allows the author to examine an impressive range of impacts, including new development and land conversion decisions, the resulting changes in residential location and employment patterns, and transportation impacts such as automobile use, aggregate travel time and costs, fuel consumption, and CO2 emissions. The RELU-TRAN Model The land use component of the model (RELU) focuses on individuals’ dual roles as labor market participants and consumers, which makes it useful for illustrating policy impacts on their labor supply decisions and the resulting job location patterns, as well as their consumption levels. While this feature complicates the analysis of household-level decisions such as residential location choices because of the importance of multiple-worker households, it significantly improves the model’s realism. One potential concern is the dependence of wages on workers’ skill levels and job location but not on industry, which may appear to make workplace locations more readily substitutable than is actually the case, thus overstating the degree of job mobility. Similarly, although RELU does allow households to have heterogeneous tastes for residential locations, it is unclear whether these vary sufficiently to reflect the observed “stickiness” in household locations that presumably reflects neighborhood and other locational amenities. RELU also treats developers’ decisions about the type and intensity of new land development and conversion of existing land uses explicitly and seemingly realistically, although I had some difficulty understanding the logic of developers’ decisions about demolition and conversion. Specifically, it is not clear whether the mechanism for simulating land use conversion or intensification requires all landowner-developers to contemplate the demolition and subsequent reconstruction of every parcel at the end of each time period, or whether it uses some threshold condition to limit the number of properties that can be converted during any one period. Nor is it clear whether the model incorporates agglomeration economies due to the co-location of mutually supportive industries or the scale of total employment in a zone. These agglomeration economies have been the 182

commentary

183

focus of much recent research, and the model’s accuracy might be improved by incorporating their effects on firms’ location decisions and consequent employment patterns. The role of demographic and economic growth in the model is also slightly unclear, in light of the model’s condition that the new construction and demolition flows of floor space must be equal for each building type within a time period. The chapter indicates that growth can be accommodated and would cause new construction to exceed demolition for some land uses, so there would be net accumulation over time of at least some types of building space, yet it is not clear what assumptions about such growth are used in calibrating the model for the application shown in the chapter. The model’s transportation component (TRAN) analyzes household travel behavior in similarly impressive detail, including vehicle ownership levels, choices of vehicle attributes such as size and fuel efficiency, commuting decisions, and the frequency of trips for shopping and other purposes. The model’s implicit assumption that all vehicles are occupied by only their driver even during commuting hours may be somewhat limiting, since the share of commuting by carpool exceeds that of transit travel even in many cities with high levels of transit service. It is also unclear why only commuting time appears to generate disutility to travelers, although this may simply be intended to reflect the greater likelihood that commuting trips will occur during congested hours. Also questionable is the assumption that the road authority understands how the average value of travelers’ time varies by road segment and that the road authority reacts by differentiating tolls accordingly. Calibration to Chicago Conditions Complex models of simultaneous decisions by many economic agents and their interactions can be intimidating, and their calibration often involves as much judgment as science, so the author’s extensive discussion of the plausibility of the baseline calibration results for Chicago is reassuring. One concern is that the marginal rates of substitution between disposable income and commuting time seem very large by comparison to those usually seen in the transportation literature. It is unclear whether this reflects the author’s assumption about the disutility of commuting in congested conditions or the fact that Chicago-area wage rates (and thus the opportunity value of time itself) are high. Another concern is whether the model includes an income elasticity of demand for housing—presumably meaning residential space and quality—in addition to its price elasticity. The wide variation in the chapter’s reported responses to road pricing by income quartile raises the question of whether those results are dominated by a relative lack of variation in the price elasticity of housing demand among respondents. Finally, the elasticities of fuel demand and vehicle use appear low in magnitude compared to those usually seen in the transportation literature, and the negative value for the elasticity of fuel economy with respect to fuel price appears implausible.

184

Don Pickrell

Design of the Pricing Policies Pigouvian tolls properly capture both the delay and fuel consumption externalities that vehicles reciprocally impose on one another in congested travel conditions. Many analyses ignore the latter component, and its inclusion represents one of the most useful features of Anas’s analysis. The fuel tax equivalents of the tolling schemes are extremely large, and although they are intended mostly to provide a frame of reference for evaluating toll policies, implementing them may be even more far-fetched than adopting ubiquitous marginal-cost pricing of highways. As Anas notes, imposing tolls only on major roads magnifies the differences between the tolling and fuel taxation alternatives because vehicles on local roads would not pay tolls but would be subject to fuel taxes. The fuel tax equivalent of even the partial tolling policy analyzed in the chapter is also very high by U.S. standards; it corresponds to a fuel tax rate of about 150 percent, which is more than 10 times the combined rate of federal and average state taxes on fuel. Results of the Policies In general, the reported effects of the pricing policies on population and employment patterns as well as changes in travel behavior appear directionally plausible if surprisingly modest, particularly for the comprehensive pricing policy. All of the pricing policies examined in the chapter are projected to reduce labor supply, although the estimated shifts in job locations are large relative to declines in total employment. While this effect seems realistic, the potential for it is conspicuously absent from most debates about the desirability of road pricing, and the author’s analysis suggests that it should instead be a major focus. Similarly, while the chapter reports variation in the job and residential location impacts by geographic zone in useful detail, it does not report them by income quartile, and these may be among the most important results. Road pricing might be expected to “sort out” workers of varying income levels—and thus values of commuting time—among existing residences in ways that might have more prominent effects on travel patterns and the resulting environmental externalities than on land development patterns themselves. The author’s analysis also demonstrates that tolling major roads is likely to cause employment to decentralize but population (residences) to centralize, exactly as theory predicts. However, I do wonder a little about the plausibility of CBD workers relocating their residences from suburban zones to the CBD to enable them to commute by transit, especially if the tolling policies were combined with restructuring transit fares so that they resembled marginal costs. In contrast, the increases in labor supply within suburban zones that are projected to result from road pricing seem more plausible, and it might be useful to highlight the relative magnitudes of these effects. Another interesting detail that the chapter does not address explicitly is the extent of variation in residential relocation in response to the various pricing policies by income quartile. I would expect such an analysis to show that higherincome CBD workers are likely to relocate to suburban residences in response to

commentary

185

the higher commuting speeds resulting from road pricing (particularly comprehensive tolling), while lower-income CBD workers relocate to reduce their toll payments. The wide variation among income quartiles in the marginal rates of substitution between commuting time and income used in calibrating the model suggests that both of these effects might be significant. One other minor reservation is that the absence of any explicit assumption about redistribution of toll or tax revenues means that the reported welfare changes, which provide impressive detail about their underlying sources, may still represent an incomplete picture. Despite these concerns, Anas’s chapter offers a thorough, internally consistent, and seemingly realistic analysis of the potentially important but often overlooked consequences of road pricing for the distribution of economic activity within urban areas. It extends the debate about such policies well beyond the more conventional focus on policy consequences for the performance of urban transportation systems themselves.

The Challenges of Large Projects

7 Chicago and Its Skyway: Lessons from an Urban Megaproject Louise Nelson Dyble

T

he Chicago Skyway’s effects on urban development and real estate values have been shaped by dramatic changes over the last half century. The eight-mile roadway and bridge facility was built in the 1950s, just as the city was entering into a long period of spatial reorganization, political transformation, and economic decline. The structure’s awkward design and anachronistic toll-based financing made it an obstacle to local development, and its lack of integration with regional highways resulted in anemic traffic and chronic revenue shortfalls. By 2005, however, its function had been redefined and its value redeemed: in the context of a new political and economic landscape, a 99-year lease to a private consortium transformed it from one of Chicago’s greatest liabilities into a major financial asset. This chapter traces the history of the Skyway, first as a component of a regional transportation system, and later as a distinct urban megaproject. The Skyway’s history reflects broad changes in the purpose of urban infrastructure for aspiring global city leaders and some of the consequences for neighborhoods and communities. Just which lessons Skyway history offers, and whether the structure should be understood as a success or a failure, depends on the perspective and values of the observer. The Chicago Skyway was one of many unsuccessful projects undertaken after World War II by city leaders hoping to somehow harness or reorient the emerging highway system to counter its powerfully centrifugal effects. In this, the Skyway clearly failed. Instead, the Skyway’s actual function as a link connecting Chicago’s business district to an expansive network of interurban toll roads and far-flung suburbs made it a prototype for a new and completely different infrastructure paradigm. Starting in the 1970s, city leaders self-consciously 189

190

Louise Nelson Dyble

adopted new strategies to achieve economic success, prioritizing investment in the central business district, and cultivating a service-based economy catering to international corporations and the financial industry. Not only did the Skyway provide a convenient conduit for traffic to bypass peripheral urban districts and connect with national and international networks, but its privatization also provided an increasingly entrepreneurial city government with immediate financial value. Globalization has transformed city governance throughout the world. The history of the Skyway represents some of the most important urban policies promoting a global city economy: the emphasis in transportation policy on longdistance connections rather than local or regional integration, the role of highprofile megaprojects in attracting investment and signaling status and power, and the shift toward financially driven urban governance. Its history also represents the uneven development associated with globalization; the Skyway was typical of large infrastructure projects dedicated to ensuring the success of the central business district and downtown elites even as they undermined the well-being of other urban populations and places, contributing to greater social and economic inequality. The history of the Skyway evokes the “splintering urbanism” described by Graham and Marvin (2001) that is reflected in international trends in the design, administration, and development of infrastructure serving internationally prominent cities. The Skyway was an early component of a transportation network that at the same time both connected and disconnected, serving Chicago business interests by “bind[ing] spaces together across cities, regions, nations and international boundaries” while defining “the material and social dynamics, and divisions” within and between Chicago neighborhoods (11).

Origins and Intention From a contemporary perspective, the Chicago Skyway fits easily into the category of “megaproject.” While its initial cost does not reach the threshold of most definitions of the term, the $1.83 billion generated by its lease certainly does. It also meets the criteria outlined by Gellert and Lynch (2003, 15–16) as a project that required “coordinated applications of capital and state power” and that transformed surrounding landscapes “rapidly, intentionally, and profoundly in very visible ways.” Today, analysts generally evaluate megaprojects based on their financial performance as distinct, independent facilities. Other functions or relationships with surrounding areas are treated as secondary if they are considered at all. Most contemporary scholars would condemn the decision to build the Skyway because of its abysmal financial performance. However, to its original backers, its primary purpose was not to generate revenue. As they addressed an impending traffic crisis in the early 1950s, Chicago city leaders sought to assert control over the form and function of a regional highway system. The impetus for the Skyway came from an egregious failure of coordination between Indiana and Illinois officials as various state agencies took the initiative

chicago and its skyway: lessons from an urban megaproject

191

to develop highways by any means available. When the chairman of the Indiana Toll Road Commission announced the final route of the state’s new east-west highway in 1953, Chicago leaders were shocked. Originally, plans had called for the cross-state route to extend from the Ohio border to “somewhere south of Gary,” where it would feed onto the Tri-State Expressway, which was already under construction (figure 7.1). Instead, the toll road was scheduled to begin unloading traffic directly onto some of Chicago’s most congested streets early in 1956 (“City protests” 1953). Indiana Toll Road traffic was destined to enter the city in the middle of a heavily industrialized bistate region known as the Calumet, which includes southeastern Chicago and the cities of northwest Indiana. Starting in the 1860s, inexpensive land combined with good transportation access attracted large industrial operations to the area. Steel mills, petroleum refineries, and chemical plants proliferated along the banks of the Calumet, Little Calumet, and Grand Calumet Rivers and along the southern shore of Lake Michigan in the late nineteenth and early twentieth century (Colten 1985; Lewis 2008). They were interspersed with working-class residential areas and company towns like Pullman and Gary. Calumet industries drew workers from Indiana cities including Gary, Hammond, East Chicago, and Whiting, as well as from the neighborhoods of Southeast Chicago, particularly Riverdale, Hegewisch, East Side, and Pullman (Buder 1967; Mohl and Betten 1986; O’Hara 2011). For industrial interests, the rivers, wetlands, and shallow lakes of the Calumet region were mixed blessings. Railroads converged in a narrow corridor, providing easy access to suppliers and markets to the east but complicating street development. Federal funding supported the expansion and engineering of Calumet waterways into an industrialized complex of docks, harbors, and canals. Marshes and streams provided convenient disposal sites for industrial wastes, but unregulated dumping also created the need for unending dredging (Colten 1985, 1986, 1994; Cutler 2006; Hurley 1995). By the 1950s, Chicago planners were warning that growth in the Calumet region posed serious challenges, particularly for transportation: “[w]ithout careful control, . . . transportation planning could become the hopeless task of attempting to build facilities for mammoth land use bodies generating hordes of traffic onto a pigmy street and transit skeleton” (Chicago Plan Commission 1956, viii). Multiple lift bridges over the Calumet River represented particularly egregious problems, stopping traffic whenever ships traveled in or out of the port and posing hazards to large lake freighters. City planners anticipated an enormous increase in shipping volume and in the size of ships with the opening of the St. Lawrence Seaway in 1959. The Illinois state legislature created a regional port district in 1951 to expand and modernize Calumet facilities, and the construction of a new ship-truck-train freight exchange terminal was already under way. However, there were no specific plans for replacing or updating nearly a dozen obsolete bridges (Chicago Plan Commission 1956; Chicago Regional Port District 1953).

192

Source: Dennis McClendon, Chicago CartoGraphics.

Figure 7.1 Map of the Chicago Skyway and Environs

chicago and its skyway: lessons from an urban megaproject

193

The route of the Indiana East-West Toll Road as announced in 1953 would severely exacerbate Calumet traffic problems. The decision-making process for the toll road took place behind closed doors as appointed commissioners conferred with financial and engineering consultants, maintaining secrecy to “prevent real estate speculation” (“Gov. Craig” 1953). The result reflected little regard for the interests of local neighborhoods or communities: the toll road sliced through Gary’s downtown, for example, and protests and lawsuits did nothing to ameliorate the damage it caused. Alderman Emil Pacini, who represented much of Chicago’s portion of the Calumet, was outraged and called for immediate action to address an “emergency situation.” The Chicago public works commissioner accused “engineers and bankers” of cynically exacerbating congestion to maximize toll revenue. Official protests from the Cook County highway department and the Chicago city council had no effect. Representatives of the banking syndicate handling toll road financing warned that any changes to the route would threaten the viability of the entire project. Commissioners insisted that “politics” should not come into consideration (“City protests” 1953). In the era before the federal highway trust fund, local and state governments struggled to pay for urban highways. In 1953 Cook County’s share of state gas taxes was already committed for the foreseeable future, and did not even come close to meeting demand for new routes. The new Illinois Toll Highway Commission, dominated by state-appointed Republicans unsympathetic to Chicago’s problems, refused to build a connection to the new Indiana terminus because of the high cost of urban land acquisition. Chicago leaders had no recourse but to finance their own solution to the impending local traffic disaster, and they took action. By the time Indiana Toll Road construction was under way in 1954, the City of Chicago had secured $88 million to build a bridge and toll road to funnel traffic from the state line, over the Calumet River, and northwest seven miles toward the Loop (Dyble 2012). The quick financing and construction of the Chicago Skyway reflected a long tradition of unwavering elite support for large-scale city-centered transportation. The city’s boosters had always viewed infrastructure as the key to achieving regional dominance and making Chicago a world-class city. Transportation infrastructure was the basis for its history of spectacular growth; a continental network of canals and railroads supported an empire, their centralization reinforced by commuter service and urban transit that converged in a downtown “Loop.” In the automobile age, Chicago leaders sought to reinforce long-standing patterns through highway development (Cronon 1991; Miller 1996). Canals and railroads helped make Chicago a major power center by the beginning of the twentieth century, and they also served a regional purpose: they defined and sustained a centralized urban territory. Based on an “integrated ideal,” the success of cities was understood to depend on effective command of a regional economy (Graham and Marvin 2001, 89). The role of government was to promote and direct development by, among other things, working closely with financial, industrial, and commercial elites to provide for the development of

194

Louise Nelson Dyble

cohesive infrastructure networks that would provide effective connections at the local, regional, and national scale. The vast railroad system that enabled Chicago’s domination of the midwestern economy was both the product of this modern paradigm of urban development and essential to its success.

Design and Purpose The unexpected Calumet crisis created by the routing of the Indiana Toll Road posed major problems for Chicago, but it also presented opportunities. The original toll road plans would have routed Indiana traffic onto the peripheral TriState Expressway, which bypassed Chicago. With responsibility for developing a solution to the Calumet traffic crisis, Chicago leaders could take the chance to ameliorate a larger problem: the city’s declining regional centrality relative to its booming suburbs. In the decade following World War II, city leaders assumed that the same basic patterns would continue as a regional highway system was added to Chicago’s transportation web. Both Daniel Burnham’s 1906 Plan of Chicago and the 1939 Comprehensive Superhighway Program described radial regional road systems centering on downtown. However, highways were fundamentally different from previous transportation facilities, both functionally and institutionally, and cities had much less influence over their form. In the years following World War II, it became increasingly clear that the effect of highways was not to concentrate population and resources in the city but rather to disperse them (Barrett 1975, 1987; Dyble 2009; Fogelson 2001; Smith 2007; Spatz 2010). The result was a highway system that bypassed and undermined the power and vitality of Chicago rather than supporting it. The Skyway was one of several infrastructure projects initiated under Mayor Martin Kennelly designed to counter the “tug of decentralization,” along with a subway system and street expansions that improved access to downtown (Hirsch 2005, 132). The Skyway’s most revealing aspects were its approaches and the system of one-way streets that fed them. The Skyway was more than just a bridge; its seven miles of three-lane approaches made it an important addition to Chicago streets. However, the Skyway’s design made it more of an obstacle than a benefit to local traffic (figure 7.2). Heading west, motorists exited onto streets that directed them toward the Loop and made it difficult or impossible to turn around or otherwise navigate local streets. Eastbound motorists could not exit before they reached Indiana. Built alongside the Pennsylvania railroad tracks to minimize land-acquisition costs and building demolition, the Skyway reinforced and widened a band of infrastructure that already obstructed local traffic (De Leuw, Cather and Company 1954; Spivey 2001). As planners recognized, the area was desperately in need of separated grades to reduce delays and increase street capacity (Chicago Plan Commission 1956). The soaring span took grade separation too far, exacerbating rather than alleviating Calumet congestion.

chicago and its skyway: lessons from an urban megaproject

195

Figure 7.2 Main Span of the Chicago Skyway

The Skyway rises 125 feet above the Calumet River and is served by seven miles of limited-access approaches. Source: Historic American Engineering Record (1999).

Of course, there were reasons for this design. Skyway underwriters paid for its engineering, and to them, limited access made sense financially. It eliminated the need for more than one collection point, and congestion would motivate drivers to pay a toll. The Skyway was not intended to serve the workers, trucks, and commercial traffic of the Calumet; its only benefit to them was to prevent significant new congestion. Its designers assumed that most users would be the same white-collar suburban commuters or long-distance travelers who used turnpikes elsewhere. These motorists would be eager to bypass the smoke, soot, and fumes emanating from Calumet smelters and smokestacks, as well as the scents of some of the city’s only active sanitary landfills (figure 7.3). The expectations and rhetoric surrounding the 1953 Skyway proposal reflected longstanding priorities of Chicago development and infrastructure policy: to integrate the regional economy around a central urban core through investments in transportation.

Figure 7.3 1960 Chicago Skyway Promotional Map

This map was issued soon after the name of the structure was changed from the Calumet Skyway to the Chicago Skyway to reduce confusion among out-of-town motorists. The image suggests a remote destination and highlights the structure’s detachment from its immediate surroundings. The Skyway was a route “to and through Chicago” but was implicitly not part of the city. Source: University of Chicago Library Map Collection.

196

chicago and its skyway: lessons from an urban megaproject

197

Traffic and Finances If the Skyway’s backers were working on the assumption that the project would have a beneficial effect on the Calumet economy while supporting the centrality of the Loop as other transportation systems had, they were quickly proven wrong. On opening day in April 1958, fewer than half of the expected motorists crossed the span, and traffic did not improve significantly for decades. There was little to suggest that the project did anything to spur development either in the Calumet area or downtown. By 1963 Skyway bonds were in default, and they remained in default through the 1990s. As with most toll roads, financiers were directly involved in Skyway planning from the beginning. The firms selected to underwrite the structure not only paid for its routing and design, but also provided all financial analysis. Coverdale & Colpitts, a transportation consulting firm based in New York that had done studies for all U.S. toll roads since 1945, developed detailed traffic and revenue projections (Ryan 1954). Analysts adopted several problematic assumptions. They predicted that most of the traffic of Indianapolis Avenue, the major Calumet area artery for commercial traffic in both Chicago and Indiana, would be diverted to the Skyway, disregarding the impracticality of the structure for local traffic. They also estimated 25 percent induced traffic, predicting that the Skyway would cause a “major relocation of industry and residential areas and [change in] travel and recreation habits of large numbers of people.” Most significantly, the analysis discounted plans to extend the Tri-State Expressway to connect with the Indiana East-West Toll Road east of Gary, which would give motorists easy access to a toll-free alternative parallel to the Skyway (Coverdale & Colpitts 1954, 10). For the most part, highways in the 1950s and 1960s experienced surging traffic, and toll roads throughout the United States generated revenue well beyond expectations. In contrast, the Skyway was uniquely positioned to avoid traffic because of its relationship with surrounding areas and its lack of integration with emerging regional highway systems. The Indiana state highway commission dealt the most devastating blow to Skyway solvency in 1964. That year, the Burns-Harbor interchange connected the Indiana East-West Toll Road to the toll-free Tri-State Expressway just east of Gary. This meant that traffic heading west could avoid the Skyway toll and never had to cross through the Calumet industrial district at all. Once the interchange opened, the Skyway primarily served traffic between the Calumet cities of Gary, Whiting, and Hammond and Chicago’s South Side or the Loop. Several better, less expensive routes were available to suburban commuters that skirted central Chicago and avoided the Skyway toll. Over the long term, the inaccuracy of traffic projections (figure 7.4) is not surprising. No one predicted the rapid economic decline and population loss that affected South Chicago and the Calumet area in the decades to come. While the Calumet region maintained its industrial economy through the 1960s, the neighborhoods of the South Side had already entered into a period of social and economic transition, as white residents and jobs left en masse for the suburbs

198

Louise Nelson Dyble

Figure 7.4 Actual and Projected Traffic on the Chicago Skyway, 1958–1990

Skyway annual traffic

30,000,000 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 1958 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1991

Coverdale & Colpitts projections, 1954

Actual Skyway traffic

Several bondholder lawsuits forced toll increases, so revenue improved relative to projections in the late 1980s. Bond interest was brought up to date for the first time in 1989, and the structure was refinanced in 1994.

(Squires et al. 1989). In addition, the Skyway was a completely independent city project. State highway officials had no interest in its financial problems and no sympathy for its financiers, and they proceeded Figure 7.4to build a series of competing “free” routes supported by gas taxes in the 1960s. Lincoln_Ingram_Infrastructure Immediate shortfalls are harder to explain. There were no precedents for the wildly inaccurate Skyway traffic projections or for its spectacular financial failure; it was the largest public revenue bond default in U.S. history until 1983 (Cohen 1989). Previous Coverdale & Colpitts reports had underestimated demand for toll roads, including a study for the New Jersey Turnpike that caused a minor scandal when the facility faced almost immediate congestion. However, manipulating cost estimates to ensure that megaprojects are approved and funded is not unusual and may even be considered a standard practice. This raises the question of whether this was a case of “lying” at the behest of financiers to ensure the sale of bonds, or if it was simply a case of incompetent analysis (Flyvbjerg, Holm, and Buhl 2002; Wachs 1986). Regardless of the answer, bondholders bore the brunt of inaccurate forecasts. There were ample incentives for consultants, engineers, and financiers to ensure that the Skyway was financed and constructed, whether or not it was financially viable. Well-connected investors could also expect that their confidence would be rewarded in other ways. There are many ways to make money on a big public project, but none on one that is canceled. For city leaders, the imperative of maintaining the centrality of downtown Chicago easily justified swift and decisive action to build the Skyway, despite

chicago and its skyway: lessons from an urban megaproject

199

financial or ethical risks. Mayor Richard J. Daley, Kennelly’s powerful successor, fully supported the project and took action to ensure that its financial problems would not jeopardize any other potential city infrastructure projects. After a few years of dismal Skyway performance, evidence emerged of insider trading and financial manipulation by city officials to ensure underwriter profits. In addition, Daley repeatedly sought state or federal takeovers that would result in full debt redemption (Dyble 2012).

Decline and Decay The Skyway failed to achieve its original goals. Rather than preventing traffic congestion in the Calumet area, it added another obstacle to movement around its struggling steel mills, factories, and port. Its dismal traffic suggested that any benefit it might have had for the Loop was negligible. Rather than stimulating development as other transportation systems had, urban highways hastened the decline of the city, dispersing and decentralizing resources and population. The Skyway’s performance was just one of many indications that the previous model of urban development was breaking down as deindustrialization and disinvestment transformed the urban Midwest into the nation’s “Rust Belt.” These changes were already well under way in Chicago when the Skyway opened in 1956. The city’s population peaked in 1950 at 3.6 million, and 1960 census numbers came as a shock. Hinterland had transcended city: 3.3 million city dwellers were surrounded by 3.6 million suburban residents. Chicago’s demographics were also transformed as the city lost 1.1 million white residents while gaining half a million black residents. Incomes plummeted, and poverty and segregation increased. These numbers marked the beginning of long-term trends: in the decades to come, Chicago’s economy declined while that of surrounding areas surged (Abu-Lughod 1999). As in many urban areas, highway development contributed to these trends, providing the fundamental infrastructure of decentralization and facilitating the shift of population and jobs away from the city (Gutfreund 2005; Jackson 1985; Jones 2008). Ongoing changes were not experienced evenly across Chicago neighborhoods, and the areas that most influenced Skyway traffic faced some of the most severe problems. South Side businesses and residents would have been among those likely to benefit from the ability to bypass Calumet area congestion, but the area’s economy was in the process of collapse by the 1970s. Increasingly poor neighborhood residents had little incentive or means to pay tolls. Industrial decline and population losses followed. Calumet port traffic grew in the 1960s, but failure to invest in new facilities rendered it uncompetitive with other Great Lakes ports for containerized shipping in the 1970s. In 1980 Wisconsin Steel became the first of Chicago’s many large industrial employers to leave the Calumet region, initiating a period of punctuated job losses and economic contractions as a series of large factories shut their doors over the next 10 years (Hurley 1995; O’Hara 2011; Squires et al. 1989; Wiewel 1986).

200

Louise Nelson Dyble

The Skyway by no means caused the catastrophic decline of the Calumet, but it did exacerbate its existing problems and reduced its chances for economic recovery. One of the reasons large industrial employers were originally attracted to the area was its isolation from the rest of the city; company towns could be established and labor relations controlled without undue influence from Chicago’s powerful and often radical unions. The Skyway reinforced this historic isolation, providing a means of avoidance rather than access. It contributed yet another physical obstruction to an area that was already sliced and segmented by converging and often conflicting transportation systems. This infrastructure landscape, along with the myriad of severe environmental problems, obstructed the recovery of neighborhoods like Hegewisch and the East Side in the wake of economic collapse. In 1966 state toll highway commissioner Donald Bonniwell remarked that the Skyway functioned as “a concrete curtain cutting off access by people in the southeast section of the city” (McMullen 1966). The Skyway came to symbolize neighborhood neglect and decline. Through the 1970s and most of the 1980s, traffic remained flat and the Skyway deficit grew. Negligent maintenance, no-bid contracts, nepotism, and toll-booth theft all generated regular scandal. The Skyway’s historical nadir occurred around 1976, when local papers reported damage to cars from potholes in the Skyway roadbed, and sloppy sandblasting left the surrounding homes covered in a layer of leadtainted coal-like paint dust (Seltzner 1976). Its accumulated net deficit peaked at $72 million in 1988 (Chicago Department of Streets and Sanitation 1989). The Skyway remained an unredeemed failure for three full decades, but during that time new ideas and approaches to urban development were tested in the vacuum left by the slow collapse of Chicago’s industrial economy. In the 1950s, city leaders expected that industry in Southeast Chicago would continue to thrive. However, they believed that downtown was in serious jeopardy, threatened by declining property values and the steady loss of retail, commercial, and financial establishments. The emphasis of city policy began to shift from promoting diverse development throughout Chicago to a strategy focused primarily on redeveloping the central business district (Miller 1996; Rast 1999, 2001; Wille 1997). Mayor Kennelly had worked with business and real estate interests on downtown-oriented transportation projects including the Skyway, but when Richard J. Daley was elected mayor in 1955 he took a much more aggressive approach to transforming the area in and around the Loop. He organized the massive Chicago Area Transportation Study (CATS), which produced an ambitious plan of multimodal transportation development that transformed the city in the decades to come (Condit 1974; Spatz 2010). Daley also backed the creation of the Chicago Central Area Committee (CCAC) in 1956, comprised of real estate, financial, and business leaders with an interest in revival and redevelopment. The CCAC emerged as one of the most influential policy-advocacy groups in the decades to come, spearheading a broad program of renewal, redevelopment, and construction focused on the central business district (Rast 1999, 2011).

chicago and its skyway: lessons from an urban megaproject

201

The vision of the CCAC was first articulated in the 1958 Development Plan for the Central Area of Chicago, which outlined a program for downtown. Drawing on earlier city highway proposals, the plan emphasized access; the Skyway would be just one spoke in a wheel of highways that would center on the central business district and feed extensive new parking facilities (Rast 2001). One of the ironies of Skyway history was that the structure was financed just before the Federal-Aid Highway Act of 1956 provided generous funding for urban highway construction. The projects proposed by the 1958 plan, in contrast, were well timed to take advantage of federal funding. They had the full support of the influential members of the CCAC and of Mayor Daley, who made public works construction, including transit and highway megaprojects, a keystone of city policy. Between 1955 and 1976, the Daley administration effectively tapped into new federal funding for a variety of ambitious urban renewal and transportation projects that physically transformed the city (Condit 1974; Suttles 1990). At the same time, the city also underwent a dramatic social and political transformation. Although the 1958 plan clearly promoted the interests of business elites, it was not generally viewed as antagonistic to the rest of Chicago; at that time, urban development was not perceived as a zero-sum game among neighborhoods. Another report released in 1966 outlined a complementary plan to address the economic problems of the South Side (Mayor’s Committee for Economic and Cultural Development 1966). However, Daley did not lend even a fraction of the political support or financial resources to this program that he devoted to downtown rehabilitation (Rast 1999, 2011). As the urban crisis intensified, it became clear that city priorities were resulting in the neglect of most of Chicago’s neighborhoods, and particularly those with majority black populations to the south of the Loop. The violence of 1968 following the assassination of Martin Luther King Jr. intensified racial antagonism and division, accelerating public and private disinvestment. In 1973, the response to the CCAC’s second major downtown plan, Chicago 21, highlighted deepening divisions within the city. Chicago residents, organized in a growing number of varied neighborhood advocacy groups, objected to its emphasis on suburban commuter rail and highways and to the absence of any regard for intracity integration. The plan described downtown as a white-collar center of finance and big business, lacking substantial ties to the rest of the city (Rast 1999, 2011). It was supported by infrastructure that, like the Skyway, provided access “to and through” the city, emphasizing long-distance connections and generally bypassing peripheral neighborhoods: regional commuter rail, airport expansion, and a series of high-speed, limited-access expressways. This was the vision for Chicago’s future that the Skyway best served. However, policies that sacrificed the interests of neighborhoods outside of the Loop met with resistance. Activists opposed the Crosstown Expressway in the 1970s, the only one of Daley’s major public works projects that was defeated by public pressure before his death in 1976 (Spatz 2010). By the time Harold

202

Louise Nelson Dyble

Washington, who was devoted to supporting neighborhood and community development, was elected mayor in 1983, a myriad of new neighborhood activist groups contributed to a clear “downtown versus the neighborhoods” orientation (Mier and Moe 1991, 67). This antagonism hindered Washington’s ability to implement policy, and although he managed to win approval for a $185 million city bond issue for neighborhood improvements in 1985, many of his proposals were stymied by downtown opposition (Ferman 1996). Between 1976 and 1989, Chicago officials had lost much of their power to shape the urban landscape or to promote economic development.

Redefinition and Redemption In the 1990s, changing economic and political circumstances transformed perceptions of the Skyway and created opportunity for its redemption as a city asset. In 1989 the toll bridge generated enough revenue to bring bond interest payments up to date for the first time since 1969. The same year, Richard M. Daley, the son of Richard J. Daley, won election as mayor. The younger Daley’s approach to urban development and transportation policy in many ways resembled that of his father, dedicating city resources primarily to promoting development and investment in the Loop. However, during his administration city government adopted a new orientation: the second Daley administration represented a model of the “entrepreneurial city,” seeking to promote the interests of city government as an independent entity, maximizing revenue, and building strategic alliances and partnerships with financial and business interests (Eisinger 1988; Hall and Hubbard 1998; Harvey 1989). There were several reasons for the shift toward entrepreneurial city government that occurred throughout the United States in the 1980s and 1990s. Contributing to the change was growing financial pressure on city government, as reduced federal appropriations and new tax restrictions added to continuing urban economic problems. Cities like Chicago faced declining revenue and chronic, structural budget deficits (Biles 2011; Eisinger 1998; Fuchs 1992). Daley distinguished himself as an innovator, identifying new resources and promoting publicprivate partnerships designed to attract investment, reduce city obligations, and tap into new sources of revenue. The second important reason had to do with Chicago’s changing aspirations: looking to the financial and economic success of global capitals like Tokyo, London, and New York, city leaders throughout the world sought to attract financial institutions, corporate headquarters, and high-tech firms (Abu-Lughod 1999; Sassen 2001). Daley and his backers hoped to build a “New Chicago” resting on a service-based economy, with highly educated elites catering to the financial and legal needs of international corporations and low-wage workers providing amenities and readily available, inexpensive labor (Koval et al. 2006). To many, this new economic paradigm offered a compelling strategy for urban economic revival after decades of industrial decline.

chicago and its skyway: lessons from an urban megaproject

203

But it came at the price of greater inequality, not only among social groups, but also among neighborhoods. Infrastructure once again became a key component of Chicago’s economic development strategy under Mayor Richard M. Daley. However, the intended function of transportation infrastructure was different than it had been during the first Daley administration, and so was the relationship of the central business district to the rest of Chicago. Rather than integrate the city and connect various elements of a diverse urban economy, infrastructure supported communication and movement among and between major cities. In addition, infrastructure was no longer primarily something to be supported and subsidized for the sake of promoting city and regional development, but instead any facilities with the potential to generate revenue exceeding the cost of their construction and operation could be financial assets for city government. Infrastructure administration changed as well, with more facilities managed separately as discrete structures, rather than centrally as part of interconnected, coordinated systems. When policy analysts began to discuss megaprojects and their implications (e.g., Altshuler and Luberoff 2003; Flyvbjerg, Bruzelius, and Rothengatter 2003), their discourse reflected trends in urban governance: decentralization, segmentation, and privatization. Whereas previously the Skyway’s administration had been anomalous, in this new context it was prototypical. It inspired a new strategy for unlocking the potential value of city transportation facilities. In many ways, Richard M. Daley continued the downtown-oriented development priorities of his father, emphasizing central city development projects catering to the interests and preferences of downtown real estate interests and the international corporations and financial, legal, and high-tech firms that they hoped to attract. Daley’s hallmark infrastructure projects—the expansion of the University of Illinois campus to the west of the Loop, the showcase Millennium Park on the downtown lakeshore, and the aggressive expansion of O’Hare International Airport—were all dedicated to promoting Chicago as a global city. However, his most influential legacy as mayor was not manifest in bricks and mortar but rather in his willingness to surrender control of revenue-generating facilities to promote the immediate financial interests of the city and to facilitate Chicago’s transformation into a global city (Bennett 2010, 2011). The potential advantages of the Skyway’s independent administration and detachment from the rest of the highway system, particularly for a cash-strapped city government with an entrepreneurial orientation, became clear as its traffic and revenue improved in the 1990s. Daley identified the Skyway as a potential financial asset, taking advantage of it in 1994. That year, Skyway revenue was sufficient to retire the original construction debt and to float an additional $110 million in revenue bonds, some of which went toward repaying city appropriations for debt payment and operating expenses dating back to the 1960s. Two years later, Skyway tolls supported another bond issue, paying for the structure’s rehabilitation as well as a popular city “neighborhood infrastructure plan,”

204

Louise Nelson Dyble

including pothole repair and park construction. By that time, Skyway motorists were generating a $17 million annual surplus above maintenance and debt payments (Dyble 2012). Although Skyway refinancing alleviated severe city budget problems and provided revenue for neighborhood development, it also produced a major political backlash. Tradition dictated that bridge tolls be used exclusively for construction or maintenance, and motorists did not view the Skyway as a legitimate source of city revenue. Critics charged that the facility was becoming the city’s new “cash cow,” and a lawsuit by toll payers challenged the legitimacy of revenue “diversion” for anything other than highway development (Dyble 2012, 74). While the suit was legally baseless, protest and pressure from motorists threatened to undermine the potential value of the Skyway as a city asset. However, global trends presented a potential alternative. Internationally, private toll roads were becoming a big business, as revenue-generating infrastructure was increasingly constructed and operated as private enterprise. Throughout Europe, Australia, Canada, and Asia, investors were financing or refinancing major tolled facilities, taking over their operation as well as their revenue. The Daley administration took the opportunity to secure a large, immediate financial return from the Skyway for the city. Rather than endure unending criticism of toll “diversion” under city management, the city could lease the Skyway to a private operator with little exposure to public pressure. In theory, the private operator could tap into the structure’s long-term value much more effectively than could public officials. While Skyway traffic remained lackluster, it continued to improve slowly in the context of severe highway congestion on surrounding routes. In 2004 Daley issued a call for bids for a Skyway lease, and his timing was ideal. International enthusiasm for infrastructure leases was being fueled by the rapid expansion and dramatic profits of the Macquarie Infrastructure Group (MIG), a private fund that was managed by an Australian investment bank (Jefferis and Stilwell 2006; Solomon 2009). In partnership with Cintra, a Spanish construction firm, Macquarie submitted a stunning offer in response to Daley’s call for bids: $1.83 billion to operate the Skyway for 99 years. It was much more than analysts had predicted, and nearly twice the next largest bid. Chicago aldermen unanimously approved the lease, transferring control and the right to collect tolls on January 24, 2005. Alderman Edward Burke called it “the greatest single financial coup in the history of Chicago” and a “windfall” comparable to the purchase of Manhattan (Dyble 2012, 74). City officials commenced distributing the funds, paying off debt, providing for “the city’s continued financial strength and stability,” and establishing longand mid-term reserve funds. One hundred million dollars of the proceeds were used to create a “neighborhood human infrastructure fund” that supported a wide variety of social programs (Chicago City Council 2004). The Skyway lease was one of the defining achievements of Richard M. Daley’s political career, and it

chicago and its skyway: lessons from an urban megaproject

205

inspired a series of similar proposals around the United States, including the lease of the Indiana East-West Toll Road to the same consortium just a few months later. Daley maximized the political and financial benefits of the deal. He was savvy enough to distribute the benefits of the windfall widely, using it to help shore up city finances and consolidate his power as mayor (Johnson, Luby, and Kurbanov 2007). By leasing the Skyway, Daley sacrificed patronage appointments and a longterm revenue source, but the venerable, ward-based political system that rallied voters at the neighborhood level with city jobs and other favors during the first Daley administration was no longer the foundation of mayoral power. By the end of the century, the outcome of elections depended more on campaign contributions from big business that paid for mass media advertisements and other publicity. Daley was successful in his efforts to reshape downtown Chicago to attract large firms and investors, and his campaign contributions testified to his success (Hogan and Simpson 2001; Simpson and Kelly 2011). Critics described his close relationships with corporate and international interests as “pinstripe patronage” (Betancur and Gills 2004, 98). As during the first Daley administration, the mayor effectively controlled and assembled alliances that allowed him to define a development program and to implement the policy it required. However, in doing so the younger Daley went much farther in prioritizing and promoting the central business district as the locus of a new urban economy based on finance, technology, innovation, and consumption in pursuit of success as a global city. Daley followed up on the Skyway success with three more transportation infrastructure lease proposals. In 2006 a 99-year lease of four downtown parking garages garnered $563 million for the city, and Chicago officials traded 75 years’ worth of parking meter revenue for $1.1 billion in 2008 (Kaplan 2012). The mayor also brokered a lease deal for Midway Airport, Chicago’s regional facility located in South Chicago, in 2008. However, a global financial crisis halted new investment, and the Midway investors withdrew their bid for a lease before the contract was executed. By that time, the value of big-ticket infrastructure leases was coming into question. In 2007 scandal erupted surrounding the management of Macquarie’s investment funds, which included incentives for large deals and significant conflicts of interest (Lawrence and Stapledon 2008; McLean 2007). Criticism led to the resignation of the bank’s CEO, Allan Moss, and a major restructuring of its investment funds to separate low-performing and heavily indebted toll roads, including the Skyway, from more profitable “high-quality assets” (Bennet 2010). The scandal, combined with lagging traffic in the context of a general economic slowdown, resulted in a rapid devaluation of many Macquarie acquisitions. By 2011 the Skyway was listed in investor reports with a carrying value of zero (Macquarie Atlas Roads 2011). There is justification to predict that the Skyway may return to its original status: a disappointing city liability with little value to surrounding neighborhoods or to the city as a whole.

206

Louise Nelson Dyble

Daley announced his retirement from politics soon after the Midway lease fell through, leaving the city with the same persistent structural budget problems that he had inherited upon taking office and fewer options for addressing them. His development priorities contributed to a celebrated revival of the Chicago economy based on the success of a few elite neighborhoods and central business district, but his priorities were also reflected in the landscape of the Calumet and the increasing poverty of the South Side and other neighborhoods. One of his very first proposals as mayor was the construction of a new Lake Calumet Airport; the airport would have mitigated some of the area’s toxic waste problems but would have destroyed thousands of homes and several local neighborhoods in the process (Chicago Department of Aviation 1991). While Daley touted the airport’s potential to generate jobs, it would have displaced some of the area’s largest remaining employers and inflicted new damage on its wetland ecology. In a sense, it would have continued the transformation of the Calumet area from an industrial center to a transportation corridor, serving the interests of global Chicago while sacrificing those of the neighborhood’s remaining residents. After Daley dropped the proposal in 1992, he never showed much interest in the fate of the Calumet again. City plans to attract new industry and ecotourism to the area lacked funding and yielded few results (Chicago Department of Planning and Development 2001). The 2010 census revealed that the Calumet and the rest of South Chicago continued to lose jobs, resources, and population. Today, the Skyway provides a means of avoiding a grim postindustrial landscape of abandoned barges and piers, slag piles, salvage yards, and the decaying ruins of factories and steel mills.

Conclusions Although the era of infrastructure lease “windfalls” may be over, the economic trends and urban problems that provided the backdrop for the Skyway lease continue to inform infrastructure policy in aspiring global cities like Chicago. Downtown interests remain a priority, and city leaders continue to court international corporations and high-tech firms, and to promote service- and consumptionbased economic development. Municipal governments continue to face ideological pressure and financial constraints that limit policy options and promote an entrepreneurial orientation. Urban transportation investments favor long-distance connections via highways and airports. Megaprojects, including bridges and tunnels as well as cultural facilities like stadiums and park complexes, remain appealing as symbols of status and power as well as for their functional or financial value. The segmentation and reorientation of urban infrastructure over the last half century have had broad social and economic consequences. It is no coincidence that megaprojects, which are scaled to the metropolis and are inevitably disruptive to urban environments and neighborhoods, have met with increasingly vigorous public protest in North America and elsewhere. They both con-

chicago and its skyway: lessons from an urban megaproject

207

tribute to and reflect policies that result in extremely uneven local and regional development outcomes. As infrastructure systems have been divided into their component parts, so have places become disconnected from regions, and political processes removed from policy. A growing literature on twenty-first-century urbanism emphasizes the resulting environmental and economic inequality and political disfranchisement (e.g., Brenner 2004; Castells 1996; Hackworth 2007; Ranney 2002; Smith 1996, 2008). The history of the Skyway highlights some of the consequences of the “splintering urbanism” that has shaped urban policy and its consequences in major cities throughout the world, as growing physical and institutional divides separate and disconnect urban places, defining their relationship with the global economy and resulting in stark patterns of economic and social disparity (Graham and Marvin 2001). Once the global city model of urban development prevailed in Chicago, the design problems of the Skyway and its lack of integration with other transportation systems were no longer important concerns to city leaders. Although its traffic and revenue continued to improve through the 1990s, the structure still had basically the same problematic relationship with its surroundings, with the regional highway system, and with the Loop. But because of a change in perspective, a slight improvement in traffic and revenue, and a major shift in economic development policy, the Skyway was redefined as a success. Its lack of integration with local and regional transportation systems was originally a liability; its city administration and financial problems were compounded by the disregard of state and federal officials, who were uninterested in mitigating the damage of its rushed design and financing. However, its institutional disconnection eventually became a benefit: it made it easy for city leaders to transfer responsibility and control while leveraging its financial value to literally capitalize on its singularity. The Skyway’s successful privatization as well as its uneven effects on city development were representative of larger trends, reflecting the splintering effects of infrastructure in the twenty-first-century global metropolis. Because of its physical detachment from its surroundings and its failure to attract or induce new traffic, the Skyway had a relatively small impact on the development of Chicago. However, the structure did contribute to the transformation of relationships between places, in this case between the central business district and the peripheral Calumet industrial district. This relationship was indicative of momentous change in the orientation and outcomes of urban policy and development patterns that is represented in metropolitan areas throughout the world. The fate of the Calumet suggests the powerful local implications of global city policies. Its history also underscores the importance of considering the effects of megaprojects on urban places and communities, in addition to their financial costs and benefits for city governments.

208

Louise Nelson Dyble

references Abu-Lughod, J. L. 1999. New York, Chicago, Los Angeles: America’s global cities. Minneapolis: University of Minnesota Press. Altshuler, A., and D. Luberoff. 2003. Mega-projects: The changing politics of urban public investment. Washington, DC: Brookings Institution. Barrett, P. 1975. Public policy and private choice: Mass transit and the automobile in Chicago between the wars. Business History Review 49(4):473–497. ———. 1987. The automobile and urban transit: The formation of public policy in Chicago, 1900–1930. Philadelphia: Temple University Press. Bennet, M. 2010. Macquarie Infrastructure to split into Intoll International, Macquarie Atlas Roads. The Australian, 22 January. Bennett, L. 2010. The third city: Chicago and American urbanism. Chicago: University of Chicago Press. ———. 2011. The mayor among his peers: Interpreting Richard M. Daley. In The city, revisited: Urban theory from Chicago, Los Angeles, and New York, ed. D. R. Judd and D. Simpson. Minneapolis: University of Minnesota Press. Betancur, J. J., and D. C. Gills. 2004. Community development in Chicago: From Harold Washington to Richard M. Daley. Annals of the American Academy of Political and Social Science 594:92–108. Biles, R. 2011. The fate of cities: Urban America and the federal government, 1945– 2000. Lawrence: University Press of Kansas. Brenner, N. 2004. New state spaces: Urban governance and the rescaling of statehood. Oxford, U.K.: Oxford University Press. Buder, S. 1967. Pullman: An experiment in industrial order and community planning, 1880–1930. New York: Oxford University Press. Castells, M. 1996. The rise of the network society: The information age. New York: Wiley. Chicago City Council. 2004. Lease ordinance. Journal of Council Proceedings (27 October):2905–3545. Chicago Department of Aviation. 1991. Lake Calumet Airport: Crossroads of the nation—future of the region. Chicago: City of Chicago. Chicago Department of Planning and Development. 2001. Calumet area land use plan. Chicago: City of Chicago. Chicago Department of Streets and Sanitation. 1989. Chicago Skyway annual report 1988. Chicago: City of Chicago. Chicago Plan Commission. 1956. The Calumet area of metropolitan Chicago. Chicago: City of Chicago. Chicago Regional Port District. 1953. First biennial report: Where two great waterways meet. Chicago: Chicago Regional Port District. City protests Indiana’s toll road terminal. 1953. Chicago Tribune, 18 December. Cohen, N. R. 1989. Municipal default patterns: An historical study. Public Budgeting and Finance 9(4):55–65. Colten, C. 1985. Industrial wastes in the Calumet area, 1869–1970: A historical geography. Champaign: Illinois Natural Resource Center. ———. 1986. Industrial wastes in Southeast Chicago: Production and disposal, 1870–1970. Environmental Review 10(2):93–105.

chicago and its skyway: lessons from an urban megaproject

209

———. 1994. Chicago’s waste lands: Refuse disposal and urban growth, 1840–1990. Journal of Historical Geography 20(2):124–142. Condit, C. W. 1974. Chicago: Building, planning and urban technology, 1930–70. Chicago: University of Chicago Press. Coverdale & Colpitts. 1954. Report on estimated traffic and revenue of the proposed City of Chicago Calumet Skyway Toll Bridge. Prepared for Blythe & Co., Inc., and John W. Clarke, Inc. New York: Coverdale & Colpitts. Cronon, W. 1991. Nature’s metropolis: Chicago and the great West. New York: W. W. Norton. Cutler, I. 2006. Chicago: Metropolis of the mid-continent. 3rd ed. Carbondale: Southern Illinois University Press. De Leuw, Cather and Company. 1954. Report on Calumet Skyway Toll Bridge: Engineering studies and estimates. Prepared for Blythe & Co., Inc., and John W. Clarke, Inc. Chicago: De Leuw, Cather and Company. Dyble, L. N. 2009. Reconstructing transportation: Linking tolls and transit for placebased mobility. Technology and Culture 50(3):631–648. ———. 2012. Tolls and control: The Chicago Skyway and the Pennsylvania Turnpike. Journal of Planning History 11(1):70–88. Eisinger, P. 1988. Rise of the entrepreneurial state: State and local economic development policy in the United States. Madison: University of Wisconsin Press. ———. 1998. City politics in an era of federal devolution. Urban Affairs Review 33(3):308–325. Ferman, B. 1996. Challenging the growth machine: Neighborhood politics in Chicago and Pittsburgh. Lawrence: University Press of Kansas. Flyvbjerg, B., N. Bruzelius, and W. Rothengatter. 2003. Megaprojects and risk: An anatomy of ambition. Cambridge, U.K.: Cambridge University Press, 2003. Flyvbjerg, B., M. S. Holm, and S. Buhl. 2002. Underestimating costs in public works projects: Error or lie? Journal of the American Planning Association 68(3):279–295. Fogelson, R. M. 2001. Downtown: Its rise and fall, 1880–1950. New Haven, CT: Yale University Press. Fuchs, E. R. 1992. Mayors and money: Fiscal policy in New York and Chicago. Chicago: University of Chicago Press. Gellert, P. K., and B. D. Lynch. 2003. Megaprojects as displacements. International Social Science Journal 55(175):15–25. Gov. Craig OK’s route for toll road in Indiana. 1953. Chicago Tribune, 2 July. Graham, S., and S. Marvin. 2001. Splintering urbanism: Networked infrastructures, technological mobilities and the urban condition. New York: Routledge. Gutfreund, D. 2005. Twentieth-century sprawl: Highways and the reshaping of the American landscape. New York: Oxford University Press. Hackworth, J. 2007. The neoliberal city: Governance, ideology, and development in American urbanism. New York: Cornell University Press. Hall, T., and P. Hubbard, eds. 1998. The entrepreneurial city: Geographies of politics, regime and representation. New York: Wiley. Harvey, D. 1989. From managerialism to entrepreneurialism: The transformation of urban governance in late capitalism. Geografiska Annaler 71B(1):3–17. Hirsch, A. R. 2005. Martin H. Kennelly: The mugwump and the machine. In The mayors: The Chicago political tradition, 3rd ed., ed. P. M. Green and M. G. Holli. Carbondale: Southern Illinois University Press.

210

Louise Nelson Dyble

Hogan, S., and D. Simpson. 2001. Campaign contributions and mayoral/aldermanic relationships. Urban Affairs Review 37(1):85–95. Hurley, A. 1995. Environmental inequalities: Class, race, and industrial pollution in Gary, Indiana, 1945–1980. Chapel Hill: University of North Carolina Press. Jackson, K. T. 1985. Crabgrass frontier: The suburbanization of the United States. New York: Oxford University Press. Jefferis, C., and F. Stilwell. 2006. Private finance for public infrastructure: The case of Macquarie Bank. Journal of Australian Political Economy 58:44–61. Johnson, C. L., M. J. Luby, and S. I. Kurbanov. 2007. Toll road privatization transactions: The Chicago Skyway and Indiana Toll Road. Bloomington: School of Public and Environmental Affairs, Indiana University. Jones, D. W. 2008. Mass motorization + mass transit: An American history and policy analysis. Bloomington: University of Indiana Press. Kaplan, I. 2012. Does the privatization of publicly owned infrastructure implicate the public trust doctrine? Illinois Central and the Chicago parking meter concession agreement. Northwestern Journal of Law and Social Policy 7(1):136–169. Koval, J. P., L. Bennett, M. I. J. Bennett, F. Demissie, R. Garner, and K. Kim, eds. 2006. The new Chicago: A social and cultural analysis. Philadelphia: Temple University Press. Lawrence, M., and G. P. Stapledon. 2008. Infrastructure funds: Creative use of corporate structure and law—but in whose interests? New York: Riskmetrics Group. Lewis, R. 2008. Chicago made: Factory networks in the industrial metropolis. Chicago: University of Chicago Press. Macquarie Atlas Roads. 2011. Annual report, 2011. Sydney: Macquarie Infrastructure Group. Mayor’s Committee for Economic and Cultural Development. 1966. Mid-Chicago economic development study. 3 vols. Chicago: City of Chicago. McLean, B. 2007. Would you buy a bridge from this man? Fortune Magazine (18 September):138–150. McMullen, J. 1966. The Skyway: High speed road to fiscal ruin. Chicago News, n.d. Chicago Public Library Newspaper File, Calumet Skyway (microfilm). Mier, R., and K. J. Moe. 1991. Decentralized development: From theory to practice. In Harold Washington and the neighborhoods: Progressive city government in Chicago, 1983–1987, ed. P. Clavel and W. Wiewel. New Brunswick, NJ: Rutgers University Press. Miller, D. L. 1996. City of the century: The epic of Chicago and the making of America. New York: Simon & Schuster. Mohl, R. A., and N. Betten. 1986. Steel city: Urban and ethnic patterns in Gary, Indiana, 1906–1950. New York: Holmes and Meier. O’Hara, S. P. 2011. Gary: The most American of all American cities. Bloomington: Indiana University Press. Ranney, D. 2002. Global decisions, local collisions: Urban life in the new world order. Philadelphia: Temple University Press. Rast, J. 1999. Remaking Chicago: The political origins of urban industrial change. DeKalb: Northern Illinois University Press. ———. 2001. Manufacturing industrial decline. Journal of Urban Affairs 23(2):179–190. ———. 2011. Creating a unified business elite: The origins of the Chicago Central Area Committee. Journal of Urban History 37(4):583–605.

chicago and its skyway: lessons from an urban megaproject

211

Ryan, J. R. 1954. Coverdale & Colpitts reaches 50: Still going way out on a limb. New York Times, 28 March: F1. Sassen, S. 2001. The global city: New York, London, Tokyo. 2nd ed. Princeton, NJ: Princeton University Press. Seltzner, M. 1976. Skyway sandblasting hit. Calumet Daily News, 5 June. Simpson, D., and T. Kelly. 2011. The new Chicago school of urbanism and the new Daley machine. In The city, revisited: Urban theory from Chicago, Los Angeles, and New York, ed. D. R. Judd and D. Simpson. Minneapolis: University of Minnesota Press. Smith, C. 2007. The plan of Chicago: Daniel Burnham and the remaking of the American city. Chicago: University of Chicago Press. Smith, N. 1996. The new urban frontier: Gentrification and the revanchist city. London and New York: Routledge. ———. 2008. Uneven development: Nature, capital and the production of space. 3rd ed. Athens: University of Georgia Press. Solomon, L. D. 2009. The promise and perils of infrastructure privatization: The Macquarie model. London: Palgrave Macmillan. Spatz, D. 2010. Roads to postwar urbanism: Expressway building and the transformation of metropolitan Chicago, 1930–1975. Ph.D. diss., University of Chicago. Spivey, J. M. 2001. Historic American engineering record: Chicago Skyway Toll Bridge. Washington, DC: U.S. Department of the Interior, National Park Service. Squires, G. D., L. Bennett, K. McCourt, and P. Nyden. 1989. Chicago: Race, class, and the response to urban decline. Philadelphia: Temple University Press. Suttles, G. D. 1990. The man-made city: The land-use confidence game in Chicago. Chicago: University of Chicago Press. Wachs, M. 1986. Technique vs. advocacy in forecasting: A study of rail rapid transit. Urban Resources 4(1):23–30. Wiewel, W. 1986. The state of the economy and economic development in the Chicago metropolitan region. Chicago: Center for Urban Economic Development, University of Illinois at Chicago. Wille, L. 1997. At home in the Loop: How clout and community built Chicago’s Dearborn Park. Carbondale: Southern Illinois University Press.

commentary Richard G. Little In her chapter, Louise Nelson Dyble develops two interrelated themes that are extremely relevant to contemporary conversations about urban form, infrastructure, and public finance. First, she makes a convincing case that the Chicago Skyway was the wrong project, in the wrong place, at the wrong time. She presents the history of this project very thoroughly and does a nice job of describing how Chicago responded to post–World War II changes in the economy, demographics, and development with a solution more befitting the early twentieth century. On top of that, by rushing the project through planning, design, financing, and construction, Chicago ended up paying for the entire project. Had the city waited a few more years, the federal government would have picked up 90 percent of the cost through the new interstate highway program, and this chapter would never have been written. As what is now termed a “greenfield” toll road, the Skyway was oversold on optimistic traffic projections, which, as Dyble points out and references, is not at all uncommon on projects of this type. On top of that, the Skyway never accomplished its primary purpose of alleviating congestion in the Calumet. In fact, it hastened the physical, economic, and social decline of Chicago’s South Side. Dyble’s second theme is far more interesting and will resonate in U.S. infrastructure and public finance for years to come: that is, the repurposing of underperforming infrastructure to convert a public liability into a financial asset through a concession agreement with the private sector, while still maintaining (and in all likelihood improving) the level of service. The $1.8 billion Skyway concession was a bombshell in the infrastructure business, and for the most part, U.S. policy makers have totally failed to adapt to it even seven years later. Governor Mitch Daniels of Indiana immediately followed up with a similar $3.8 billion concession agreement for the Indiana Toll Road, which was enough to finance most of Indiana’s planned transportation improvements. Since then, however, few similar deals have closed. Chicago, Indianapolis, and Ohio State University have awarded noteworthy concessions for parking, but similar deals collapsed in Los Angeles, Pittsburgh, and Hartford. As the United States faces an ever-increasing backlog of infrastructure maintenance needs with little or no money to pay for them, the primary public policy concern has been to “protect the public interest,” as though inefficient and poorly maintained infrastructure benefited the public. The repurposing of infrastructure assets like the Skyway offers communities a way to leverage the past to build for the future, but there has been great reluctance to pursue this option. This chapter should be asking “Why is this the case?” but does not. Did the Skyway succeed just because of Chicago mayor Richard M. Daley, a strong, authoritarian executive? Governor Daniels was successful, and he is no Richard Daley. Interestingly, Governor Ed Rendell of Pennsylvania, a much more Daleyesque figure than Dan212

commentary

213

iels, was unable to close an extremely lucrative concession for the Pennsylvania Turnpike (to the great relief of investors, as it turned out). Despite the excellent historical treatment of the Skyway and its impacts on urban form and development, the chapter would have been much more effective if Dyble had looked in more depth at the history of the “brownfield” infrastructure concession and why it has worked in Australia and elsewhere better than in the United States. My opinion is that this is mostly a result of the partisan and highly conflicted political discourse currently surrounding public finance, and more specifically infrastructure finance, in the United States. During the latter half of the twentieth century, excise taxes on motor fuels (the “gas tax”) provided the bulk of funding for highways and other transportation projects: a 90 percent federal share in most cases. However, this tax is based on consumption and is not indexed to inflation. As a result, increasing vehicle fuel efficiency and the reluctance of lawmakers to increase the 18.4¢ per gallon levy have gradually depleted the Highway Trust Fund to the point where transfers from the federal general fund have been necessary to maintain solvency. The recently enacted MAP-21 transportation act does nothing to reverse this trend and is evidence of a broader national reluctance to fund infrastructure through general tax measures at the federal level. Faced with this reality, state and local officials like Daley and Daniels have sought to increase revenue for infrastructure (and other public services) by any means possible. The brownfield concession is one such method. One of the criticisms leveled at asset concessions is that the public sector is giving up monetary benefits that governments could reap on their own. Although attractive as a headline, this is an incomplete and mostly incorrect assessment. Yes, the government as owner is turning over future revenue potential to the private sector in exchange for cash today, but is this really the same as Esau selling his birthright for a bowl of porridge? The concession agreements require the concessionaire to maintain the facility in a specified condition, usually far better than the public operator had provided, and to refurbish the facility as necessary. Over the course of a 75- or 99-year concession period, the Skyway and the Indiana Toll Road will essentially need to be rebuilt at least once and perhaps twice. The cost of this ultimately will be borne by the future users of the facility, which is efficient from the standpoint of intergenerational equity. In addition, the risk that the facilities will actually be generating revenue into the twenty-second century will be borne by the concessionaire, not the government. If, in a distant future, people are no longer using individual mobility devices along a dedicated right-of-way, financial viability will be more the concern of private investors than of public institutions. In the conclusion to her chapter, Dyble correctly notes that separating the financial aspects of infrastructure has much broader physical, economic, and social implications. However, the shift away from the tax-allocation model employed at the federal level for the past 50 years has made urban entrepreneurism a financial necessity. Whether this will or should become the normative model

214

Richard G. Little

for infrastructure finance for the next 50 years remains to be seen. At the very least, the days of cheap federal money are effectively over, and U.S. governments at all levels will need to become both more innovative in how they provide services and more efficient in how they pay for them. Overall, this chapter is a good beginning to sorting out a complex series of public actions that are still playing out and that could truly become a routine method of capturing stranded public investment and allowing the funds to be reallocated to today’s needs, whether for infrastructure maintenance and renewal or for other purposes.

8 Assessing the Infrastructure Impact of Mega-Events in Emerging Economies Victor A. Matheson

S

porting mega-events like the Summer and Winter Olympic Games and soccer’s World Cup focus the world’s attention on the city or country hosting the event, and the competition among cities and countries to host these events is often as fierce as the competition on the playing field. Increasingly, developing countries have entered the bidding process, chasing after the riches and the glory that presumably accrue to the city where the events will take place. However, with great events come great responsibilities, and the cost of operating, organizing, and building infrastructure for the Olympic Games or World Cup can be daunting. From an economic standpoint, the question is whether mega-events represent a good investment for developing countries; this chapter addresses that question. The modern Summer Olympics began in 1896 and take place every four years at a location selected through an elaborate bidding process well before the event. The Winter Olympics, held since 1924, follow an identical procedure. In recent times, the host cities for both the summer and winter games have been selected six or seven years before the events are to take place. Historically, hosting the Olympic Games has been almost exclusively the domain of rich, industrialized nations. Between 1896 and 1952, all of the summer and winter games were held in either Western Europe or the United States, with cities in Japan, Canada, and Australia joining the mix over the next two decades (table 8.1). In 1968 Mexico City was the first city outside the industrialized world to host the games. Eastern European countries were awarded the summer games in 1980 (Moscow) and the winter games in 1984 (Sarajevo, Yugoslavia). Seoul, South Korea, was selected to host the summer games in 1988, a time when South Korea might have been classified as rapidly industrializing rather than industrialized. Shortly after the 215

Table 8.1 Hosts of the Summer and Winter Olympic Games and FIFA World Cup Year

Summer Olympics

Winter Olympics

1896 1900 1904 1908 1912 1916 1920 1924 1928 1930 1932

Athens, Greece Paris, France St. Louis, United States London, United Kingdom Stockholm, Sweden Not held Antwerp, Belgium Paris, France Amsterdam, The Netherlands

Not held Not held Not held Not held Not held Not held Not held Chamonix, France St. Moritz, Switzerland

Los Angeles, United States

Lake Placid, United States

Berlin, Germany

Garmisch, Germany

Not held

Not held

Not held

Not held

1934 1936 1938 1940 1942 1944 1946 1948 1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972

216

World Cup

Uruguay

Italy France Not held Not held London, United Kingdom

St. Moritz, Switzerland Brazil

Helsinki, Finland

Olso, Norway Switzerland

Melbourne, Australia

Cortina, Italy

Rome, Italy

Squaw Valley, United States

Tokyo, Japan

Innsbruck, Austria

Mexico City, Mexico

Grenoble, France

Sweden

Chile England Mexico Munich, Germany

Sapporo, Japan (continued)

infrastructure impact of mega-events in emerging economies

217

Table 8.1 (continued) Year 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2022

Summer Olympics

Winter Olympics

World Cup Germany

Montreal, Canada

Innsbruck, Austria

Moscow, Soviet Union

Lake Placid, United States

Los Angeles, United States

Sarajevo, Yugoslavia

Seoul, South Korea

Calgary, Canada

Barcelona, Spain

Albertville, France Lillehammer, Norway

United States

Nagano, Japan

France

Salt Lake City, United States

South Korea/ Japan

Turin, Italy

Germany

Vancouver, Canada

South Africa

Sochi, Russia

Brazil

Argentina

Spain Mexico Italy

Atlanta, United States Sydney, Australia

Athens, Greece Beijing, China London, United Kingdom Rio de Janeiro, Brazil Russia Qatar

Olympics, however, the country was admitted to the Organization for Economic Cooperation and Development (OECD), a sort of de facto dividing line between industrialized and developing nations. More recently, the International Olympic Committee (IOC) has encouraged bids from developing countries and has awarded the games on several occasions to nontraditional countries outside the OECD. The 2008 summer games were hosted by China, and the 2016 Summer Olympics will be played in Rio de Janeiro, the first time the event has taken place in South America. The 2014 Winter

218

Victor A. Matheson

Table 8.2 Number of Bids for Summer and Winter Olympic Games Event Summer Olympics: 1896–1996 Summer Olympics: 2000–2016 Winter Olympics: 1924–1998 Winter Olympics: 2002–2014

Bids from Industrialized Countries

Bids from Developing Countries

Bids from Eastern Bloc or Former Soviet States

71 (82%)

9 (10%)

7 (8%)

21 (49%)

19 (44%)

3 (7%)

51 (93%)

1 (2%)

3 (5%)

18 (56%)

3 (9%)

11 (34%)

Olympics will take place in Sochi, Russia, leaving Western Europe, North America, and Japan for only the second time. As seen in table 8.2, the list of countries submitting formal bids has changed dramatically in recent decades. Eighteen percent of the bids submitted for the summer games prior to 2000 came from outside Western Europe, Japan, Australia, Canada, and the United States. Since 2000, however, more than half of all bids have come from this group, including applications by Istanbul, Bangkok, Havana, Buenos Aires, and Cape Town, as well as the successful bids from Beijing and Rio. For the Winter Olympics, the past decade has witnessed bids from Kazakhstan, Georgia, China, Slovakia, and Poland for the first time. The world’s other major international mega-sporting event is the Fédération Internationale de Football Association (FIFA) World Cup. This event takes place every four years, like the Olympics, and features national soccer teams. The World Cup1 began in 1930 in response to soccer’s growing prominence in the Olympics. Due to the number of large stadiums required to accommodate the tournament, FIFA selects a host country for the event, as opposed to the IOC’s tradition of choosing a single host city. As shown in table 8.1, for the first 60 years of the competition, the World Cup essentially alternated between the two centers of soccer interest, Europe and Latin America; unlike the Olympics, numerous countries in Central and South America have hosted the World Cup, including Uruguay, Brazil, Chile, Argentina, and Mexico.

1. Other international sporting organizations, notably in cricket and rugby, also host similar international tournaments that are dubbed “the World Cup.” These events are typically smaller than the FIFA World Cup, and for the purposes of this chapter, the term World Cup is meant to describe the soccer tournament unless specifically noted otherwise.

infrastructure impact of mega-events in emerging economies

219

This rotation scheme lasted until 1994, when FIFA, in an attempt to expand world interest in the game, awarded the World Cup to the United States, a huge untapped market for the sport. Japan and South Korea followed in 2002, the first tournament cohosted by two countries and the first World Cup played in Asia. More “firsts” followed. South Africa became the first African host in 2010, Russia will become the first Eastern European host in 2018, and Qatar, a nation with no domestic soccer league and little soccer history or tradition, will become the first Middle Eastern host in 2022. In 2014 the World Cup will return to a Latin American country for the first time in nearly 30 years when Brazil hosts the event. Economically, the world’s attention has increasingly shifted from the socalled G-7 nations—which include the world’s largest industrialized economies, such as the United States, Japan, the United Kingdom, and Germany—to the BRICS nations, an acronym for the five rapidly developing nations of Brazil, Russia, India, China, and South Africa. When the 2010 Commonwealth Games hosted by India are included, each of these countries will have held at least one of the world’s top sporting events between 2008 and 2018. The shift to a more egalitarian system of awarding mega-events to nontraditional hosts has it proponents. Supporters of South Africa’s failed bid to host the 2006 World Cup were bitterly disappointed with the controversial selection of Germany as the host nation. With the growing interest in soccer throughout Africa, it was thought that the continent deserved a chance to host the tournament, and proponents of hosting a South African World Cup also pointed to the potential for large economic benefits that would accrue to the country. However, an in-depth analysis of both the short-run and long-run economic impact of hosting mega-events demonstrates that in a direct economic sense, the World Cup is something of a poisoned chalice. Similarly, the Olympics often prove to be an expensive burden, providing a short-run economic boost that is well below what the event’s proponents typically predict—and few long-run economic benefits.

Short-Run Benefits Mega-events undoubtedly result in significant tourism expenditures, but in the vast majority of cases the observed increases in economic activity fall well short of the economic impact predicted by event organizers. Table 8.3 shows commissioned ex ante economic impact studies for various Olympic and World Cup events. Table 8.4 shows ex post estimates of economic impact performed by independent economists examining actual economic data before, during, and after the events. In most cases, the independent economists have found little or no direct economic impact of mega-events on host communities. The disconnect between ex ante predictions and ex post reality results from numerous factors. As authors like Matheson (2008) indicate, economic impact studies may be based on inflated, unrealistic, or best-case predictions, but even

220

Victor A. Matheson

Table 8.3 Examples of Mega-Event ex ante Economic Impact Studies Event

Year

Impact

Source

World Cup (Japan)

2002

$24.8 billion

World Cup (South Korea)

2002

$8.9 billion

World Cup (South Africa) World Cup (South Africa) Summer Olympics (Atlanta) Winter Olympics (Vancouver)

2010 2010 1996 2010

$7.5 billion, 198,400 jobs $12 billion, 483,000 visitors $5.1 billion, 77,000 jobs C$10.7 billion, 244,000 jobs

Dentsu Institute for Human Studies; Finer (2002) Dentsu Institute for Human Studies; Finer (2002) Grant Thornton SA; Rihlamvu (2011) Grant Thornton SA; Voigt (2010) Humphreys and Plummer (2005) InterVISTAS Consulting (2002)

Table 8.4 Examples of Mega-Event ex post Economic Impact Studies Event

Year

Variable

Impact

Source

Summer Olympics (Atlanta) 1996 Employment Summer Olympics (Atlanta) 1996 Employment

3,500–42,000 jobs Approx. 75,000

Winter Olympics (Salt Lake City) Winter Olympics (Salt Lake City) World Cup (United States)

2002 Employment

4,000–7,000 jobs

2002 Retail sales 1994 Employment

Positive, hotels Negative, retailers Not statistically significant

World Cup (Germany) World Cup (United States) World Cup (Germany) World Cup (Germany)

2006 1994 2006 2006

Not statistically significant Down $4 billion Not statistically significant Not statistically significant

Baade and Matheson (2002) Feddersen and Maennig (forthcoming) Baumann, Engelhardt, and Matheson (2012a) Baade, Baumann, and Matheson (2010) Baumann, Engelhardt, and Matheson (2012b) Allmers and Maennig (2009) Baade and Matheson (2004) Allmers and Maennig (2009) Allmers and Maennig (2009)

Employment Personal income Personal income Employment

when appropriate data are used, many economic impact estimates suffer from three features that exaggerate the numbers. First, any money spent by local sports fans is money not being spent by these residents elsewhere in the local economy. Spending by local citizens does not represent new money in the economy; rather, it is simply money that is reallocated within the city or country. Crowds of local fans cheering for the home team might make for a festive atmosphere, but it does little to encourage new spending in the economy or to promote economic growth.

infrastructure impact of mega-events in emerging economies

221

Second, money spent in a local economy during a mega-event may not remain in the local economy. Mega-events are frequently characterized by capacity constraints and high prices for accommodations and other services. Hotel rooms can frequently sell at three or four times their normal rates during mega-events, but the desk clerks and room cleaners who service these establishments generally do not see their wages triple or quadruple. Thus, the tourist industry should see an increase in returns to capital, but to the extent that hotels or other service industries are owned by individuals outside the local economy, event spending leaks out of the host economy. Third, sports fans can crowd out regular visitors, displacing economic activity that would have occurred in the absence of the sporting event. While a city’s hotels and restaurants may be full of sports fans during a tournament, had those same hotel rooms and restaurants been full of business travelers or other vacationers in the absence of the mega-event, the tournament would not have resulted in a net increase in economic activity. Yogi Berra’s famous quote “No one goes there anymore—it’s too crowded” is largely true when applied to tourism and mega-events. An examination of tourist arrivals in South Africa around the time of the 2010 World Cup illustrates these issues. The 64 games of the tournament attracted an average of 49,670 spectators per match, for a total of nearly 3.2 million fans. As noted previously, only foreign visitors should be included in any economic impact estimates. In addition, many fans are likely to attend more than one game, so the number of people that should be included in any impact figures is likely to be significantly below 3.2 million. The consulting firm Grant Thornton South Africa initially predicted 483,000 international visitors for the 2010 World Cup but later revised their figures downward to 373,000. Even this number turned out to be too optimistic. FIFA reported that just “309,554 foreign tourists arrived in South Africa for the primary purpose of attending the 2010 FIFA World Cup” and that they spent 3.64 billion rand (US$482 million) during their stay (FIFA 2010). Thus, the substitution effect, combined with overly rosy attendance figures, reduced 3.2 million fans in the stadiums to just 310,000 actual overseas visitors. The bad news for South Africa does not stop there. Total tourist arrivals in June and July 2010 were only 273,000 above the same months the year before, suggesting a degree of crowding out. Furthermore, 2009 was a particularly poor year for tourism to South Africa due to the worldwide economic crisis. Econometric analysis of tourist arrivals suggests an increase of only 123,000 to 202,000 above what would have been expected with the World Cup (Matheson, Peeters, and Szymanski 2012). Visitor numbers like these are unlikely to be sufficient to cover the high costs of putting on a mega-event of this magnitude. South Africa’s experience is far from unique. Beijing reported total visitor numbers in August 2008 during the Summer Olympics similar to those in the same month the previous year. Shops, restaurants, and tourist attractions outside the areas immediately adjacent to the Olympic venues in London reported a tourist drought during the 2012 summer games (CBC 2012).

222

Victor A. Matheson

Short-Run Costs Hosting mega-events can be an enormously expensive affair, and governing bodies like the IOC and FIFA typically require that host countries bear most of the costs. The Olympics require very specific sports infrastructure in order to accommodate a large range of events. For the World Cup, FIFA requires host countries to have at least 12 modern stadiums capable of seating at least 40,000 spectators; one of the stadiums must seat at least 80,000 for the opening and final matches. Operating costs can be massive, in large part due to the extensive security requirements that mega-events require. The security budget alone for the Athens Olympics in 2004 ran to more than $1.5 billion, nearly six times the budget for the Sydney games just four years earlier. For the 2010 FIFA World Cup, South Africa bore $3.9 billion in expenses, including at least $1.3 billion in stadium construction costs (Baade and Matheson 2012; Voigt 2010). Costs for Brazil’s 2014 World Cup are estimated to exceed $10 billion. As is common in sporting events, costs have escalated drastically in just a few short years: Back in 2009, the Brazilian Football Confederation estimated the 12 stadiums being refitted or built for the World Cup would cost about 2.2 billion reais [US$1.14 billion]—a figure that two years later seems quaint. The government now sees them costing more than triple that, at 6.9 billion reais [US$4.1 billion]. (Grudgings 2011)

Table 8.5 shows the sports infrastructure, other infrastructure, and operational spending for various recent mega-events. Full information is not available for all events. Sports infrastructure includes spending on stadiums and sports venues, while other infrastructure includes construction costs for transportation, tourist and athlete accommodations, and public spaces. The dividing line between sports infrastructure and other infrastructure is not entirely clear. For example, 20 percent of the total budgeted cost for London’s new Wembley Stadium was $150 million in general infrastructure improvements, including new roads and a renovated Underground station designed to better accommodate stadium traffic. While the roads and subway station are clearly not a part of the stadium, without the stadium they would not have been required (Matheson 2008). The entire Wembley project, which played a significant role in the 2012 London summer games, ended up costing 798 million pounds (US$1.24 billion) (2007), over twice its original budget—yet another example of optimistic accounting in sporting events. Given the huge costs associated with mega-events and the relatively small number of visitors, it is virtually impossible for the direct revenues associated with these events to cover the expenses. This is less true if little new infrastructure is needed. For example, total infrastructure costs for the 1994 World Cup held in the United States were only $30 million, as the existing stadiums were more than adequate for the event. Similarly, the 1984 summer games in Los Angeles made a

infrastructure impact of mega-events in emerging economies

223

Table 8.5 Costs of Hosting Mega-Events Event

Year

Type

Spending (millions, 2011 $)

Source Preuss (2004)

2006 Sports infrastructure 2010 Sports infrastructure

$2,856 $4,870 $1,731 $14,517 $798 $999 $1,672 $1,725 $13,813 $1,758 $45,000 $15,000–$20,000 Over $14,000 $4,100 C$5,900 $10,000 (est.) $2,000 (S. Korea) $4,000–$5,600 (Japan) $1,870 $1,300

Total 2014 Sports infrastructure General infrastructure 2018 Total

$3,900 $3,680 $13,000 (est.) $10,000 (est.)

Summer Olympics (Seoul)

1988 Sports infrastructure General infrastructure Summer Olympics (Barcelona) 1992 Sports infrastructure General infrastructure Summer Olympics (Atlanta) 1996 Sports infrastructure General infrastructure Summer Olympics (Sydney) 2000 Sports infrastructure General infrastructure Summer Olympics (Athens) 2004 Total cost Summer Olympics (Beijing) 2008 Sports infrastructure Total spending (est.) Summer Olympics (London) 2012 Total cost Winter Olympics (Nagano) 1998 Total cost Winter Olympics (Turin) 2006 Total cost Winter Olympics (Vancouver) 2010 Total cost Winter Olympics (Sochi) 2014 Total cost World Cup 2002 Sports infrastructure (South Korea/Japan) World Cup (Germany) World Cup (South Africa)

World Cup (Brazil) World Cup (Russia)

Preuss (2004) Preuss (2004) Preuss (2004) Preuss (2004) Preuss (2004) Baade and Matheson (2012) Burns (2012) Longman (1998) Payne (2008) Economist (2011) Estimates, very preliminary Sloan (2002)

Downie (2012) Baade and Matheson (2012); Voigt (2010) Downie (2012) Estimates, very preliminary

large profit for the organizers, again because existing facilities were used for most events. Given the post-9/11 need for increased security, however, it is uncertain whether a mega-event today would have short-run net benefits for the host, even with no capital outlays. Thus, economic rationality rests on the long-run effects of the events in terms of branding or economic growth based on infrastructure legacies.

224

Victor A. Matheson

Long-Run Benefits While the short-run tourism boost that mega-events provide is clearly limited, especially in relation to the large expenses involved, event organizers typically claim that mega-events result in a lasting legacy that will provide significant economic benefits for many years to come. Just as the short-run benefits of mega-events are overblown, so too are the claims of long-run benefits from sports infrastructure. Supporters often claim that stadiums and sports facilities can serve as an anchor to promote local economic development. They envision stadiums serving as an integrated component of a thriving and diverse local economy. One example of this economic model is the Wrigleyville neighborhood on the north side of Chicago, home to Major League Baseball’s Chicago Cubs. Wrigley Field, the second-oldest major league sports stadium in the United States (behind Boston’s venerable Fenway Park), was built in 1914 and rests comfortably within the existing street grid. The Cubs generate significant spillover effects for the surrounding community by attracting sports fans to the area. The 81-game season brings into the local neighborhood roughly 3 million baseball fans who frequent bars, restaurants, and souvenir shops both before and after home games. Figure 8.1 clearly shows how Wrigley Field serves to promote local businesses. A thriving entertainment district has grown up around the stadium, with dozens of eating and drinking establishments within just a few blocks of the Cubs’ home. Unfortunately for proponents of sports-based economic development, Wrigley Field is the exception rather than the rule. Just 10 miles south of Wrigley is U.S. Cellular Park, home of the Chicago White Sox, Chicago’s other Major League Baseball team. Built in 1992 to replace the aging Comiskey Park, U.S. Cellular is more in line with most modern stadiums designed to maximize instadium revenue. As exemplified by U.S. Cellular Park (figure 8.2), the modern stadium is like a walled fortress with a moat of parking lots driving fans inside the castle and away from the barbarian hordes of shops and businesses in the local neighborhood. Indeed, most studies on the economic benefits of new stadiums have found little or no positive impact on local economies (Baade 1996; Coates and Humphreys 1999, 2008), although neighborhood effects are evident in some cases (Feng and Humphreys 2008; Tu 2005). Most studies of stadium economics have examined facilities in the United States and to a lesser extent in Europe. If the economics are poor for facilities in the industrialized world, they are even worse in developing countries. Rich countries usually have well-developed professional sports leagues, meaning that in many cases existing sports infrastructure can be used, and many new facilities can be put to use after the event. For example, currently all 12 of the stadiums used in Germany in the 2006 World Cup are regularly filled to capacity by the Bundesliga soccer teams that have become full-time tenants. In contrast, the South African Premier Soccer League averages only 7,500 fans per match, hardly the crowds for which the World Cup stadiums were designed. Other events at South African stadiums have rarely filled the venues. Atlanta’s newly constructed

infrastructure impact of mega-events in emerging economies

225

Figure 8.1 Wrigley Field

Source: Baade, Matheson, and Nikolova (2007). Reprinted courtesy of Geographische Rundschau International Edition.

Centennial Olympic Stadium was renovated after the 1996 games and is currently home to Major League Baseball’s Atlanta Braves, while Beijing’s National Stadium (better known as the “Bird’s Nest”) sits largely unused. Without regular, well-attended events at the newly constructed sports facilities, the stadiums are unlikely to give rise to urban development in their local neighborhoods. Indeed, an overhead image (figure 8.3) of the area in Beijing around the Bird’s Nest and the National Aquatic Center (or “Water Cube”) shows a beautifully landscaped area but little automobile or pedestrian traffic and few new businesses. Similarly, a view of Soccer City (figure 8.4) on the outskirts of Johannesburg, South Africa, the site of the 2010 World Cup Final, shows a string of administrative buildings next to the stadium but little else. For the most part, new stadiums in developing countries mirror the experience of Chicago’s U.S. Cellular Park, not the more development-friendly Wrigley Field. Sports facilities are generally quite difficult to convert to other uses. Housing for athletes or officials can be easily converted to residential facilities for students

226

Victor A. Matheson

Figure 8.2 U.S. Cellular Park

Source: Baade, Matheson, and Nikolova (2007). Reprinted courtesy of Geographische Rundschau International Edition.

or other residents, as was done in Atlanta following the 1996 Summer Olympics and in Los Angeles in 1984. Such conversions are rare, however, for athletic venues. The famous Water Cube in Beijing, home to most of the aquatic events in the 2008 summer games, was opened for public swimming the next year, making it the world’s most expensive lap pool. It subsequently underwent significant renovations and reopened as a large water park. While that is a fine long-term use for an otherwise underutilized venue, it is also an extraordinarily expensive way to build a water park. If the creation of new or improved sports infrastructure cannot be seen as a savior for mega-events, then one is left to appeal to the creation of other infrastructure as an economic justification for hosting mega-events. As can be seen in table 8.5, non-sports-related infrastructure expenditures often far exceed spending on sports venues, and unlike sports venues, expenditures on transportation networks and other types of general infrastructure can encourage future growth. Mega-events can serve as an impetus to engage in needed infrastructure invest-

infrastructure impact of mega-events in emerging economies

227

Figure 8.3 Bird’s Nest, Water Cube, and Olympics Sports Center in Beijing

Source: Reprinted courtesy of Astrium GEO-Information Services.

ments held back by a lack of political will. Brazil, for example, is engaging in massive investment spending in its run-up to the 2014 World Cup and 2016 Summer Olympics. The words of Brazilian Football Confederation president Ricardo Teixeira echo those of many proponents of mega-events: Over the next few years we will have a consistent influx of investments. The 2014 World Cup will enable Brazil to have a modern infrastructure. In social terms it will be very beneficial. . . . Our objective is to make Brazil become more visible in global arenas. The World Cup goes far beyond a mere sporting event. It’s going to be an interesting tool to promote social transformation. (CNN 2007)

There is an element of truth to Teixeira’s words; however, two caveats are in order. First, spending millions or billions of dollars on unproductive sports infrastructure simply in order to have the political will for needed infrastructure investments is a distinctly second-best economic strategy. Public capital would be more efficiently allocated if governments would simply make reasonable public investment choices without a mega-event hanging over their heads. In addition,

228

Victor A. Matheson

Figure 8.4 Soccer City Near Johannesburg

Source: KARI 2010/Distribution Spot Image. Reprinted courtesy of Astrium GEO-Information Services.

mega-events can place surprisingly tight deadlines on major public works projects. These deadlines can increase costs due to rushed schedules, relaxed bidding rules, and potential corruption. Finally, preparations for a mega-event can result in too high a level of investment in non-athletic infrastructure. An airport, transportation network, and hotel capacity that are the right size for three weeks of tourist insanity may be extensively overbuilt for the post-event period. For example, two major luxury hotels built for the 1994 Winter Olympics in Lillehammer, Norway, filed for bankruptcy shortly after the close of the games. The final potential benefit of mega-events is that they can serve to “put the host on the map,” leading to higher levels of future tourism, trade, and investment. As noted by Matheson, The other major intangible benefit of mega-events claimed by sports boosters is that of national and international exposure. Sports fans may enjoy their visit to the city and return later raising future tourist revenues for the area. Corporate visitors, it is claimed, may relocate manufacturing facilities and company headquarters to the city. Television viewers might

infrastructure impact of mega-events in emerging economies

229

decide to take a trip to the host city at some time in the future based on what they see during the broadcast of the mega-event. Finally, hosting a major event might raise perceptions of the city so that it becomes a “world class” city and travel destination. All of these claims are potentially true although little empirical research has conclusively demonstrated any long-run connections between hosting mega-events and future tourism demand. There are not even any anecdotal examples of companies moving corporate operations to a city based on the hosting of a sporting event. (2008, 86)

There are individual cases where mega-events do seem to have had a major influence on future demand, but it appears that a “perfect storm” is needed. Cities that are already on everyone’s map, like London, gain little in exposure from a major event because they are already at nearly maximum exposure. Other cities like Atlanta and many Winter Olympics hosts also gain little from exposure because they have little to offer potential tourists. Advertising without a subject to advertise is largely ineffective. In a perfect situation, a “hidden gem” can raise its international profile with the right situation. This appears to have been the case with Barcelona, a city with great artistic, cultural, and architectural treasures that had long been overshadowed by European capitals and 40 years of fascist rule. By 2012, two decades after its moment on the world stage, Barcelona was the fourth most visited city in Europe. Barcelona’s tourism experience, however, has not been shared by most Olympic hosts. Rose and Spiegel (2010) find that international trade increases significantly when a country hosts a major event. Typically, this would lend strong evidence to the idea that the Olympics or the World Cup has a large advertising effect, but the authors also find that the simple act of bidding for the Olympics increases capital inflows. They chalk this up to a signaling effect: bidding for the Olympics lets other countries know that the nation is “open for business.” If Rose and Spiegel’s findings are truly more than spurious correlation, the findings of other economists suggest that an optimal strategy would be to bid for the Olympics but not win them. Subsequent analysis of foreign trade flow, however, indeed suggests that Rose and Spiegel’s findings are likely the result of selection bias. Countries that are in a position to bid for the Olympics are typically the sort of rich, growing countries that generally experience trade growth. When Olympic hosts and bidders are compared to otherwise similar countries that did not bid for the games, the so-called Olympic effect disappears (Maennig and Richter 2012). It should also be noted that the presence of a mega-event may bring with it intangible costs as well as benefits. For example, the publicity associated with a sporting event may not place a city in a positive light. The bribery scandal that surrounded the 2002 Winter Olympics in Salt Lake City certainly didn’t enhance the city’s reputation. Similarly, the international reputations of Munich and Atlanta were tarnished by the terrorist events that occurred during the Olympic Games in those cities.

230

Victor A. Matheson

Of course, sporting events have been used for centuries to provide entertainment for the masses. The term bread and circuses dates from the first-century Roman Empire when extravagant games were held in conjunction with giveaways of subsidized food in order to pacify the citizenry and reduce urban unrest. Sports boosters often cite civic pride or national exposure as a primary benefit of mega-events and of sports in general. In many cases, it is undoubtedly true that mega-events bring intangible psychological value to the communities that host them. The 1995 Rugby World Cup in South Africa represented an opportunity for the country to announce its reemergence as a full member of not only the world’s sporting community but also its political community. The picture of South African president Nelson Mandela wearing the jersey of the white South African captain Francois Pienaar while presenting him with the championship trophy was a powerful image to the world, indicating that South Africa had emerged from its years of racial oppression, and served to unify the country (Baade and Matheson 2004). Similarly, New Orleans mayor Ray Nagin pointed to the return of the National Football League to the city in September 2006 as an important symbol to the rest of the country that the city was fully on the road to recovery from Hurricane Katrina, which had struck the year before. Allmers and Maennig (2009) also found that the largest identifiable effect from the 2006 World Cup in Germany was a “feel-good” effect: a clear increase in self-reported happiness among German residents.

Conclusions Empirical research into the true economic impact of mega-events on host economies tends to show that major sporting events bring high costs with low rewards. The return in developing nations may be even lower. Probably the best that can be said for mega-events is that they allow governments to overcome political constraints and make beneficial infrastructure investments. However, overcoming these political constraints comes at a very high cost in terms of money spent on unproductive investments in sports infrastructure and tournament operations, and there is no guarantee that all of the general infrastructure investments will provide a net positive return for the cities involved. While the recent trend has been to “reward” developing countries with the opportunity to host mega-events such as the World Cup and the Olympics, the empirical evidence suggests that if rich countries want to promote economic development in poor countries, it would make more sense to keep these events out of the developing world and instead continue to award the games to rich countries that are better able to absorb the associated costs. Alternatively, the industrialized world could subsidize these events when they are held in poor countries through sponsorship or by direct foreign assistance. However, it seems unlikely that rich countries would be willing to aid poor countries when they are often in direct competition with one another for the rights to host in the first place.

infrastructure impact of mega-events in emerging economies

231

It remains a widespread belief among countries that substantial national gains result from hosting these global events, but the evidence indicates that this is rarely the case. Samuel Johnson once wrote that second marriages reflect “the triumph of hope over experience.” Such thinking pervades the vigorous competition among countries to host these exciting but economically questionable events.

references Allmers, S., and W. Maennig. 2009. Economic impacts of the FIFA Soccer World Cups in France 1998, Germany 2006, and outlook for South Africa 2010. Eastern Economic Journal 35(4):500–519. Baade, R. 1996. Professional sports as a catalyst for metropolitan economic development. Journal of Urban Affairs 18(1):1–17. Baade, R., R. Baumann, and V. Matheson. 2010. Slippery slope: Assessing the economic impact of the 2002 Winter Olympic Games in Salt Lake City, Utah. Région et Développement 31:81–91. Baade, R., and V. Matheson. 2002. Bidding for the Olympics: Fool’s gold? In Transatlantic sport: The comparative economics of North American and European sports, ed. C. Barros, M. Ibrahimo, and S. Szymanski, 127–151. London: Edward Elgar. ———. 2004. The quest for the cup: Assessing the economic impact of the World Cup. Regional Studies 38(4):343–354. ———. 2012. Financing professional sports facilities. In Financing for local economic development, 2nd ed., ed. Z. Kotval and S. White, 323–342. New York: M. E. Sharpe. Baade, R., V. Matheson, and M. Nikolova. 2007. A tale of two stadiums. Geographische Rundschau International Edition 3(1):53–58. Baumann, R., B. Engelhardt, and V. Matheson. 2012a. Employment effects of the 2002 Winter Olympics in Salt Lake City, Utah. Journal of Economics and Statistics 232(3):308–317. ———. 2012b. Labor market effects of the World Cup: A sectoral analysis. In International handbook on the economics of sporting mega events, ed. A. Zimbalist and W. Maennig, 385–400. Cheltenham, U.K.: Edward Elgar. Burns, J. 2012. After warnings of an Olympic crush, businesses suffer in a deserted London. New York Times, 2 August. CBC. 2012. London businesses bemoan Olympic slump. CBC, 2 August. www.cbc.ca /news/world/story/2012/08/02/london-olympics-business-slump-ghost-town.html. CNN. 2007. Brazil to host 2014 World Cup. CNN, 30 October. http://edition.cnn .com/2007/SPORT/football/10/30/brazil.cup. Coates, D., and B. Humphreys. 1999. The growth effects of sports franchises, stadia, and arenas. Journal of Policy Analysis and Management 14(4):601–624. ———. 2008. Do economists reach a conclusion on subsidies for sports franchises, stadiums, and mega-events? Econ Journal Watch 5(3):294–315.

232

Victor A. Matheson

Downie, A. 2012. Soccer: Brazil World Cup stadiums on track, but costs soar. Reuters, 3 April. www.reuters.com/article/2012/04/03/soccer-world-brazil-idUSL2E8F2GG 820120403. Economist. 2011. Up false creek: The cost of a property deal gone sour. Economist, 13 January. www.economist.com/node/17906069. Feddersen, A., and W. Maennig. Forthcoming. Mega-events and sectoral employment: The case of the 1996 Olympic Games. Contemporary Economic Policy. Feng, X., and B. Humphreys. 2008. Assessing the economic impact of sports facilities on residential property values: A spatial hedonic approach. Working Paper No. 08-12. International Association of Sports Economists. FIFA. 2010. Study reveals tourism impact in South Africa. 7 December. www.fifa.com /worldcup/archive/southafrica2010/news/newsid=1347377/index.html. Finer, J. 2002. The grand illusion. Far Eastern Economic Review (7 March):32–36. Grudgings, S. 2011. Brazil’s World Cup rush fuels spending blowout. Reuters, 27 September. www.reuters.com/article/2011/09/28/us-brazil-worldcup-idUSTRE78R01D 20110928. Humphreys, J., and M. Plummer. 2005. The economic impact on the state of Georgia of hosting the 1996 Summer Olympic Games. Athens: Selig Center for Economic Growth, University of Georgia. InterVISTAS Consulting. 2002. The economic impact of the 2010 Winter Olympics and Paralympic Games: An update. Victoria: British Columbia Ministry of Competition, Science and Enterprise. Longman, J. 1998. Nagano 1998: Seven days to go—High costs and high expectations. New York Times, 30 January. Maennig, W., and F. Richter. 2012. Exports and Olympic Games: Is there a signal effect? Journal of Sports Economics 13(6):635–641. Matheson, V. 2008. Mega-events: The effect of the world’s biggest sporting events on local, regional, and national economies. In The business of sports, vol. 1, ed. D. Howard and B. Humphreys, 81–99. Westport, CT: Praeger. Matheson, V., T. Peeters, and S. Szymanski. 2012. If you host it, will they come? Megaevents and tourism in South Africa. Paper No. SPO2012-0037. Athens, Greece: ATINER’S Conference Paper Series. Payne, B. 2008. The Olympics effect: When the games are over, which cities win big—and which stumbled? MSNBC, 3 August. http://today.msnbc.msn.com// id/26042517#.UBwk104gcsd. Preuss, H. 2004. Economics of the Olympic Games. London: Edward Elgar. Rihlamvu, E. 2011. 2010 FIFA Soccer World Cup. Africa Travel. www.africa-ata .org/sports.htm. Rose, A., and M. Spiegel. 2010. The Olympic effect. Economic Journal 121(553): 652–677. Sloan, D. 2002. Cup offers Japan economic free kick. XtraMSN, 25 January. http:// xtramsa.co.nz/sport/0,,3951–1071885,00.html. Tu, C. C. 2005. How does a new sports stadium affect housing values? The case of FedEx Field. Land Economics 81(3):379–395. Voigt, K. 2010. Is there a World Cup economic bounce? CNN.com, 11 June. http:// edition.cnn.com/2010/BUSINESS/06/11/business.bounce.world.cup/index.html.

commentary David E. Luberoff Reflecting on Victor A. Matheson’s stimulating chapter, I find the phrase “old wine in new bottles” fitting in two different ways. First, for a student of urban mega-projects, particularly in transportation, Matheson’s assertions, analyses, and conclusions on the minimal economic value of mega-events have a familiar ring. Second, the chapter repeats familiar laments from students of public investment policies who regularly ask, “Why do seemingly rational actors keep acting in irrational ways?”

Are Mega-Events Good Investments? While backers of mega-events tout their economic benefits, Matheson notes that independent, ex post analyses generally show that these events rarely make good investments. Rather, he reports that “in most cases, independent economists have found little or no direct economic impact of mega-events on host communities.” This conclusion closely echoes both the rhetoric about and analyses of publicly supported stadiums and arenas in the United States (Danielson 2001; Long 2012; Zimbalist 2003). Given the lack of direct, short-term economic payoffs, Matheson suggests three longer-term rationales for mega-events: promoting economic development, spurring needed infrastructure investments, and putting places on the global map. The desire to significantly transform urban areas is clearly a key factor driving local efforts to host the Olympic Games. For example, the 2008 Beijing Olympics reportedly resulted in the displacement of approximately 1.5 million people (Centre on Housing Rights and Evictions 2007). Less dramatically, but still quite ambitiously, while the 2012 London Olympics made use of many existing facilities (such as Wembley Stadium), the games were also used to spur the redevelopment of the Lea Valley, a major brownfield site. In fact, while this pattern is not ubiquitous, it is increasingly common, having started with the 1960 Olympics in Rome and having been at least part of the agenda of many subsequent Olympic bids (Gold and Gold 2008). Table 8.5, which shows that spending on non-sports-related infrastructure often far exceeds spending on sports infrastructure, is one of the most striking parts of Matheson’s chapter. As Matheson notes, this is, at best, a poor way to make infrastructure investment decisions. More worrisome is that the history of infrastructure investment suggests that, at worst, it is an extraordinary opportunity for rent-seeking behavior by those who will benefit from these construction projects (via contracts, jobs, etc.) and by landowners who may benefit from the provision of new forms of infrastructure. Such forces were key drivers of many recent U.S. transportation mega-projects, such as Denver International Airport and Boston’s Big Dig (Altshuler and Luberoff 2003). 233

234

David E. Luberoff

Matheson cites some provocative studies in support of the argument that cities that host mega-events often see subsequent increases in tourism and private investment. These findings should be viewed cautiously because the opportunity costs associated with such events may well be too high. In particular, in the long run the fortunes of cities and regions probably are more closely tied to the quality of their workforce, the effectiveness of their governments, the overall quality of life, and the extent to which their institutions respect and protect property rights (Glaeser 2011).

Why Do “Fools” Fall in Love (with Mega-Events)? Given the lack of strong positive impacts, why do countries keep falling over themselves to host mega-events? The answer is that we are asking the wrong question. Matheson focuses on the question of why national governments pursue the Olympics and the World Cup. But what if national bids for these events are actually the products of intensely localized activity by those who stand to benefit from hosting them? This would be the case if the cost of hosting the games is primarily borne by others, such as the nation’s taxpayers. This is not a new idea. Such bottom-up federalism has been a staple of national infrastructure policies for many years (see Altshuler and Luberoff 2003; Maass 1947). Is it possible to improve the local and national decision-making process for mega-events? The easy answer is that policy makers should use better analyses in making those decisions. But given the role of local rent-seeking behavior, this is not likely to occur. Even so, cities or regions that host mega-events should be required to provide at least half of the funding needed to build and operate the facilities required for those events. Moreover, local voters should have the opportunity to weigh in on those plans with some sort of referendum. This would allow those who might ultimately foot the bill to decide if the tangible and intangible benefits that come from hosting such events outweigh the potentially high cost. As George Peterson has shown, voters will support reasonably well-crafted packages and are likely to reject plans and projects that egregiously benefit a narrower group of interests (Peterson 1990). In the end, as the late John Quigley and his colleagues noted in a marvelous essay on the long debate and many referenda over a new baseball stadium for the San Francisco Giants, the median voter may decide to support projects that policy analysts judge to be unwise. But as Quigley and his colleagues also note, it is hard to argue with such decisions if they come after reasonable, informed discussions and debates about a carefully crafted proposal’s costs and benefits (Agostini, Quigley, and Smolensky 1997).

commentary

235

references Agostini, S. J., J. M. Quigley, and E. Smolensky. 1997. Stickball in San Francisco. In Sports, jobs, and taxes: The economic impact of sports teams and stadiums, ed. R. G. Noll and A. S. Zimbalist, 385–426. Washington, DC: Brookings Institution Press. Altshuler, A. A., and D. Luberoff. 2003. MegaProjects: The changing politics of urban public investment. Washington, DC: Brookings Institution Press and Cambridge, MA: Lincoln Institute of Land Policy. Centre on Housing Rights and Evictions. 2007. Fair play for housing rights: Megaevents, Olympic Games, and housing rights. Geneva: Centre on Housing Rights and Evictions. Danielson, M. N. 2001. Home team: Professional sports and the American metropolis. Princeton, NJ: Princeton University Press. Glaeser, E. L. 2011. Triumph of the city: How our greatest invention makes us richer, smarter, greener, healthier and happier. New York: Penguin Press. Gold, J. R., and M. M. Gold. 2008. Olympic cities: Regeneration, city rebranding and changing urban agendas. Geography Compass 1(2):300–318. Long, J. G. 2012. Public/private partnerships for major league sports facilities. New York: Routledge. Maass, A. 1947. Muddy waters: The Army Engineers and the nation’s rivers. Cambridge, MA: Harvard University Press. Peterson, G. E. 1990. Is infrastructure being undersupplied? In Is there a shortfall in public investment?, ed. A. Munnell, 113–130. Boston: Federal Reserve Bank of Boston. Zimbalist, A. S. 2003. May the best team win: Baseball economics and public policy. Washington, DC: Brookings Institution Press.

9 Involuntary Resettlement in Infrastructure Projects: A Development Perspective Robert Picciotto Change is an ordeal; its only cure is action. —eric hoffer There’s nothing easier, there’s nothing cheaper than taking care of the poor. —lula da silva

T

his chapter puts the spotlight on a dark corner of the development enterprise: involuntary resettlement. In the increasingly integrated world economy, infrastructure promotes the free flow of ideas, people, and goods that underlies economic convergence. In particular, public works (e.g., roads, highways, bridges, dams, canals, and urban renewal schemes) are essential for economic and social progress, but when implemented in vulnerable social contexts, they leave shattered lives and community upheaval in their wake. This chapter first examines the intersection between development and human displacement within a rapidly changing development policy context. It then probes the logic of involuntary resettlement, explores its context, reviews its antecedents, and delineates its scope and impact. Next, it sketches the emerging challenges created by current infrastructure provision and financing trends. The chapter concludes with policy recommendations.

Development Policy at a Crossroads Ample evidence suggests that forcible eviction to accommodate construction projects in the zones of turmoil and transition in the developing world disrupts 236

involuntary resettlement in infrastructure projects

237

livelihoods. In fact, this disruption is similar to that caused by violent conflicts or natural disasters. This is wrong and unnecessary. New policy priorities are needed; they should reflect the lessons of experience and take into account recent changes in the authorizing environment. The reduction of poverty has supplanted growth as the universally accepted aim of the development enterprise, and the human rights agenda is now at the center of the policy stage. Tolerance for incurring social and environmental risks has shrunk. The north-south model of international relations, which lumped together emerging middle-income economies with low-income and fragile countries, has become anachronistic. Developing countries are now in the driver’s seat of development programs. However, development-induced resettlement practices have yet to adapt to these new policy directions. Everywhere the basic assumptions that underlie development programs are being reconsidered. The growing interconnectedness of the global economy and the dominance of private investment in the development process remain uncontested, but standard “big picture” policy prescriptions like the Washington Consensus have been discredited. Within the academic world, the lack of robust research findings from a decade’s worth of cross country regressions has shifted the focus of development research from macroeconomics to microeconomics. In particular, experimental methods geared to figuring out what works in development have become fashionable, but they too are failing to come up with unambiguous results applicable across countries and continents (Cohen and Easterly 2009). Only one big idea appears to have survived: immune from experimental testing, it holds that infrastructure is a fundamental determinant of economic growth and productivity.1 But infrastructure investment remains controversial given its social and environmental consequences.

The Return of Infrastructure-Driven Development Evidence supports the proposition that economic growth is highly correlated with the stock of infrastructure assets, especially early in the development process (Yoshino and Nakahigashi 2000). Equally, income inequality has been shown to decline with an increase in infrastructure quantity and quality (Calderon and Serven 2004). The current development paradigm holds that human well-being is the ultimate development prize, that private enterprise is the engine of a developing economy, and that infrastructure provides its wheels. This notion amounts to a rediscovery of the capital-driven models that characterized development economics in its pioneering years. Policy studies have also disclosed that equitable access to basic infrastructure services is a crucial prerequisite for accelerated progress toward the Millennium Development Goals endorsed by all United

1. World Bank (1994b) provides a comprehensive treatment of the role of infrastructure in development.

238

Robert Picciotto

Nations members (Leipziger et al. 2003). Finally, there is no denying that environmentally sustainable development hinges on a much greater investment in clean and renewable energy sources. More than 1 billion people lack access to roads, safe water is out of reach for 1.2 billion people, 2.4 billion are not served by proper sanitation facilities, and 2.3 billion are not connected to reliable energy sources. Hence, infrastructure investment is now considered an essential instrument of global poverty reduction. Justin Lin, former chief economist for the World Bank, recently identified infrastructure investment as a “low hanging fruit” in the development process.2 Yet, with the exception of structural adjustment lending, no aid vehicle has given rise to more civil society protests or more divisive policy controversies than the large infrastructure projects funded by development assistance agencies. This deep public distrust is associated with a long history of mismanaged and misdirected resettlement initiatives.

The Checkered History of Resettlement Colonial history has shaped the history of resettlement. Forced displacement has systematically been used to control groups that harbor opposition to the state. The removal of Native American tribes to reservations in the United States was achieved through resettlement. The British government used resettlement to fight Kikuyu tribes resistant to colonial rule in Kenya. Relocation to protected enclaves was an essential part of the military strategy that defeated the communist insurgency in Malaya. The U.S. Army famously tried to emulate this approach through its Strategic Hamlets program in its vain attempts to pacify the countryside during the Vietnam War. Resettlement has also been used for social engineering. For example, colonial rulers and later the Indonesian government relocated poverty-stricken landless Javanese farmers to the outer islands of the Indonesia archipelago to achieve a more balanced population distribution and to promote a single national identity. Similarly, following independence and under one-party rule, Tanzania resettled rural populations in order to collectivize agriculture in line with President Julius Nyerere’s defunct model of African socialism (Ujamaa). These ill-fated social experiments were driven by nation-building programs that gave pride of place to forced modernization. Most policy makers in charge of developing countries’ economies viewed informal urban settlements, nomadic lifestyles, and scattered farming as obstacles to development. Under authoritarian 2. Elaborating on this proposition at the North South Institute Conference on the Future of Multilateral Development Cooperation in a Changing World Order (June 2011, Ottawa, Canada), he said, “A global infrastructure investment initiative would be a ‘win-win’ for the world. It would boost growth and reduce poverty in developing countries, rejuvenate advanced economies, increase the demand for their capital goods exports and create much needed jobs.”

involuntary resettlement in infrastructure projects

239

rule, and with strong support from the fledgling aid industry, the newly independent states used large-scale infrastructure projects to accelerate capital formation and industrialization. The demise of central planning models paralleled the growing influence of rights-based approaches in development and eventually led to a reconsideration of resettlement policies and practices. This chapter gives special attention to the World Bank not only because of its pioneering social research contributions to the resettlement debate, but also because of its early adoption of a range of social and environmental safeguard policies that put it in the vanguard of development policy reform.

The Peculiar Logic of Involuntary Resettlement For decades, the World Bank has exercised remarkable intellectual leadership in social and environmental policy based on its innovative and activist policy research agenda. Its operational manual policy statement 4.12 on involuntary resettlement stresses the dire consequences of unmitigated development-induced resettlement for affected people and communities. The policy statement acknowledges the reduced incomes, unemployment or underemployment, homelessness, dismantled production systems, environmental risks, disrupted social networks, and weakened cultural identity associated with involuntary resettlement (World Bank 2011). But the actual implementation record shows that in seeking to walk a fine line between the capital formation imperatives of economic growth and its disruptive consequences for vulnerable segments of society, the policy has glossed over major ethical and policy dilemmas. For example, the policy statement defines “involuntary” as actions taken without the displaced person’s informed consent or power of choice, while also stating that “resettlement” is made up of actions to ensure, among other things, that the displaced are not only informed about their options and rights but also meaningfully consulted on, offered choices among, and provided with alternatives. It follows that involuntary resettlement, thus defined, embodies a fundamental contradiction: if, as posited by economist Amartya Sen, development is the freedom to realize human capabilities, how can involuntary displacement practices that devastate communities and destroy livelihoods be made compatible with the mandate of a development assistance agency? Is development-induced relocation ever voluntary under current practices? The participatory process mandated by the policy implies that a genuine choice is available to individuals and groups facing relocation, whereas in reality principled negotiations about relocation are rarely carried out on a level playing field. More often than not, affected individuals and groups are deprived of access to full and reliable information about the infrastructure projects that will affect their livelihoods and lifestyles. Refusing to relocate is not usually a realistic option. Given power imbalances, relocation is often achieved through misleading communications, threats of persecution, intimidation, or bribery. When such tactics

240

Robert Picciotto

fail, coercion is the default option. To be sure, potentially displaced people may be able to pressure the authorities and prevent construction through public protests and political influence, but the implicit threat of force is always present. Violence is structural more often than explicit; refusal to relocate may mean being flooded out by rising waters or witnessing the destruction of one’s home by a bulldozer. The choices available to affected individuals are severely restricted and akin to offers that cannot be refused. Without legal recourse, relocation is inevitable, and without independent verification one cannot determine whether relocation has been voluntary, induced, or coerced. Similarly, the word resettlement is not neutral. Given its technocratic flavor, it evokes assisted migration, planned relocation, and harmonious reintegration in the community. It also connotes mobility, forward thinking, and social adaptation. But in the real world, resettlement is subject to bureaucratic rules and manipulation, and powerful interests and political realities shape all resettlement transactions. Thus, the World Bank policy does not specify when and how the resettlement process ends. Does it end when livelihoods have been fully restored and communities have been revitalized? Or is the mere prospect of achieving such a state the relevant standard? Who decides, and on what basis? Almost invariably, the criteria that mark the beginning and end of the resettlement period are set arbitrarily and unilaterally by the authorities (Muggah 2008). To be sure, the World Bank policy statement makes clear that involuntary resettlement should be avoided wherever possible or that it should be minimized, but the policy does not linger on this topic or elaborate on how this might be done. Instead, it instructs World Bank staff in extraordinary detail about the measures, instruments, and practices that they should put in place to make way for the projects it funds. The subtext is that involuntariness is the rule rather than the exception.

A Disciplinary Divide Disciplinary fault lines characterize the resettlement policy debate. Economists point to the great benefits to the society at large from infrastructure investment. They argue that opposition by civil society critics and activist scholars is detrimental to the public interest. While not blind to the downside social risks, economists tend to view them as undesirable side effects to be addressed on a best effort basis rather than as fundamental policy concerns that should inform project choices and engineering designs. In response, sociologists stubbornly insist that displaced individuals and groups bear a disproportionate share of the costs. They point to eight interrelated consequences of poorly managed human displacement (Cernea 1997): 1. Landlessness (linked to land expropriation). 2. Joblessness (connected to loss of wage employment). 3. Homelessness (loss of shelter, disrupted communities).

involuntary resettlement in infrastructure projects

241

4. Marginalization (human capital loss, downward social mobility). 5. Food insecurity (associated with loss of land). 6. Increased morbidity and mortality (increased reliance on unsafe water sources, increased exposure to disease, psychological stress). 7. Reduced access to social services and common property (schools, health centers, pastures, forest lands, burial grounds). 8. Social disarticulation (unraveling of social ties, loss of cultural capital, etc.).3 On the one hand, methodological blinders have induced myopia among economists about the consequences of infrastructure investment on third parties. Naive cost-benefit analyses of project investments have failed to capture multifaceted impacts, such as the psychological trauma inflicted on displaced individuals, the decline in their social status, the irreversible costs to the environment, and the destruction of social capital. On the other hand, sociologists and anthropologists have relied on one-sided surveys and heartrending narratives about the plight of displaced individuals while ignoring the enormous social benefits associated with infrastructure services like transportation, electricity, housing, water, sanitation, and flood control. A rapprochement is overdue in order to capture the full effects of infrastructure projects and learn from them. Typically, infrastructure investments yield widely dispersed direct benefits. The positive externalities include economically advantageous spillover effects (forward linkages like industrial investment induced by a new road and backward linkages like rising demand for capital goods). The same projects are frequently saddled with negative and cumulative externalities, mostly borne by those unlucky enough to live on land required for development.4 Resettlement policy regimes that came about as a result of global advocacy campaigns and local protests have had the stated purpose of internalizing and alleviating these effects. But their efforts have failed, and this is why the controversy still rages.

The Emergence of Safeguard Policies in Development The World Bank’s first policy on involuntary resettlement was approved in 1980. It was revised several times,5 and the organization issued a handbook that went

3. Other authors mention abuse of human rights and violence from security forces and communal conflict in resettlement areas. 4. The entire society and future generations may also be affected where infrastructure development results in cultural or environmental damage. 5. The original policy statement was substantially reworked in 2001 following five years of debate. It was revised in August 2004 to ensure consistency with the development lending policy.

242

Robert Picciotto

through several editions. The handbook promulgated a host of rules and regulations for project identification, preparation, appraisal, approval, and supervision (World Bank 2004). But whereas the World Bank’s environmental safeguards had been backed by an environmental assessment policy, the social safeguards (the indigenous peoples and involuntary resettlement policy statements) were not embedded within an overarching policy framework. As it stands, the involuntary resettlement policy includes measures to ensure that the displaced are informed of, consulted on, and offered choices among resettlement alternatives. The affected populations are meant to receive prompt and effective compensation at full replacement cost for asset losses attributable directly to the project. They are entitled to assistance during relocation and to residential housing, housing sites, or agricultural sites that are at least equivalent to what they have lost in terms of productive potential, location, and other factors. The goal is to provide the displaced with the support they need during the transition period to restore their livelihoods and standards of living, including assistance for preparing the land, obtaining credit, and training and accessing job opportunities. In designing its social and environmental safeguard policies, the World Bank consulted with governments, private sector entities, and civil society actors with a view to striking a judicious balance among conflicting interests. This process was intended to consolidate the World Bank’s development leadership and to strengthen its public legitimacy. But the process became extraordinarily costly and cumbersome; it took five years to enact a revised policy, and neither the World Bank critics nor the World Bank borrowers were satisfied with the outcome. Similarly, the setup of a fiercely independent Inspection Panel in 1993 did little to enhance the World Bank’s reputation since the Panel systematically ruled against the World Bank in response to most of the complaints it was called on to adjudicate.

Global Prescriptions Versus Local Realities The detailed rule-based approach of the World Bank’s involuntary resettlement policy reflects the project-based technocratic principles characteristic of the organization in its early years. Throughout the 1950s, 1960s, and 1970s, project lending was the instrument of choice: projects proved to be flexible vehicles of development assistance and pragmatic tools of policy influence. Consequently, the involuntary resettlement policy was framed to connect with the elaborate processes of the World Bank project cycle. This approach was intended to give teeth to the policy since projects are backed up by loan or credit agreements

It was revised again in March 2007 to take into account new policies governing World Bank responses to crises and emergencies. The most recent update in February 2011 clarified how escrow accounts can help reduce implementation delays and fund grievance mechanisms.

involuntary resettlement in infrastructure projects

243

deemed to have the force of international law and to supersede the strictures of national legislations. But in the real world, loan and credit agreements have limited enforcement value since the only practical remedy available to the World Bank is the cancellation of undisbursed loan or credit amounts.6 For resettlement—as for many other policy objectives promoted by the World Bank—conflicts between country systems and standard policy guidelines have proved to be the weak link of the accountability chain. World Bank officials must exercise the “art of the possible.” The developing countries’ civil servants, entrusted with the responsibility of administering World Bank projects, cannot realistically be expected to ignore or flagrantly circumvent the rules and edicts of their own government. Nor can they single-handedly overcome the enormous handicaps created by large regulatory gaps or deeply rooted governance weaknesses. Unsurprisingly, underperformance against the ambitious stretch targets of World Bank policies has been the pattern in the infrastructure sector. The World Bank’s unrealistic goal of 100 percent compliance and its vision of resettlement as a “development opportunity”7 could not mask a chronic inability to overcome the myriad domestic obstacles that stand in the way of realizing its policy intentions. Its mediocre implementation record and the unusual transparency of its processes fueled advocacy campaigns concentrated on the World Bank rather than on developing countries. The iconic development policy status achieved by the World Bank, combined with its patent failure to implement in full the “first do no harm” intent of its safeguard policies, made it a lightning rod for public protests.8 The pursuit of the World Bank global mandate had proved inconsistent with domestic political and institutional realities. This is not to say that the World Bank’s safeguard policies made no difference. The policies did set the stage for more humane infrastructure development standards, and they helped to mitigate a great deal of human hardship. But they did not fully succeed in preventing impoverishment and misery for hundreds of thousands of forcibly relocated individuals.

6. Theoretically, the World Bank has the legal authority to recall the full loan or credit (i.e., secure the full and immediate reimbursement of disbursed funds), but this right has never been exercised. 7. Ian Johnson, vice president for environmentally and socially sustainable development at the World Bank, notes, “Resettlement can have serious repercussions that cannot be exclusively measured in economic terms. . . . Well-designed and well-implemented resettlement can, however, turn involuntary resettlement into a development opportunity” (World Bank 2004). 8. The Sardar Sarovar (Narmada River) project in India has long been emblematic of the World Bank’s inability to implement its own resettlement policies, as documented by an independent review commissioned by the World Bank’s president (the Morse Commission). The damaging review findings fed into a global advocacy campaign and induced cancellation of the World Bank loan approved in 1985. It also facilitated the establishment of the independent World Bank Inspection Panel.

244

Robert Picciotto

The Rights-Based Critique International human rights law establishes norms and principles touching on virtually all facets of life. The United Nations General Assembly adopted the Declaration on the Right to Development on 4 December 1986. The declaration proclaimed the inalienable right of everyone to participate in, contribute to, and enjoy economic, social, cultural, and political development. From this perspective, forced evictions came to be viewed as a gross violation of human rights, in particular the right to adequate housing. For civil society activists, the World Bank, as a specialized agency of the United Nations, is charged with the protection and promotion of human rights. Hence, it must ensure that everyone affected by the projects it finances not only receive assistance that will help them restore their incomes and secure ready access to adequate land and other resources, but also derive incremental benefits from the projects, including special provisions for ethnic minorities, indigenous populations, and women disproportionately affected by the projects. Civil society activists highlight four defects in the World Bank relocation policy. First, they argue that the current policy focuses on mitigating the direct displacement impact of projects, leaving governments with the task of addressing indirect impact. They stress that the World Bank should be equally accountable for indirect effects, such as the downstream effects of a dam that affects the livelihoods of fishermen, disrupts ecosystems, and/or leads to impoverishment or social dislocation. Second, social activists deplore that the policy aims to “improve or at least restore” pre-project standards of living. Strictly interpreted, this language limits the World Bank’s obligation to aim at a baseline requirement of income restoration. Depending on the time required to achieve this goal (and since the displaced would have had opportunities to enhance their incomes in the absence of the project), the provision could in effect imply significant impoverishment of the affected communities.9 Third, whereas the policy gives preference for land-based resettlement strategies (thus allowing cash compensation), critics insist that land-for-land compensation should be mandatory for displaced farmers given the cultural significance of land, the low economic and social status accorded to landless laborers and sharecroppers, the insuperable difficulties of securing adequate land where it is at a premium, and the difficulty of retraining farmers for nonagricultural occupations.

9. This is why a review of involuntary resettlement carried out by the World Bank’s independent evaluation unit recommended that “the emphasis should shift from restoring income levels which suggest stagnation at pre-project lifestyles, to improving income levels which brings the displaced into the development process along with the project’s primary beneficiaries” (Operations Evaluation Department 1998).

involuntary resettlement in infrastructure projects

245

Fourth, critics reject the World Bank’s reliance on market valuation of replacement costs as the appropriate compensation standard for lost assets. Market valuation does not take into account nonmarket losses, including the loss of status, cultural deprivation, ruptured social networks, and the resulting insecurity associated with resettlement (Clark 2000).

The Eminent Domain Precedent The governments of developing countries champion rights-based approaches to development but only as admirable aspirations rather than as practical guides to decision making. In particular, they resist the pious exhortations of civil society activists and point out that the World Bank’s involuntary resettlement policies demand more of them than what industrialized country governments would tolerate on their own territory. Specifically, they see no reason why they should be denied the long-standing prerogative of a sovereign state to expropriate land under the eminent domain doctrine. The idea that the government has the power of eminent domain goes back to the Magna Carta of 1215. It states that the government can dispossess its subjects of land and property but only according to the law of the land. Similarly, Hugo Grotius, a Dutch jurist, opined in 1625 that the property of subjects is under the eminent domain of the state so that the state may use and even alienate or destroy such property in the public interest. In such instances, however, the state is bound to make good the loss to those who lose their property (Nowak and Rotunda 2004). To be sure, the eminent domain doctrine is not a universally accepted legal principle.10 Japan, for example, does not recognize eminent domain. But the legal regimes of most countries are in line with the eminent domain principles, especially regarding compensation. In Australia the Land Acquisition Act grants the state unlimited powers for any purpose whatever. In the United States according to the Fifth Amendment, land taking by a public or private body is entirely acceptable when the acquisition is for public use; the only proviso is that just compensation must be paid. The U.S. Supreme Court has deferred to the right of states to define and interpret “public use,” but it has had to adjudicate several appeals. Its rulings have confirmed that public use only means public benefit, rather than public occupation, and that just compensation is no more than the fair market value of the land. Similarly, based on historical precedent and in line with its Convention on Human Rights, the European Union protects private property and prohibits

10. Since land cannot be produced by human labor, the notion of land as private property (rather than common property) is not universally accepted. In this vein, Henry George advocated taxing away all land value gains resulting from economic development (George 1953).

246

Robert Picciotto

state interference unless it is necessary in the interests of national security, public safety, economic well-being of the country, prevention of disorder or crime, protection of health or morals, and so forth. Thus, deprivation of private property is allowed under European law if it is in the general or public interest (including payment of taxes), but just compensation must be paid where interference takes place.11

The Free-Market Alternative Just as land takings in developing countries have elicited protests, eminent domain actions in developed countries have triggered court challenges. In the United States new local statutes have been introduced to restrict intrusive government intervention in land use planning and regulation. Opposition groups dismiss the “market failure” justification for a dominant state role in land policy. Instead, they deplore the government failures associated with unchecked regulatory powers. In fact, many jurists, scholars, and libertarians have taken the view that eminent domain is despotic. The arguments they have put forward coincide with those advanced by social activists in the development sphere: (1) market value is only the value that the marginal owner attaches to the property and not the personal and cultural value that the actual owner attaches to the confiscated land; (2) the assembly value generated by an infrastructure project belongs to the original occupants rather than the state or the private developers since it is inherent in the land itself; and (3) the disproportionate gains that urban renewal and other investment projects generate for the state and for the well-to-do at the expense of the poor and powerless are unethical. In fact, value capture has generated a vast literature and considerable debate in policy circles, as illustrated by the deliberations of the Lincoln Institute’s sixth land policy conference. Fundamentally, proponents of free-market solutions believe that liberty is rooted in the right to private property and that state efforts to correct land market failures (whether in support of public or private development projects) usually make matters worse. They reject the notion that only government enjoys the coordination powers needed to assemble land for use in the public interest. They favor voluntary exchange as the right approach since, they believe, a recalcitrant owner will eventually yield to a profitable offer—if the price is right (Benson 2010). There is certainly room for market-based solutions in resettlement activities. Such solutions are used as a matter of course by developers and private corpo-

11. In England and Wales, under a patchwork of statutes and case law, compulsory purchase requires compensation at a price approved by or stipulated by a court. In Germany expropriation for the public good is allowed, but just compensation must be paid, and dispossessed owners may ask a court to review the compensation offered.

involuntary resettlement in infrastructure projects

247

rations to purchase contiguous land parcels for industrial or commercial use. The use of dummy buyers helps them achieve land assembly at reasonable cost; option contracts facilitate the workings of land markets, and the allocation of fractional and tradable land development rights helps distribute project benefits more equitably. But such solutions presuppose that property rights are protected, that the individuals subject to displacement have legal title to the land they occupy, and that they can expect redress from the courts if they are subjected to unlawful treatment. This is not the situation in many developing countries where land records are scarce, titling is not widespread, and judicial processes are weak or corrupt. In such situations, governance reform and capacity building have priority, and in the interim the civil society may be induced to fill the institutional gap. For example, if voluntary organizations are satisfied with the fairness of proposed relocation and compensation arrangements, they may choose to partner with the state to ensure that free and informed consent is secured, that participatory methods are put to work, and that peer group pressure is applied to win over recalcitrant holdouts. Of course, such processes imply prior identification of legitimate claimants, compensation packages set at attractive levels, effective relocation support, and employment assistance. They ultimately aim to make resettlement as voluntary as possible.12

How Many People Are Displaced? How significant is the resettlement issue? The scarcity of data is a major cause of uncertainty regarding the social outcomes of infrastructure projects. But although no reliable count is available, it is possible to venture informed guesses that point to 11 to 23 million people affected annually. A study of World Bank–assisted development projects (World Bank 1994a) estimated that displacement due to World Bank projects between 1986 and 1993 accounted for 3 percent of the total US$ cost for dam construction and 1 percent for transportation and urban development. Extrapolating from these data, the review concluded that 10 million people were displaced annually by development projects worldwide. The construction of high dams accounted for 4 million; urban, water, and transport infrastructure projects accounted for another 6 million. The same study showed that the impact was heavily concentrated on the rural and urban poor, indigenous communities, and ethnic minorities. For example, Guatemala’s Chixoy Dam project involved the forced relocation of 2,500 Maya Achi indigenous people and the massacre of 369 tribal people by local armed

12. In Romania the Rosia Montana gold-mining project initially involved forced displacement, but the process became largely voluntary based on a “willing buyer, willing seller” principle induced by local protests (Maldonado 2010).

248

Robert Picciotto

groups and the army (Witness for Peace 1996). In Mexico the Miguel Alemán Dam displaced and impoverished at least 20,000 Mazatec people (Barabas and Bartolome 1973). Tribal peoples are estimated to make up 40 to 50 percent of development-related involuntary displacement in India (Colchester 2000). A more recent inventory of World Bank projects (1999) identified 223 active projects that displaced 2.6 million people (Clark 2000). Given an average project life of seven years, this is equivalent to annual displacements of 371,000 people for average annual investments estimated at $7.1 billion (or 52,000 people displaced per $1 billion). But a significantly lower estimate emerges from the latest review of World Bank safeguard policies (Independent Evaluation Group 2010). It estimates that World Bank projects involved displacement of about 166,500 people annually for average annual commitments of $6.9 billion (24,000 people displaced per $1 billion). This lower estimate (based on projects approved between 1999 and 2008) could reflect improved project selection, but it might also result from a lower percentage of large dams in the World Bank portfolio. By the mid-1990s, the World Bank was financing about four dam projects a year—half the number it had funded in the 1970s and 1980s.13 A similar slowdown in the rate of large dam construction also appears to have taken place globally: the number of large dams completed declined from 1,000 a year from the 1950s to the mid-1970s to around 260 a year during the 1990s. Considering that infrastructure investment in developing countries stood at $450 billion in 2008, the earlier Independent Evaluation Group (IEG) estimate (in line with ICOLD’s hypothesis of fewer but larger dams increasingly concentrated in developing countries)14 yields about 23 million people displaced annually by development projects. In contrast, the most recent IEG estimate points to less than half of this number—about 10.9 million development-related displaced people a year. The range of these estimates is consistent with Cernea’s estimate of more than 15 million development-induced displaced people a year (M. Cernea, pers. comm.; Oliver-Smith 2009). Though impressive, these numbers are limited to those who have been physically displaced by infrastructure projects. They exclude those who were indirectly affected by the project (e.g., fishermen downstream of a dam) or by the resettlement process (e.g., residents experiencing increased 13. According to John Briscoe, former World Bank water adviser and Harvard professor: “In the last 15 years the World Bank has financed only two major dam projects: one of them was Bujagali in Uganda, which took over a decade to win approval; the other was the Nam Theun 2 plant in Laos, which underwent 14 reviews by independent panels before finally being approved” (Delli Priscoli 2011). 14. The International Committee on Large Dams (ICOLD) notes that the dams currently under construction are much larger than previous dams and that the bulk of large dam investment is taking place in developing countries (www.hydrocoop.org/publications/Role_of_Dams _new.pdf).

involuntary resettlement in infrastructure projects

249

population pressure in resettled areas), and they do not take into account voluntary resettlement. Unsurprisingly, India and China are responsible for a significant share of development-related dispossessions. Together, they account for 130 million displaced people between 1950 and 2005—that is, 2.3 million displaced people annually15 (Cernea 2006). In relative terms, the impact of infrastructure development tends to be larger in smaller countries. For example, the Akosombo Dam flooded 3.5 percent of Ghana’s territory and displaced 1 percent of the population, compared to 0.01 percent of India’s territory flooded and 0.013 percent of its population displaced as a consequence of the notorious and very large Sardar Sarovar (Narmada River) project (Cernea 1997).

Impacts Vary The number of people affected by resettlement is growing in line with the growth of infrastructure investment in developing countries. Together, public-private partnerships, official development assistance, and World Bank lending for infrastructure reached $186 billion in 2008, up from $38 billion in 1990. But actual resettlement footprints vary with the type of infrastructure. Beyond their impact on directly affected communities, large dams have dominated the public debate because of their irreversible ecosystem effects and their symbolic embodiment of a particular development model.16 There is little doubt that the physical, social, and cultural alterations associated with a large dam are deep and lasting. While large dams generate clean power, irrigation, flood control, and navigation benefits shared by millions, the inundation caused by their reservoirs can have catastrophic and large-scale effects on local communities and individuals. It deprives them of access to shelter, land, and common property resources. It disrupts their ability to work and trade. And it upsets their social and cultural links. A desk review of 50 World Bank–funded large dams (Operations Evaluation Department 1996) found that they displaced 290,000 people and that threefourths of the projects fell short of the World Bank’s safeguards. This finding is not surprising since most of the dams had been approved prior to the World Bank’s resettlement guidelines. But the review also found that only one-fourth of the projects would have failed to reach the 10 percent economic return threshold had enough been spent on them to meet World Bank guidelines. This confirmed

15. The World Commission on Dams estimates 26 to 58 million displaced people for both countries between 1950 and 1990. 16. In her comment appended to the World Commission on Dams report, one of the commissioners (Medha Patkar, Struggle to Save the Narmada River) asserted that “the problems of dams are a symptom of the larger failure of the unjust and destructive dominant development model.”

250

Robert Picciotto

that following the ethical course in resettlement need not mean failing economic tests. But the challenges involved in doing the right thing should not be underestimated. A later desk review (Scudder 2005) of large dams that together displaced 1.5 million people found that incomes were improved or restored in only 16 percent of the 50 cases. Landlessness was an issue in 86 percent of the cases; joblessness in 80 percent; food insecurity in 79 percent; and marginalization and reduced access to common property resources in 77 percent. This disappointing outcome was traced to one or more of the following factors: lack of implementation capacity, lack of finance, lack of political will, lack of opportunities available for resettling households, and lack of participation in decision making. Infrastructure-related displacement caused by urban, water supply, and transportation projects—while less visible and more dispersed than the displacement attributed to dams—might well be larger in the aggregate when the sharply reduced pace of dam construction is taken into account. Such investments are growing rapidly as a result of burgeoning industrialization and spreading urbanization in emerging market countries. The extent of resettlement per unit of expropriated urban land is higher than in rural settings due to higher population densities. Forced eviction of poor and disadvantaged people has been on the rise in the developing world as part of ill-conceived slum eradication and urban renewal programs. One billion people worldwide reside or squat in slums, and their number is expected to double by 2030. Most lack access to electricity, clean water, sanitation, and other basic services. Distorted perceptions of slums as lawless, diseased, and dangerous areas have contributed to misguided urban policies that seek to eradicate slums rather than upgrade and rehabilitate them. Under authoritarian regimes (but far less so where democracy is taking hold, such as in Latin America), the practice of forced evictions in urban settings seems to have been accelerated by the fully justified priority that the Millennium Development Goals have placed on improving the dire living conditions of slum dwellers. Slum improvement and land titling programs are far better alternatives than forced evictions. Most slum inhabitants have no option but to live where they do since they rely for subsistence on informal urban employment close to their abode. Needless to say, slum dwellers rarely receive employment and relocation assistance when they are forcibly evicted, forcing them to resettle in other slums. Thus, despite its implementation problems, involuntary resettlement is a far better policy option than forced eviction, especially when it is unaccompanied by effective relocation support. Of course, voluntary resettlement is the superior option. It should be based on the free, prior, and informed consent of those affected and backed up by independent verification mechanisms. Voluntary resettlement requires accurate baseline social surveys and calls for compensation that is higher than current market values. Quite apart from the social value of this rights-based approach,

involuntary resettlement in infrastructure projects

251

the incremental costs may be compensated by the benefits of timelier project implementation. Following the liberalization of land markets in developing countries, evictions need not be forced since they are market driven. But in reality, the “creative destruction” associated with the dynamics of liberalized land markets is eerily reminiscent of the effects of unmitigated displacement associated with infrastructure projects. More profitable land use driven by urban growth induces rising property rents. When poor tenants cannot afford these higher rents, they opt to move out “voluntarily.” Thus, rising property values trigger large-scale market displacement from informal settlements. Permits to occupy (common in Africa and Asia) do not confer security of tenure because they have a short shelf life and can be abrogated by administrative fiat. Market-based evictions often take place without compensation since the occupants do not normally hold a property title or a lease. Should the occupant refuse to move, a court order may be secured. Alternatively, negotiations may take place, but the outcome depends on the extent to which occupants enjoy quasi-ownership rights, social status, or political protection. Compensation is often unfair and far less than even the property or rental value would justify, and market-driven evictions do not entitle the affected households to resettlement assistance or in situ upgrading. The social impact of urban renewal varies considerably from country to country depending on the regulatory context. Where land is the exclusive property of the state, right of use may be recognized, but security of tenure remains precarious. For example, in Cambodia the new government that took over from the Khmer Rouge in 1979 granted right of use to urban dwellers provided they registered with the authorities. Although the state retains ownership of the land, an informal property market has developed. In Rwanda the state is entitled to recover the land if leaseholders do not meet official construction standards within five years. In both countries, evictions are frequent, access to land is restricted, and new informal settlements are spreading at the periphery (Durand-Lasserve 2007).

Lessons of Experience Not all resettlement programs fail, but when they do fail, they each fail differently. The overall record is not one of systematic and unmitigated disaster, as pictured by critics. However, the aggregate performance has left a great deal to be desired for various reasons, most of which point to the inadequacy of compensation arrangements and the comparative disadvantage of public sector involvement on the supply side of resettlement (site selection, housing, employment, etc.). In India the resettlement record was abysmal for the Dhom and Kanher Dams in Maharashtra and the Narayanpur project in Karnataka. In Brazil resettler dissatisfaction has also been high for the Itaparica Dam. Even though compensation

252

Robert Picciotto

and relocation were handled properly, the irrigation projects that were supposed to provide alternative farm employment for displaced farmers failed. Physical relocation worked well for the Nangbeto Dam in Togo, but here too income restoration was not achieved due to lack of advance planning and a depressed regional economy. In contrast, the farmers displaced by the Shuikou and Yantan Dams in China were satisfied with the relocation assistance they received. Reservoir levels were set to minimize human displacement, and household incomes were restored. In Thailand and Indonesia, dynamic regional economies minimized the adverse economic and social impact of the Pak Mun and Kedung Ombo Dams despite settlers’ resistance (Picciotto, van Wicklin, and Rice 2001). ALTERNATIVES ARE NOT SYSTEMATICALLY EXAMINED

Avoiding resettlement altogether (or reducing its scope by examining alternative options) is rarely considered. The delayed examination of social aspects leads to the neglect of development opportunities, such as those associated with catchment management, natural reserves, reservoir fisheries, aquaculture, and tourism, in the case of large dams. Most of all, resettlement programs are developed too late in the project cycle. In short, human displacement is treated as an externality to be considered as part of project implementation rather than earlier as a project design consideration and a high priority in its own right. FREQUENT MISMATCH BETWEEN INSTITUTIONAL COMPETENCE AND RESPONSIBILITY

Public sector agencies responsible for resettlement are often limited in their capacity and their commitment to handle a task widely perceived to be unglamorous and unrewarding. Typically, project implementation agencies are equipped to manage civil works and to address engineering problems rather than to design and implement social programs. Accordingly, resettlement units are frequently understaffed or consist of neophytes with no field experience or training in social assessment, participatory methods, or poverty analysis. More often than not, resettlement units lack the leverage needed to obtain assistance from large and powerful agencies adept at the delivery of social services. POLITICAL WILL IS OFTEN ABSENT

Beyond adequate skills and budgets, strong and principled leadership is essential for resettlement to succeed. Along with limited agency capacities and weak commitment, a lack of political will emerges as a dominant cause of resettlement policy failure. Commitment may be strong in the field but weak at the commanding heights of the bureaucracy—or solid at high government levels but flagging at the ground level. It may also be absent at all levels. THE CIVIL SOCIETY IS NOT SUFFICIENTLY INVOLVED

Office-bound planning at the top rather than decentralized and flexible participatory approaches on the ground tend to dominate. Yet resettlement cannot easily

involuntary resettlement in infrastructure projects

253

succeed without the involvement of self-help organizations, community groups, and local governments. Where authority is not devolved to the lowest efficient level, resettlement programs are bound to suffer. Large bureaucracies seek to satisfy all constituents and do not tolerate special treatment of a particular group for the extended period of time often needed to restore the incomes of the displaced. LAND-FOR-LAND IS NOT ALWAYS THE ANSWER

Setting compensation levels at the right replacement level is tricky since land prices are not static and land speculation is fueled by the competition for land caused by infrastructure projects. Some of the most disappointing outcomes can be explained by the pursuit of land-for-land strategies under inauspicious circumstances and the lack of flexibility needed to resort to enhanced cash options combined with skills upgrading or employment assistance. THE PSEUDO-SETTLER PROBLEM IS REAL

In poverty-stricken areas, substitution effects may undermine the best-laid plans. A flood of pseudo-settlers claiming benefits for which they are not entitled may add to spiraling claims on scarce resettlement budgets. Bogus fishermen pleading for handouts, landless latecomers keen to secure a plot, and inhabitants of nearby slums seeking compensation in an urban renewal scheme may overwhelm hardpressed compensation mechanisms. This partly explains why the treatment of squatters is such a contentious issue. Under the World Bank resettlement policy, claimants who are considered illegal occupants under the national legislation (i.e., untitled people without established customary rights) must be compensated. This can make the policy unworkable. For example, a local administration that wishes to implement a slum clearance and rehabilitation scheme in a high-density urban area may be unable to control a flux of rent-seeking illegal occupants demanding compensation for involuntary resettlement and rehabilitation. INCOME RESTORATION (LET ALONE IMPROVEMENT) IS HARD TO ACHIEVE

The restoration or improvement of the income of the displaced is an exceptionally demanding challenge unless the regional economy is booming. It takes time and effort for displaced individuals to readjust to a new environment. Except in countries like China with vast experience in the management of a command economy, public sector agencies are usually ill equipped to identify income-generating opportunities suited to resettlers’ aspirations and capacities. Public-private partnerships and civil society organizations are far better equipped to handle the task. BUDGETS DO MATTER

Inadequate funding for resettlement is a frequent problem. Successful resettlement is highly correlated with ample budgets. Conversely, the diversion of resettlement funding by corrupt officials is a recipe for failure. Systematic underestimates of

254

Robert Picciotto

displaced persons, optimistic projections of voluntary resettlement, and a lack of contingency provisions to address unexpected resettlement obstacles are common causes of budget inadequacy. Adequate funding requires political will and ingenuity. Freestanding development projects upstream of infrastructure investments may be combined with a variety of benefit-sharing methods that compensate the displaced, such as the automatic transfer of revenues from hydropower sales, equity sharing with indigenous communities, taxation for redistribution to the dispossessed, the granting of land leases to affected communities, and safety nets for especially vulnerable displaced individuals (Oliver-Smith 2009).

An Evolving Enabling Environment Ironically, the same large infrastructure projects that once symbolized central planning and state-led development strategies have become emblematic of the neoliberal development models opposed by radical environmentalists and antiglobalization activists. The same nongovernmental organizations, think tanks, and private foundations that had deployed extraordinary advocacy efforts to induce the World Bank to design and adopt safeguard policies remain the most vocal in their public criticisms. They keep raising the bar regarding the scope, intent, and intrusiveness of the World Bank’s safeguard policies, and they are pitiless in highlighting the gaps between policy goals and their realization on the ground. The enabling environment of the development enterprise has been transformed, adding to the obstacles standing in the way of achieving adequate compliance with safeguard policies. In particular, the commitment to project-level conditionality began to flag after the World Bank underwent two successive reorganizations. The 1987 reorganization sought to turn the World Bank into a policy-driven rather than project-based bank. The 1996 reorganization sought to make it a knowledge-based bank. Suddenly, the prior notion of projects as privileged particles of development was no longer in vogue, and the legitimacy of top-down, process-based blueprint approaches to social transformation processes drawn up in Washington had become deeply problematic. The stakes went up again at the turn of the century, when new global development imperatives were unveiled: the Millennium Development Goals of 2002 and the Paris Declaration of 2005 brought to the fore results orientation, country ownership, alignment with country systems, and reduced transaction costs. These new policy directions have contributed to a new aid culture in which developing countries have become immune to funding carrots, conditionality sticks, and policy sermons. The advent of south-south development cooperation has also helped put the mutual accountability principle to work.

Recent Policy Changes In this new authorizing environment, the World Bank established a pilot program to encourage the use of country legal and regulatory systems for the implementa-

involuntary resettlement in infrastructure projects

255

tion of safeguard policies. In 2005 seven projects in six countries were included in the first phase. In 2009 an additional eight pilot programs were introduced in seven more countries. Initially, country interest was high, but it quickly waned once it became clear that the World Bank was not ready to integrate its multiple safeguard policies under a single umbrella, that it would not relax its standards to accommodate borrowers’ systems, and that it would not relinquish its rigid approach and process-ridden controls. Bureaucratic gridlock has evidently prevailed, and the World Bank is no longer a trailblazer in social policy innovation. Within the World Bank Group, the torch has been passed to the International Financial Corporation (IFC) and the Multilateral Investment Guarantee Agency (MIGA). They have inaugurated a distinctive policy regime that integrates all the safeguard policies under a social and environmental sustainability umbrella focused on performance standards implemented by their private sector clients and monitored by IFC/ MIGA. These IFC/MIGA principles cover labor issues. They have informed the Equator Principles adopted by private investors. Both sets of principles are resultsoriented and provide a sound management framework for determining, assessing, and mitigating environmental and social risk associated with infrastructure project finance. They have been endorsed voluntarily by financial institutions for projects with capital costs that exceed US$10 million. Currently, 76 financial institutions in 28 countries have officially adopted the guidelines. They cover more than 70 percent of international project finance debt in emerging markets. Despite the lack of independent verification and the disquiet expressed by nongovernmental organizations about the weakened accountability that this approach might have generated, no evidence has emerged that the standards are being systematically diluted (Independent Evaluation Group 2010). Public-private infrastructure partnerships are multiplying, and they now dwarf those that are aid financed (Ingram, Liu, and Brandt 2013). Outside the World Bank Group, the European Bank for Reconstruction and Development has adopted the same approach as IFC/MIGA, although it has assumed more responsibility for ensuring implementation and has added transparency provisions and performance requirements for financial intermediaries. The Asian Development Bank has also adopted a single safeguard policy statement. A single sustainability policy framework provides more comprehensive and coherent treatment of environmental and social risks. While the Inter-American Development Bank and the African Development Bank have yet to join the movement toward a single policy framework, the writing on the wall is clear: the current World Bank policy regime is too fragmented, inflexible, and process-oriented. It privileges environmental concerns over social concerns. By stressing the “first do no harm” agenda and channeling enormous staff resources toward enforcement of rigid and detailed conditions at the project level, it discourages policy dialogue, analytical work, and capacity building geared to broader social issues.

256

Robert Picciotto

Other Human Displacement Challenges Forced relocation is not the exclusive domain of development projects. Big sporting events hosted by developing countries to boost their international image have generated large-scale construction projects built with little regard to their longterm social and environmental repercussions. Natural disasters, war, and civil unrest are major causes of forced relocation. Globally, natural disasters led to the displacement of 36 million in 2008, and 20 million of those were displaced as a result of sudden climate-related events.17 In the same year, the total number of refugees and displaced individuals as a result of conflict was 42 million18 (United Nations 2009). If we add those forced from their homes by infrastructureinduced resettlement and by extractive industries, the total number of displaced people probably reaches 100 million. Addressing the involuntary resettlement problem has become a security imperative. In India infrastructure and resource extraction projects threaten widespread displacement of tribal communities and poverty-stricken populations. The resulting social disruption has contributed to festering Maoist insurgencies in 10 states. In China poorly compensated land seizures explain the rapid growth of mass disturbances: hundreds of brutally suppressed popular protests have been taking place daily (Keidel 2006). In Africa acquisition of cheap farmland by private foreign and local investors as well as large-scale infrastructure and mining projects are inducing forced evictions and threatening livelihoods (Cotula et al. 2009). Conflict prevention, disaster preparedness, and infrastructure development are closely intertwined. In postconflict settings, infrastructure rehabilitation and social reintegration of refugees, displaced individuals, and former combatants call for well-targeted development interventions. Similarly, temporary housing of displaced populations following a peace settlement or a natural disaster presents opportunities for infrastructure development geared to economic recovery and employment creation. The resulting social disruption also offers scope for speculative and commercially oriented land acquisition. For example, after the 2008 tsunami wiped out fishing settlements in Sri Lanka, the shoreline suddenly became available for state-sanctioned private investment in tourist resorts and luxury hotels. When the poverty-stricken fishing families displaced by the tsunami sought to return and rebuild their homes, the authorities prevented them from doing so (Klein 2007).

17. This estimate excludes population movements due to slow-onset disasters, such as droughts and sea level rises. 18. Conflict-related refugees and displaced people of concern to the United Nations High Commissioner for Refugees totaled 36.5 million in 2009. In Africa alone, the number of internally displaced people was 13 million, and refugees totaled 3.5 million.

involuntary resettlement in infrastructure projects

257

Given the basic human need for shelter and security, the ordeal of massive human displacement should not be compartmentalized in artificial categories. Cyclones and droughts trigger and prolong civil strife. Growth and reduced inequality reduce conflict risks. Infrastructure is a critical ingredient of community resilience to natural hazards. Restored livelihoods and enhanced employment prospects are facilitated by good infrastructure. Given this complementarity, pressures are building toward a unified policy framework geared to human security that addresses all facets of forced migration and displacement.

Conclusions and Recommendations Winds of change are sweeping the development scene. Following years of decline, infrastructure investment is being rediscovered as a priority for development assistance. But new donors have joined the fray, and developing countries are increasingly assertive in rejecting donors’ economic and social conditions. Most of all, private sector provision and financing of infrastructure services are on the rise. Quite apart from learning the hard-won lessons of field experience noted above, the transformation of the enabling environment has major implications for the future of infrastructure-related involuntary resettlement policies. Seven major implications are outlined below. EMBRACE VOLUNTARY RESETTLEMENT

In the current fiscally tight environment, neither the public sector nor the private sector can afford the delays and costs (including reputational risks) associated with poorly managed resettlement. At the project level, the conception of resettlement as a side issue requiring rearguard actions should be jettisoned. Beyond engineering considerations, attention to all the human consequences of infrastructure projects should guide project selection, design, and evaluation. Voluntary resettlement grounded in informed and free consent (combined with full use of market-based solutions) should increasingly supplant coercive approaches to resettlement.19 While voluntary relocation is best, eminent domain is needed . . . but as a very last resort. RECONSIDER THE ROLE OF THE STATE AND THE CIVIL SOCIETY

Public agencies are best at identifying beneficiaries and providing vouchers or subsidies, but their record at income restoration and supplying sites or dwelling units is poor (Angel 2000). Accordingly, market solutions on the supply side

19. The World Commission on Dams set up as a result of the World Bank’s Operations Evaluation Department review (2006) was tasked with securing a policy consensus among developing countries’ governments, the civil society, and private companies involved in dam construction. It utterly failed to do so, in large part because it was captured by civil society activists shrewdly led by the International Rivers Network (McCully 2001).

258

Robert Picciotto

(housing, employment, etc.) should be favored, and affected individuals and communities should be empowered to bargain effectively with the help of civil society organizations; while the displaced may lack proper title, they have rights that should be protected. Compensation should be set after taking into account the future stream of estimated project benefits. ADOPT A CONSOLIDATED POLICY FRAMEWORK

Currently, two policy paradigms are vying for influence: (1) a process-oriented approach (safeguards) focused on public sector interventions; and (2) a resultsoriented approach (performance standards) adopted for private sector interventions. Given the need for public-private sector partnerships as major instruments of infrastructure development, this dichotomy is no longer justified. Ideally, a common set of policy principles and practices drawing strength from the advantages of both paradigms should be secured. This new posture would be facilitated by a consolidated policy geared to socially and environmentally sustainable development (including resettlement) combined with intensified efforts to induce private sector investors to adopt the Equator Principles. Client ownership and accountability for results should be emphasized. Ex ante assessment of social and environmental risks should inform project selection and design. Community impacts and labor relations should receive adequate attention. Transparency backed by independent verification should apply across the board. In sum, a performance focus emphasizing risk management, innovation, and adaptability should replace detailed and rigid process strictures. ADOPT A COUNTRY-BASED APPROACH

Successful public-private partnerships assemble a synergistic mix of financial assets and skills. But they also depend on adequate regulatory frameworks, participatory mechanisms, and implementation capacities on the ground. To help improve the performance of infrastructure project finance, development assistance agencies will have to join together to complement their project-based approaches to infrastructure development with a country-based approach focused on relaxing the legal and institutional constraints that hamper a shift from involuntary to voluntary resettlement. A major commitment of aid providers to capacity development is required, especially for countries where the regulatory system is weak, property rights are insecure, and economic growth is accompanied by growing inequality. Mutual accountability and reciprocal obligations should become the bedrock of infrastructure development partnerships. Poverty-reduction strategies should make room for legal reform and land titling to help prevent land grabs and protect livelihoods. RELY ON PRINCIPLED PUBLIC-PRIVATE PARTNERSHIPS

For the developing world, infrastructure investment combined with operations and maintenance financing needs are estimated to require total expenditure re-

involuntary resettlement in infrastructure projects

259

quirements of 6.6 percent of gross domestic product (GDP) compared to actual levels of 3.0 to 4.5 percent.20 The size of the gap underlines the need for innovative infrastructure financing involving the private sector. Given the systemic problems faced by private banks following the 2008 financial crisis, long-term project finance is currently scarce, and thus multilateral development finance and insurance will continue to play a major catalytic role. Public-private partnerships will continue to multiply because major commercial and political risks are involved in the private provision of infrastructure finance and services. Full cost recovery for expenditure levels required to meet the Millennium Development Goals implies that the citizens of developing countries would have to allocate an unrealistic 25 to 35 percent of their incomes for infrastructure services, whereas as a rule of thumb the poor cannot bear to pay more than 15 percent of their income for such services (Estache 2010). Private financing is unlikely to materialize in the amounts required without greater involvement from the public sector and aid agencies and increased readiness to fund targeted subsidies. Over the long run, the increased access of developing countries to global capital markets will call for international mechanisms to address cross-border regulation, competition rules, and compatibility between national competition laws. RESPECT THE PRINCIPLES OF THE PARIS DECLARATION

Harmonization of policy frameworks across the development system has also become imperative in order to avoid further aid fragmentation, to reduce transaction costs, and to enhance alignment with country systems. In this context, reaching out to new development partners, including the emerging market donor countries (many of which favor infrastructure financing), is a critical priority.21 For example, increased south-south cooperation is illustrated by the fact that 35 African countries are engaging with China to fund and help construct large-scale infrastructure projects with emphasis on hydropower generation and railways (Foster et al. 2008). WORK TOWARD A UNIFIED HUMAN DISPLACEMENT POLICY REGIME

Equitable and sustainable development through socially responsible resettlement springs from the same altruistic impulse as mitigating the pernicious impacts of natural disasters and violent conflict. The constituencies for development assistance, humanitarian aid, and peace building overlap. Reduction of inequality,

20. China’s public infrastructure expenditure as a share of GDP was estimated at 9 percent from 1998 to 2002. 21. A conservative estimate of aid provided by emerging market countries and private sources that have not subscribed to the Paris Declaration is US$28–29.5 billion annually, compared to US$129 billion of official development aid provided by OECD’s Development Assistance Committee donors.

260

Robert Picciotto

protection of vulnerable groups, and increased help to poor people trapped in zones of fragility are mere strands in a single policy tapestry focused on human security. From this perspective, policy convergence between development-induced, conflict-induced, and natural emergency–induced displacement regimes would be desirable. Under a single human security policy umbrella, such a regime would provide guidance to states, international and nongovernmental organizations, and private corporations about the principles and values that should inform mitigation of human displacement risks and costs. To be sure, the norms, rules, and processes applicable in diverse contexts and for different actors would vary, but they would be anchored in the same shared values and the same universal aspirations for human security and human progress.

references Angel, S. 2000. Housing policy matters: A global analysis. New York: Oxford University Press. Barabas, A., and M. Bartolome. 1973. Hydraulic development and ethnocide: The Mazatec and Chinantec people of Oaxaca (Mexico). Document 15. Copenhagen: International Work Group for Indigenous Affairs. Benson, B. L., ed. 2010. Property rights: Eminent domain and regulatory takings reexamined. New York: Palgrave Macmillan. Calderon, C., and L. Serven. 2004. The effects of infrastructure development on growth and income distribution. Research Working Paper No. 3400. Washington, DC: World Bank. Cernea, M. 1997. African involuntary resettlement in a global context. Social Assessment Series 045. Washington, DC: World Bank Environment Department. ———. 2006. Development-induced and conflict-induced IDPs: Bridging the research divide. Special Issue. Forced Migration Review (December):25–27. www.fmreview .org/FMRpdfs/BrookingsSpecial/15.pdf. Clark, D. 2000. Resettlement: The World Bank’s assault on the poor. Washington, DC: Center for Environmental Law. Cohen, J., and W. Easterly, eds. 2009. What works in development? Thinking big and thinking small. Washington, DC: Brookings Institution Press. Colchester, M. 2000. Dams, indigenous peoples and ethnic minorities. Thematic Review 1.2. Cape Town: World Commission on Dams. www.forestpeoples.org/sites /fpp/files/publication/2010/08/damsipsethnicminoritiesnov00eng.pdf. Cotula, L., S. Vermeulen, R. Leonard, and J. Keeley. 2009. Land grab or development opportunity? Agricultural investment and international land deals in Africa. Rome: Food and Agriculture Organization, International Fund for Agricultural Development, and IIED. Delli Priscoli, J. 2011. Two decades at the center of world water policy: Interview with John Briscoe. Water Policy 13:147–160. Durand-Lasserve, A. 2007. Market-driven eviction processes in developing country cities: The cases of Kigali in Rwanda and Phnom Penh in Cambodia. Global Urban Development 3(1):1–14.

involuntary resettlement in infrastructure projects

261

Estache, A. 2010. Infrastructure finance in developing countries: An overview. European Investment Bank Papers 15(2):60–88. Foster, V., W. Butterfield, C. Chen, and N. Pushak. 2008. Building bridges: China’s growing role as infrastructure financier for sub-Saharan Africa. Washington, DC: World Bank. George, H. 1953. Progress and poverty. London: Hogarth Press. Ingram, G. K., Z. Liu, and K. Brandt. 2013. Metropolitan infrastructure and capital finance. In Financing metropolitan governments in developing countries, ed. R. Bahl, J. Linn, and D. Wetzel. Cambridge, MA: Lincoln Institute of Land Policy. Independent Evaluation Group. 2010. Safeguards and sustainability policies in a changing world: An independent evaluation of the World Bank group experience. Washington, DC: World Bank. Keidel, A. 2006. China’s social unrest: The story behind the stories. Policy Brief No. 48. Washington, DC: Carnegie Endowment. Klein, N. 2007. The shock doctrine: The rise of disaster capitalism. London: Penguin. Leipziger, D., M. Fay, Q. Wodon, and T. Yepes. 2003. Achieving the Millennium Development Goals: The role of infrastructure. Research Working Paper 3163. Washington, DC: World Bank. Maldonado, J. K. 2010. A new path forward: Researching and reflecting on forced displacement and resettlement. Journal of Refugee Studies 25(2):193–220. McCully, P. 2001. The use of a trilateral network: An activist’s perspective on the formation of the World Commission on Dams. American University Law Review 16(6):1453–1475. Muggah, R. 2008. Relocation failures in Sri Lanka: A short history of internal displacement and resettlement. London: Zed Books. Nowak, J. E., and R. D. Rotunda. 2004. Constitutional law. Hornbook Series. St. Paul, MN: Thomson/ West. Oliver-Smith, A., ed. 2009. Development and dispossession: The crisis of forced displacement and resettlement. Santa Fe, NM: School for Advanced Social Research. Operations Evaluation Department. 1996. World Bank lending for large dams: A preliminary review of impacts. OED Precis No. 125. Washington, DC: World Bank. ———. 1998. Recent experience with involuntary resettlement. Washington, DC: World Bank. Picciotto, R., W. van Wicklin, and E. Rice. 2001. Involuntary resettlement: Comparative perspectives. New Brunswick, NJ: Transaction. Scudder, T. 2005. Future of large dams: Dealing with social, environmental, institutional and political costs. London: Earthscan. United Nations. 2009. Monitoring disaster displacement in the context of climate change. Geneva: United Nations Office of Coordination of Humanitarian Affairs and the Internal Displacement Monitoring Centre. Witness for Peace. 1996. A people damned: The impact of the World Bank Chixoy Hydroelectric Project in Guatemala. Washington, DC: Witness for Peace. World Bank. 1994a. The bankwide review of projects involving resettlement, 1986– 1993. Washington, DC: World Bank Environment Department. ———. 1994b. World development report. Washington, DC: World Bank. ———. 2004. Involuntary resettlement source book. Washington, DC: World Bank.

262

Robert Picciotto

———. 2011. Operational policy 4.12: Involuntary resettlement. Operational Manual. Revised February. Washington, DC: World Bank. http://web.worldbank.org /WBSITE/EXTERNAL/PROJECTS/EXTPOLICIES/EXTOPMANUAL/0,,content MDK:20064610~menuPK:64701637~pagePK:64709096~piPK:64709108~theSite PK:502184,00.html. Yoshino, N., and M. Nakahigashi. 2000. The role of infrastructure in economic development. Tokyo: Keio University.

commentary Dolores Koenig Infrastructure construction, other development projects, natural disasters, and civil conflict all force large numbers of people to move. The negative effects on those displaced by infrastructure projects are well known and are nicely summarized in Robert Picciotto’s chapter; they include loss of livelihood, social capital, networks, and cultural meaning, particularly meaning linked to specific places. The more affluent among the displaced can use existing resources to fight displacement or to address some of its negative repercussions. In contrast, the poor are less able to fight successfully. When displaced, they suffer disproportionately because they lack resources to make up for their losses. While natural disasters and civil conflict have few favorable effects, infrastructure construction and economic development are generally considered beneficial. As Picciotto notes, infrastructure construction is highly correlated with broader economic development; there is positive change for the groups who benefit directly from it and, often, for society as a whole. While many governments try to avoid civil conflict and the worst effects of natural disasters, they usually welcome development programs, including infrastructure construction. Although the consequences of forced displacement from different causes may be similar, it is possible to do more to address the negative effects of infrastructure construction. Even with preparation, it is impossible to predict exactly when and where natural disasters and conflict will occur. In contrast, planners and implementers, with good design and adequate budgets for development programs, can avoid or mitigate the impoverishing consequences of development-caused displacement and resettlement. The key issue is to ensure that the displaced do not lose from infrastructure development, but instead gain benefits from it. To be sure, in the best cases, the displaced are included among the beneficiaries. For example, new water, sewage, or rapid transit built to serve an entire city can be used by those displaced by their construction. People would still face losses linked directly to the forced relocation, but the losses would be partially offset by new facilities. More commonly, however, the displaced do not benefit from the new infrastructure; they may be moved to new peripheral neighborhoods without access to water or sewage infrastructure, or the cost of new rapid transit may be too high for them. In the case of a dam that generates electricity, the power may be sent miles away, while farmers lose land and common pastures. Thus, the method of choice to address impoverishment has become the resettlement program: coordinated interventions intended to improve the lives of those forced to move. Infrastructure projects that generate income should also include formal benefit-sharing mechanisms that channel benefits directly to the displaced (Cernea and Mathur 2008). A society interested in equitable development needs to create resettlement initiatives that not only avoid the impoverishment of those displaced by new 263

264

Dolores Koenig

infrastructure or other economic development programs, but also improve their lives. Picciotto’s chapter, which focuses on the international donor-country client relationship, raises this issue only obliquely. He points out the effects of involuntary resettlement on the displaced, but he does not always keep improvement of their lives and livelihoods at the center of the discussion. Nevertheless, the major purpose of involuntary resettlement policies and guidelines is to improve outcomes for the displaced. Therefore, their needs and an understanding of social justice for them must be central. This raises the complex issue of what criteria should be used to judge successful resettlement. Possible strategies include the use of international human rights law, implicit in Picciotto’s discussion of the critiques of human rights activists, or Cernea’s (2000) model, which outlines major displacement risks and how to counteract them. The creation and evaluation of involuntary resettlement programs need to start from a clear understanding of the goals. Even if the goals of involuntary resettlement policies and programs are clear, the question remains of the best strategies to achieve them. A few suggestions offered in Picciotto’s chapter appear to be somewhat problematic. To be sure, the goal of putting development decisions into the hands of developing countries is worthwhile. But resettlement guidelines were put into place precisely because countries did not always look out sufficiently for their own poor populations. Picciotto discusses the power differentials between those deciding on infrastructure projects and those forced to move, the political use of resettlement by some countries, and the lack of political will among some governments. In this context, some practitioners, like Cernea (2005), remain skeptical of the ability of most countries to implement policies that serve the displaced. Moreover, activists in developing countries may use the existence of international regulations to push for better country policies. There are also limits on rendering involuntary resettlement more voluntary. The primary strategy suggested is economic: better compensation for those affected. No doubt, levels of compensation have historically been insufficient to achieve the desired improvements in livelihoods (Cernea and Mathur 2008). Thus, the affected virtually always try to stop or limit displacement or to make projects work more in their favor. When they bring lawsuits, carry out demonstrations, or try to take advantage of project benefits for which they do not qualify, their actions are not always appreciated by planners and implementation teams. Bringing in the affected at earlier stages of project design and implementation and giving them the ability to make significant decisions about the future parameters of their lives should improve resettlement. Moreover, people’s actions in response to displacement and resettlement need to be understood through a more complex vision of civil society and social action that is not limited to formal associations. Both higher compensation and more participation in decision making and implementation of resettlement initiatives will make projects more voluntary and move them toward displacement and resettlement based on free, prior, and in-

commentary

265

formed consent. Nevertheless, as Picciotto mentions, the losses sustained by the displaced are not simply economic, but also social and cultural. Strong attachments to people and places are significant obstacles to voluntary resettlement, and people suffer physiological and psychological stress when they are required to move (Scudder 2005). It is also difficult if not impossible to achieve consent in heterogeneous communities where people have divergent interests according to class, age, gender, and often ethnicity, religion, and other characteristics. Gaining agreement among the affected is a slow process, too slow to be considered realistic in many infrastructure construction projects. Thus, even though the situation can be improved, analysts like Scudder (2005) remain skeptical that involuntary resettlement can ever be made completely voluntary.

references Cernea, M. 2000. Risks, safeguards, and reconstruction: A model for population displacement and resettlement. In Risks and reconstruction: Experiences of resettlers and refugees, ed. M. Cernea and C. McDowell, 11–55. Washington, DC: World Bank. ———. 2005. The “ripple effect” in social policy and its political content: Social standards in public and private sector development projects. In Privatising development: Transnational law, infrastructure and human rights, ed. M. Liskosky, 65–101. Leiden: Martinus Nijhoff. Cernea, M., and H. M. Mathur, eds. 2008. Can compensation prevent impoverishment? Reforming resettlement through investments and benefit-sharing. New Delhi: Oxford University Press. Scudder, T. 2005. The future of large dams: Dealing with social, environmental, institutional and political costs. London and Sterling, VA: Earthscan.

Improving Sustainability and Efficiency

10 Sustainable Infrastructure for Urban Growth Katherine Sierra

I

n June 2012, world leaders made their way to Rio de Janeiro for the 20th anniversary of the 1992 United Nations Conference on Environment and Development—the Earth Summit—to attempt to provide a new political momentum for sustainable development. They faced a vastly different, and more difficult, terrain than their predecessors encountered when they met 20 years earlier. Increases in population—today’s global population of 7 billion is projected to reach 9 billion by 2050—are putting pressure on both the natural and built environments. Urbanization is surging, and Asia and Africa in particular will experience tremendous growth in their urban populations, which will strain urban service delivery systems and infrastructure (table 10.1).

The Challenge The effects of climate change make these challenges more acute. Global greenhouse gas (GHG) emissions continue to grow to dangerous levels, and cities are responsible for most of the emissions and their projected growth (Solomon et al. 2007). The United Nations climate negotiations set a goal of limiting global temperature increases to no more than 2°C over preindustrial levels. However, this goal is not likely to be reached under current policies. Indeed, the world’s GHG emissions growth is on a path consistent with a long-term average temperature increase of more than 3.5°C (International Energy Agency 2011).1 The effects of

1. The International Energy Agency’s most recent World Energy Outlook 2011 concluded that the world “cannot afford to delay further action to tackle climate change if the long-term

269

270

Katherine Sierra

Table 10.1 The Urban Transition by the Numbers World population Urban population Urban population share Urban share of greenhouse gas emissions

2011

2050

7.0 billion 3.6 billion 51 percent 75 percent

9.3 billion 6.3 billion 68 percent

0.4 billion 1.3 billion

1.2 billion 3.3 billion

Urban Population in the Most Rapidly Urbanizing Regions Africa Asia

Increase in Urban Dwellers by 2050 in the Top Five Countries India China Nigeria United States Indonesia

487 million 342 million 200 million 103 million 92 million

Sources: United Nations (2012); United Nations Environment Program (2011).

a changing climate are already being experienced, prompting a struggle to find ways to enable societies to achieve more resilient growth. Cities are at the center of the growth in GHG emissions. As centers of economic activity, cities are responsible for 75 percent of GHG emissions.2 Cities, especially those located on seacoasts and deltas, are also particularly vulnerable to the impacts of climate change (Fuchs et al. 2011). Yet cities are also well placed to take advantage of the opportunities that new technologies, innovative financing, and new partnerships with the private sector and civil society bring that will allow a new kind of infrastructure investment to support a sustainable future.

target of limiting the global average temperature increase to 2°C, as analysed in the 450 Scenario, is to be achieved at reasonable cost. In the New Policies Scenario, the world is on a trajectory that results in a level of emissions consistent with a long-term average temperature increase of more than 3.5°C. Without these new policies, we are on an even more dangerous track, for a temperature increase of 6°C or more.” The New Policies Scenario assumes that recent government policy commitments will be implemented in a cautious manner—even if they are not yet backed up by firm measures in place. 2. United Nations Environment Program (2011), citing Kamal-Chaoui and Robert (2009).

sustainable infrastructure for urban growth

271

This chapter describes how sustainable urban infrastructure is being redefined in the wake of these trends. It looks at how transformative city sustainability strategies and associated infrastructure investment plans are being supported by a menu of policy tools and innovative financing. The chapter then discusses how these trends are playing out in the provision of sustainable urban infrastructure in three settings: scaling up sustainable urban transportation in Mexico, building resilience in Jakarta, and driving innovation through new energy business models in Austin, Texas.

New Directions: Green Growth and Sustainable Urban Infrastructure Sustainable development was defined by the Brundtland Commission as development that meets the needs of the present without compromising the ability of future generations to meet their own needs (United Nations 1987).3 Sustainability looks for balance between three dimensions of development: economic, environmental, and social. Although this definition still stands after more than 25 years, the sense of urgency resulting from today’s economic, demographic, and environmental trends has prompted a debate on how to best achieve sustainable development with a “green growth” focus. The Organisation for Economic Co-operation and Development (OECD) describes green growth as a way to foster “economic growth and development while ensuring that natural assets continue to provide the resources and ecosystem services on which our well-being relies. To do this it must catalyse investment, competition and innovation, which will underpin sustained growth and give rise to new economic opportunities” (2011b, 9). The World Bank calls for inclusive green growth as necessary, efficient, and affordable (2012). The vision of the United Nations Environment Program is for green cities that “combine greater productivity and innovation capacity with lower costs and reduced environmental impact” (2011, 454). Green growth advocates call for sustainable, green urban infrastructure strategies that provide a platform for U U

efficiency, with more compact and denser cities with lower infrastructure and energy costs per capita as density rises, and with further efficiencies in energy and water use through demand management; high quality of life, with better air quality attained by reducing emissions and investing in cleaner energy solutions; fast, affordable, and lowemission mass transportation to employment and schools; and efficient provision of clean water and use of advanced waste systems;

3. The Brundtland Commission’s formal name was the United Nations World Commission on Environment and Development.

272

U U

U

U

Katherine Sierra

inclusion, with affordable and accessible transportation, modern energy, and clean water and waste management services for the poor; resiliency, with infrastructure built to withstand a wider range of weather risks given uncertainties about the local effects of climate change and the use of ecosystem services, such as preservation of mangroves or other natural buffers, and institutional and social development such as that needed to support early warning systems, instead of always turning to hard infrastructure solutions; innovation, taking advantage of the proximity of economic activities to support technological breakthroughs, particularly in the area of renewable energy technologies, and capitalizing on advances in information and communications technologies; and productivity and competitiveness, with green growth strategies creating new jobs and growth based on new technologies and new ways of doing business.4

Leading cities are taking action without waiting for global policies. These first-mover cities, such as those that have come together as part of the C40 Climate Leadership Group,5 are starting to design and implement new strategies that reduce their carbon and water footprints, build resilience, and address social inclusion while also looking to create jobs and grow their economies. Sometimes these cities are working in concert with supportive national policies, and sometimes they are innovating ahead of national policies (Carbon Disclosure Project 2011). Cities are taking advantage of new energy technologies and promoting energy efficiency as a way to reduce their carbon footprint. A revolution in information and communications technologies has allowed connectivity to drive new “smart” solutions that are bringing infrastructure costs down through increased efficiency. This revolution has also upended business models and has prompted entrepreneurs and policy makers alike to look for new approaches that can leapfrog old technologies. Cities in emerging economies like China, India, Brazil, and South Africa are joining with those in countries that belong to the OECD as sources of innovation. New types of multi-stakeholder partnerships are forming, with businesses, universities, and civil society organizations joining with public 4. A recent study found that U.S. clean technology industries—sectors like the solar thermal, solar photovoltaic, wind, fuel cell, biofuel, and smart grid industries—together grew at twice the rate of the rest of the economy, adding jobs at over 8 percent a year between 2003 and 2010 (Muro, Rothwell, and Saha 2011). The study also concluded that there were more than 830,000 jobs in energy efficiency (including public mass transit, energy-saving building materials, green architecture and construction, and other categories). 5. The C40 Cities Climate Leadership Group is a network of large cities from around the world that are committed to implementing meaningful and sustainable climate-related actions locally that will help address climate change globally. See http://live.c40cities.org/.

sustainable infrastructure for urban growth

273

policy makers to look for game-changing solutions to local, national, and global challenges. These strategies need to be pursued in ways tailored to the particular context. Urban infrastructure networks have already been largely laid down in the developed world. Planners can only make incremental changes to city form, but they can focus on strategies for rehabilitation and retrofitting, along with the use of information and communications technologies to drive efficiencies and more aggressive demand management to reduce resource use and lower emissions.6 In the developing and emerging economies, particularly those in Africa and Asia that are the center of today’s urban transition, cities can make infrastructure decisions now that will use emerging technologies and processes to create a more efficient, denser footprint. They can also take advantage of more advanced technologies where they are affordable. A transformative strategy is needed, given the scale of the challenge. Thus, green and sustainable urban growth will require a supportive enabling environment and innovative financing methods to facilitate the introduction of new infrastructure technologies and new ways of doing business.

The Policy Challenge: An Enabling Environment for Transformative Action THE NATIONAL POLICY FRAMEWORK

Although urban policy makers can work within their own spheres of influence to develop and implement sustainability strategies, supportive national policies are critical to achieve change at the scale and speed needed.7 National policies can best do this by setting the right pricing signals, particularly by setting a price on carbon through either a cap-and-trade arrangement or a carbon tax. Australia has just joined Europe in introducing a cap-and-trade system, while China is using multiple local experiments to develop its own cap-and-trade policy. The elimination of fossil fuel subsidies is also needed to level the playing field with new technologies. The Group of 20 (G-20) leaders agreed on this at a meeting in Pittsburgh, but implementation has been elusive (Hultman, Sierra, and Carlock 2011). Governments are supporting an acceleration of renewable energy investments through subsidies, but these are under pressure in the countries that belong to the OECD because of fiscal constraints.

6. The International Energy Agency (IEA) estimates that four-fifths of the total energy-related CO2 emissions permissible by 2035 to maintain the global temperature increases to within 2°C are already “locked in” by existing global capital stock, such as power plants, buildings, and factories (International Energy Agency 2011). 7. See Organisation for Economic Co-operation and Development (2011a), which provides a rich analysis of policy options, and Organisation for Economic Co-operation and Development (2012).

274

Katherine Sierra

City leaders have also called for a simplification of national policy frameworks, citing the high costs and uncertainties they face in dealing with complex sets of policies. For example, in the United States the absence of a national climate policy has given way to a web of national and state energy sector mandates and incentive programs. Many of these are set to expire in the next few years, introducing both significant uncertainty for public and private investments alike and also fiscal risks for local authorities (Jenkins et al. 2012).8 National water strategies are also laying the groundwork for improved sustainability of local infrastructure investment and services, with examples coming from countries that are experiencing acute water stress (Wallis et al. 2009). For instance, Chile was a pioneer in the use of water markets; it made legislative reforms in 2005 that seek to improve both river basin governance and the coordination of multiple water uses, solutions to water conflicts, and environmental protection. In Australia, the National Water Initiative aims to deal with overallocated and overstressed water systems by introducing registers of water rights and standards for water accounting, improving pricing for water storage and delivery, and expanding water trade.9 China’s Five-Year Plan calls for investing $600 billion over a period of 10 years on measures to cope with serious water scarcity in both urban and rural settings. This plan relies heavily on infrastructure solutions (e.g., the South-North Water Transfer Project, a massive infrastructure project that would transfer water from the Yangtze River to relieve acute water stress and environmental pollution in China’s northern regions and cities).10 THE LOCAL POLICY FRAMEWORK

City sustainability strategies can set the vision, provide a mandate for action, and guide the development of incentives for consumers and business, while also driving infrastructure investment priorities. An analysis of C40 city master plans carried out by the Carbon Disclosure Project (2011) looked at sustainability strategies that leading cities have adopted to reduce their GHG emissions and enhance resilience (figure 10.1).11 Common themes in city master plans include first focusing on energy savings and building efficiency, and then investing in in-

8. Jenkins et al. (2012) argue for a reform of U.S. energy deployment subsidies and policies so that they reward technology improvements and cost reductions while strengthening energy innovation systems to drive down costs. 9. See http://nwc.gov.au/nwi and www.environment.gov.au/water/australia/nwi/index.html. 10. See www.wantchinatimes.com/news-subclass-cnt.aspx?cid=1105&MainCatID=&id=2011 0711000040. 11. This section draws, where noted, on the findings from 42 of 58 C40 cities, which submitted responses to a questionnaire from the Carbon Disclosure Project covering governance, greenhouse gas emissions, adaptation, and strategy. The findings were documented by the Carbon Disclosure Project (2011).

sustainable infrastructure for urban growth

275

Figure 10.1 C40 Cities’ Mitigation Measures Technical solutions

3%

Water management

9%

Tree planting

9%

District heating

9%

Permitting incentives

13%

Waste treatment

25%

Retrofitting

25%

Renewable energies

25%

Transport

28%

Infrastructure and urban planning

34%

Awareness and consultation

53%

Building standards

59%

Subsidies and fiscal incentives

66% 0

10

20

30

40

50

60

70

Percentage (Number of responding cities = 32)

frastructure and creating incentives to increase the share of renewable energy and green transportation measures. Figure 10.1 For example, Seoul has adopted Lincoln_Ingram_Infrastructure a citywide target of 40 percent GHG emissions reduction by 2030, focusing on buildings, energy efficiency, transportation, and an increased share of renewable energy in its mix. The Sustainable Sydney 2030 plan calls for a 70 percent reduction in emissions by 2030. This is expected to be achieved through Sydney’s partnerships with businesses to increase energy efficiency; investments in LED streetlights and solar panels on city buildings; and the use of combined cooling, heating, and power trigeneration techniques to supply local power, heating, and cooling in city buildings. The city’s new water master plan anticipates that half of its water supply will come from a citywide decentralized recycled water network. It is also investing in light rail and bikeways

276

Katherine Sierra

(Sydney 2011). Jakarta has adopted climate change goals, preparing a road map for a 30 percent reduction in GHG emissions across the transportation, waste, and energy sectors (Soehodo 2011). THE LOCAL POLICY TOOL KIT

Local policy makers, in turn, have a wide set of tools that they can use to build an attractive regulatory environment for sustainable development (table 10.2). Indeed, subsidies and fiscal incentives, along with building standards, were the most popular GHG reduction activities reported by C40 cities (figure 10.1) and were commonly used by the cities studied by researchers at the OECD (2011a, 2011b). Incentives include property tax reductions for high-density development

Table 10.2 Greening the Local Fiscal Tool Kit Goal

Tools to Incentivize Sustainable Behavior and Investment

More compact cities

V

Sustainable transportation

Efficient environmental services

Improved building and industrial energy efficiency

Increased share of renewables in energy mix

V V V V V V V V V V V V V V V V V V V

Reformation of property tax on multifamily dwellings to incentivize high-density development Split-rate property tax to incentivize infill development Cascading property taxes that rise with distance from core Development charges that internalize externalities of sprawl Reformation of land sale plans Congestion pricing and cordon tolls Higher and variable parking charges Taxes on vehicle ownership and use Value-capture tax to finance public transportation Cost-recovery water and waste charges to incentivize efficient use Tariff-based incentives for waste recycling Water rebates for conversion to water-efficient appliances and fixtures Tax incentives, credits for investment in building energy efficiency Tax credits, subsidies for industrial energy efficiency; tradable permits Tax incentives, rebates for investment in Smart Grid Public utility charges to fund energy conservation programs Feed-in tariffs and long-term power purchase agreements Tax rebates for installation of solar photovoltaic systems Net metering that allows consumers and producers that produce energy to sell to grid Programs that give households and businesses option to purchase renewable energy at premium price

Sources: Adapted from Banks et al. (2011); Carbon Disclosure Project (2011); Organisation for Economic Co-operation and Development (2011a, 2011b).

sustainable infrastructure for urban growth

277

and the reformation of land sale plans, which in countries like China fund infrastructure development yet also exacerbate sprawl. Congestion pricing, variable parking rates, and value capture taxes can support sustainable transit. Water and wastewater recovery charges, along with incentives for recycling and investment in water-efficient appliances, help manage demand while also reducing the cost of infrastructure investments and improving utilities’ finances (Sykes et al. 2010). In those cities that manage energy utilities, feed-in tariffs and power purchase agreements support renewable goals. Decentralized energy is being encouraged with net metering programs and tax incentives for investments in solar photovoltaic systems and participation in community-based smart grid programs. The use of these tools will be important for sustainable infrastructure strategies because they will help set the course for city density and the intensity of public improvements, and services like public transit (which in turn drives infrastructure costs and emissions); improve demand management and efficiency; provide incentives for businesses and consumers to invest in new approaches, like renewables; and enhance financial sustainability through cost recovery.

The Financing Challenge: Scaling Up Through Innovative Finance Global infrastructure investment requirements over the next several decades will be significant. The OECD (2007) estimates that investments in electricity, transportation, water, and telecommunications could average some 2.5 percent of the world’s gross domestic product. The International Energy Agency (2011) estimates that $38 trillion in global investment in energy-supply infrastructure will be required from 2011 to 2035. Two-thirds of this is required in non-OECD countries.12 The amount of financing needed to reduce GHG emissions and for adaptation is large, and public finance is insufficient to meet this need, particularly in developing countries. Net climate change mitigation costs in developing countries, over and above the cost of business-as-usual investment needed for economic development, are estimated in the range of $60 to $175 billion a year (World Bank 2010d). Even if the United Nations’ 2°C goal were to be achieved, countries would already be facing the costs of a changing climate. Adaptation costs are estimated to range from $75 to $100 billion a year over and above the investment costs of a business-as-usual development trajectory (World Bank 2011a). The menu for financing urban infrastructure that will shift communities to a sustainable path shares the same core elements well known to urban policy makers. Cities raise revenue from taxation, fees, and (especially in developing countries) land sales; and utilities cover their costs through user fees. As discussed earlier, these can enhance sustainable outcomes through demand management while also

12. This is according to the IEA “New Policies Scenario” (see International Energy Agency 2011).

278

Katherine Sierra

bolstering city and utility finances. Intergovernmental transfers through grants or subsidized loans cofinance public investments when there are broader national, state, and global public good considerations that are not captured by local benefits. The green stimulus packages introduced in South Korea, the United States, and China in response to the financial crisis are one example. Cities can benefit from renewable energy subsidies—directly where they are responsible for power generation or indirectly through national clean energy investment programs. Private capital is leveraged through public-private partnerships, while municipal bonds and investments by pools of private capital (e.g., pension funds13) provide long-tenor financing for low-risk, steady returns from infrastructure investments for cities and utilities with investment-grade policies. However, though the core financing menu may be similar, new types of sustainable infrastructure and associated services face financing barriers beyond the already-large challenge of meeting growing demands for infrastructure to simply meet population growth and to replace aging infrastructure stocks. Four broad categories of barriers are most commonly cited:14 t t

t

t

Investment climate and borrowing capacity at the national and local levels. Sector-specific barriers, which include concerns about the stability and certainty of the sector policy and regulatory framework—such as the longevity of power purchase agreements or feed-in tariff programs; technology risks for investments in new and relatively untried technologies and systems; and execution and unfamiliarity risks where there are concerns about capacity to execute projects. Other barriers include stakeholder resistance to the often-complex changes implied by new approaches that may undermine project delivery. Capacity and knowledge gaps, which include the low capacity available to prepare project pipelines and to structure projects,15 a lack of skilled and semiskilled labor for new industries, and a lack of established engineering, procurement, and construction contractors. Technology cost gaps, which are the residual cost gaps between high- and low-emission alternatives after accounting for the costs of carbon that are built into existing international and national policies (e.g., efficiency standards, carbon taxes, removal of fossil fuel subsidies, and intergovernmen-

13. OECD pension funds in 2007 were estimated to total $18 trillion, up from $13 trillion in 2001 (Organisation for Economic Co-operation and Development 2007). 14. This section has been adapted from Sierra (2011b), and it also draws on work by Brown and Jacobs (2011) and United Nations (2010). 15. Grants for advisory services and technical assistance can be used to accelerate the development of a viable pipeline through feasibility studies, including technical, engineering, economic, financial, social, and environmental studies, and to provide support for legal and advisory services.

sustainable infrastructure for urban growth

279

tal grants or subsidies like feed-in tariffs). These costs may also be derived from inadequate network infrastructure, such as transmission lines linking renewable resources to the main grid. Given these barriers to the introduction of new technologies, cities are hardpressed to move to higher-cost or higher-risk approaches in the absence of financial support and capacity building. The 2008 global financial and economic crisis continues to have an impact on the economies of the United States, the European countries, and Japan, and worrying fiscal deficits mean that public resources for investment will be scarce for the near to medium terms. At the international level, though developed countries have pledged as part of the United Nations Framework Convention on Climate Change negotiations to provide $100 billion per year in climate finance by 2020 to help developing countries move to low-carbon and resilient growth pathways, the source of that funding is highly uncertain. As a result of these trends, policy makers are looking for innovative financing solutions that seek to create new, or to redirect existing, international climate funds to support transformation in developing countries. Experiments are under way on the use of public funds to leverage private capital. Finally, carbon markets could also support investments in urban sustainable infrastructure, but this avenue is still relatively underdeveloped (World Bank 2011b). INTERNATIONAL CLIMATE FINANCE

International transfers of grants or highly concessional loans aim to accelerate the introduction of new technologies to developing countries by reducing costs and risks. In theory, they also support the additional costs imposed by a changing climate, but the level of funding available is far below needs. Support for developing countries was initially channeled through the Global Environment Facility (GEF), which helped pilot new renewable energy and energy-efficient technologies by providing grants to cover the additional costs associated with their deployment (Global Environment Facility 2009). In 2008, the Climate Investment Funds (CIFs) were created to channel more than $6 billion in highly concessional funds to mitigate and adapt to the effects of climate change.16 The majority of the funding goes to the Clean Technology Fund (CTF), which finances programs that aim to make a transformative impact in encouraging renewable energy, energy efficiency, and clean transportation.17 The CTF supports both public and private investment. Support for private investment aims to reduce the barriers for early market entrants so that later

16. The author managed the negotiations and creation of the CIFs. 17. Sector-specific country investment plans financed by the Clean Technology Fund were undertaken for Egypt, Mexico, the Philippines, Vietnam, Colombia, and Nigeria (see Clean Technology Fund 2009a, 2009b, 2009c, 2009d, 2010a, 2010b, 2010c, and 2011).

280

Katherine Sierra

investors, developers, and financial intermediaries will subsequently enter the market without additional support. The Adaptation Fund and the CIF Pilot Program for Climate Resilience are the main multilateral sources of funding for adaptation. A new Green Climate Fund (GCF) was agreed to at the 2010 climate negotiations in Cancún and was finalized in Durban in 2011. The board for this fund met for the first time in August 2012, and work to operationalize the fund is ongoing. The multilateral development banks—like the World Bank, the European Investment Bank, and the regional development banks—cofinance climate-related projects in cooperation with the GEF and the CIFs. They are also looking to direct more of their own financing to sustainable infrastructure.18 Funding also comes from bilateral sources, such as the German International Climate Initiative and the United Kingdom’s International Climate Fund, and from bilateral institutions, such as the United States’ Overseas Private Investment Corporation, national export-import banks, and credit guarantee agencies. National development banks in the developing world are also supporting sustainable infrastructure investments, though this is still nascent. BNDES (Banco Nacional de Desenvolvimento Econômico e Social), Brazil’s national development bank, and Banobras (Banco Nacional de Obras y Servicios Públicos), Mexico’s national development bank, are examples of domestic institutions that are beginning to channel intergovernmental transfers for sustainable development. LEVERAGING PRIVATE CAPITAL

Despite an increased interest in specialized funding for investments related to climate and sustainable infrastructure, the current tight fiscal situation means that public funds will fall far short of the need. As a result, there is growing interest in using scarce public funds to leverage private capital. A number of new instruments are being rolled out or are under development (table 10.3): U

Green banks. Green investment banks could provide debt financing and, depending on the national legal setting, issue bonds and seek patient investors looking for a long-term conservative rate of return, such as pension fund investors (Berlin et al. 2012). In the United Kingdom a green bank is being created that will be capitalized with national government funds but allowed to raise its own financing. It is expected that the bank will fill a gap in the market for government-backed bonds, bring in banking expertise, and offer a range of commercially driven interventions—loans, equity, and risk-reduction finance (Environmental Audit Committee 2011). In June 2011 Connecticut created the Clean Energy Finance and Investment

18. The World Bank’s Sustainable Infrastructure Action Plan is one example; see http://water .worldbank.org/publications/world-bank-group-sustainable-infrastructure-action-plan.

sustainable infrastructure for urban growth

281

Table 10.3 Using Public Finance to Leverage Private Capital for Sustainable Infrastructure Goal

Tool

Increase local access to large pools of private capital—pension funds, private equity, sovereign wealth funds—for sustainable infrastructure

Green or sustainable infrastructure banks of funds to attract private capital while lowering debt service costs and increasing tenors: V V

V V V

Direct intergovernmental or intragovernmental transfers to bankable projects. Lower debt service costs where investors perceive more risk in projects or approaches without a track record, including concerns that cities may lack capacity or implementation experience. Support private sector participation by reducing risk through policy guarantees, insurance, first-loss instruments, subordinated debt, or equity. Promote local commercial banking capacity building by providing risk sharing for new products (energy efficiency lines of credit). Bundle and securitize small and dispersed investments (building energy efficiency).

Pledge funds: Public fund pledges to provide a small amount of equity to pooled funds to encourage much larger pledges from private investors like sovereign wealth funds, private equity, and pension funds for investments in sustainable urban infrastructure under public-private partnership arrangements.

Speed up and deepen bond market for sustainable investments allowing access to large pools of capital, reduce the average cost of capital, and provide a low-cost exit for construction-phase capital and for bank long-term debt

Fund of funds: The public funder invests as a limited partner in a private fund that holds a portfolio of other private investment funds. Green bonds: Public financing (through public institutions like the green banks or international financial institutions) supports first-loss tranches or partial guarantees from early bond issuances in new asset classes and/or in countries with less developed capital markets; these have not yet been adapted for the municipal bond market.

Source: Adapted from Sierra (2011b).

Authority.19 This quasi-public green bank combines several existing state clean energy and energy efficiency funds in a structure that allows these

19. Section 99 of Public Act 11-80, An Act Concerning the Establishment of the Department of Energy and Environmental Protection and Planning for Connecticut’s Energy Future. For more information, see www.cga.ct.gov/2011/act/pa/pdf/2011PA-00080-R00SB-01243-PA.pdf.

282

U

U

U

Katherine Sierra

funds to be leveraged. Private investment in the bank is permitted, and the investors receive a reasonable rate of return on their investments.20 A proposal to use the state green bank concept more broadly is being developed for the United States (see Berlin et al. 2012). Greening of private domestic banks. Specialized climate funds are being channeled through intermediary local banks with the objective of meeting the needs of small and medium-scale project sponsors while building the capacity of the domestic banking system to appraise and price lowemission projects. An example is the Inter-American Development Bank’s Planet Banking,21 which is focusing on providing lines of credit and technical assistance to private banks that want to develop new climatecompatible products. Investing alongside private capital. A number of new initiatives for developing countries that aim to scale up by tapping pools of capital—private equity or institutional investors like pension and sovereign wealth funds— are currently being tested or planned. Pledge funds aim to mobilize private equity, sovereign wealth funds, and pension funds by investing equity or near equity alongside pooled funds (see Brown and Jacobs 2011; Center for American Progress and Global Climate Network 2010). An example is the U.S. Overseas Private Investment Corporation, which supports energyrelated investments in developing countries. Green bonds. Public finance could support green or climate bonds by holding first-loss tranches or partial guarantees from early bond issuances in developing countries, thereby helping to create a market. This market is still nascent. Green bond issuances from the World Bank have shown investor appetite for sustainable investments, but their application to municipal finance is not yet developed.

CARBON MARKETS

Cities have not been able to fully participate in carbon markets, but some participate in national cap-and-trade arrangements (e.g., through the voluntary markets) as a means to supplement resources. Others are instituting cap-and-trade systems to reduce GHG emissions (Tokyo) or local pollutants (Los Angeles, Santiago). However, it is difficult to implement GHG cap-and-trade systems at the local level without national regulations due to the global nature of emissions (Clapp et al. 2010). Another way that cities in developing countries have sought to enhance revenues is through the Clean Development Mechanism (CDM). The CDM is a

20. Senate Bill No. 1243. 21. See www.iadb.org/en/resources-for-businesses/beyondbanking/planetbanking,2081.html.

sustainable infrastructure for urban growth

283

market mechanism developed under the Kyoto Protocol that allows entities with emission reduction requirements in developed countries to buy Certified Emission Reduction (CER) credits from projects in developing countries. The voluntary carbon market is another, albeit small, source of finance. It is being used, among other things, for energy and water projects that deliver climate change mitigation co-benefits (Nakhooda et al. 2011). The CDM provides support alongside other sources of funding (e.g., tax revenues and user fees) to complete a project’s financing package. Payments are received after projects have started being implemented and contribute to the project’s cash flow. Because of uncertainties about the post-2012 carbon market plan and carbon prices in the future and the complexities of project CER certification, these flows are often not seen as integral to a financing decision but instead are a way to improve an already-viable financial rate of return. An OECD review (Clapp et al. 2010) noted that whereas carbon markets could offer support to viable urban mitigation projects, market activity has been limited. Projects focused on urban mitigation account for only 10 percent of all projects in the compliance market (Clapp et al. 2010). These are concentrated in a few sectors (energy efficiency, waste management, and energy distribution networks). Barriers include limited city authority to regulate GHG emissions, limited budgets and access to start-up capital, and lack of knowledge and institutional capacity. Further, many types of mitigation programs do not lend themselves to easy measurement (e.g., citywide transportation strategies). New products, such as those being developed by the Carbon Partnership Facility (CPF), are looking to develop new methodologies for scaled-up, programmatic approaches while targeting areas that have not been reached effectively by the Clean Development Mechanism. These products aim to pilot citywide carbon finance programs. The first CPF operation currently being planned would support the Green Growth Program of Jordan’s Amman Municipality, which includes potential opportunities in municipal waste, urban transportation, sustainable energy, and urban forestry.22

Moving to Practice The following three cases illustrate several of the strategies and approaches discussed in this chapter. In Mexico the national government is setting a policy framework that, along with innovative financing, is setting the stage to scale up sustainable urban transportation. Jakarta, which has suffered from devastating floods, is working to build resilience by improving its flood management infrastructure while it copes with weak institutional capacity. Austin, a city with vibrant technological and entrepreneurial capacities, is testing new partnerships

22. See http://wbcarbonfinance.org/Router.cfm?Page=Projport&ProjID=65753.

284

Katherine Sierra

and business models to drive innovation as it promotes sustainable energy infrastructure. SCALING UP BUS RAPID TRANSIT IN MEXICO

Bus rapid transit (BRT) has become increasingly popular among cities seeking to move large numbers of people at a low cost.23 Mexico’s climate change strategy looks to accelerate a shift to energy-efficient, low-carbon mass transit systems, building on the BRT demonstration projects in Mexico City and León.24 Mexico’s transportation sector represents 18 percent of the country’s GHG emissions, with emissions from transportation increasing by more than 2 percent a year. This increase is driven by motorization rates that are increasing by about 10 percent a year. Mexico’s goal is to have national deployment of BRT and other mass transit solutions in place by 2040, with the country’s carbon footprint from the transportation sector at 2007 levels despite expected growth in the country’s economy (Clean Technology Fund 2009b). The $2.7 billion program (table 10.4) supports the introduction of citywide BRT programs as part of an integrated package of land use planning, traffic management, and infrastructure investment. The plan is to expand the program in the larger GHG emitters like Mexico City, Guadalajara, Monterrey, Puebla, and León and begin to introduce it in small and medium-size cities like Chihuahua and Mexicali. The program calls for introduction of lower carbon emission vehicles, reorganizing fragmented owner-operated services into more efficient privately owned companies, and scrapping displaced vehicles. Institutional and Political Economy Risks Globally, there are good examples of BRT systems and useful knowledge about implementing them. But individual cities and policy makers seeking to implement BRT systems continue to face knowledge risks (i.e., how to adapt to local circumstances), political risks (resistance from fragmented, informal private operators and motorists), and institutional risks (the complexity of managing across the jurisdictions and agencies needed to implement a fully integrated sustainability plan). Mexico’s program is designed to address these barriers by providing a national vision accompanied by supportive policy and regulatory frameworks. This includes a policy framework for reorganizing private operators and promoting public-private partner-

23. This trend was documented in a recent review of the BRT literature, which confirmed that BRT systems have mostly been successful in combining the characteristics of rail systems but in a relatively cost-efficient way and with short implementation time. The effects on land use, city form, and land values are not yet well understood (Deng and Nelson 2011). 24. This section draws on project documents prepared for the CTF and World Bank loans for this program (Clean Technology Fund 2009b; World Bank 2010c). It also benefited from input from the World Bank team leader, Arturo Ardita-Gomez. See also www.cambioclimatico.gob .mx/index.php/en/politica-nacional-sobre-cambio-climatico.html.

sustainable infrastructure for urban growth

285

Table 10.4 Mexico’s Urban Transportation Transformation Program Investment Plan Financing Source FONADIN Local governments Private sector World Bank Clean Technology Fund Total program cost

Millions of Dollars (US$) 768.5 738.5 839.0 150.0 200.0 2,694.0

Note: The GEF is also supporting the program through grants for capacity building. Source: World Bank (2010c).

ships in infrastructure development. A national interagency working group was established to facilitate better coordination of planning and implementation at the city level, along with a funding mechanism, which is discussed below. Competition for Scarce Public Investment Funds These programs require significant capital investment, potentially crowding out other public investment priorities. Mexico is providing incentives to local authorities through the Federal Fund for Infrastructure (FONADIN), which is run by the national development bank (Banobras) and is funded by revenues from toll concessions. PROTRAM, an arm of FONADIN, helps direct resources to finance studies and transport infrastructure investments through the provision of grants, loans, and guarantees. These resources are complemented by low-cost financing from the CTF, which will further lower capital costs and mitigate risks.25 High Up-Front Costs of Low-Emission Technologies Low-carbon, highcapacity bus technologies have higher capital costs than conventional technologies, and introducing them entails incurring a number of transition costs. These costs include fleet scrapping and replacement programs as well as investment in fueling stations. Mexico has estimated that by adopting advanced hybrid systems for its buses, it could reduce GHG emissions by 40 percent, as compared with standard articulated diesel-powered buses. Yet this would mean incurring high up-front capital costs for procurement that would be expected to be 30 to 40 percent more than for conventional vehicles. Although operating and maintenance costs would be expected to offset the higher capital costs, the transition costs would remain a barrier. 25. The CTF’s terms are no-interest 20-year loans, with 10 years’ grace on principal repayments, a 0.75 percent service charge, and a 0.25 percent management fee.

286

Katherine Sierra

Nascent Domestic Financing for Private Sector Engagement Mexico has implemented a national policy framework for private investment in infrastructure, including provisions for competitive tenders for service provision and construction concessions. At the same time, the experiences of Mexico and other developing countries indicate that innovative financing—such as financial products supported by public finance that provide first-loss provisions, equity or quasiequity, and other forms of risk-sharing guarantees—will be needed to unlock private capital (Sierra 2011b). FONADIN is tasked with facilitating private capital for infrastructure projects, and it is prepared to take risks that the market will not yet take, including attracting private investors to projects with low yields but high social impact, while providing long-term financing at competitive rates. BUILDING RESILIENCE IN JAKARTA

Jakarta’s floods are an example of the stress experienced by the 890 million people living in cities that face a high risk of exposure to natural hazards (United Nations 2012).26 A major event in 2007 resulted in the flooding of 70 percent of Jakarta’s metropolitan area, with financial losses topping $880 million. Factors that contribute to the city’s vulnerability include its topography, with 40 percent of its area below sea level and 13 rivers flowing through the city, and land subsidence averaging 5 to 10 centimeters a year in the north of the city. The encroachment of the built-up area on critical water catchment areas has resulted in increased rainwater runoff and a lack of natural water retention both in the city and upstream. Canals and drainage systems are in poor condition due to a backlog of maintenance, with significant sediment in the canals and solid waste buildup contributing to flooding (Soehodo 2011).27 Adaptation will require the integration of climate risks into urban management strategies and investment plans through a combination of soft measures that focus on policies, institutions, knowledge and social cohesion, and ecosystem management and hard measures that focus on capital investments. Hard measures will include investment in infrastructure, such as flood protection and seawalls, as well as measures to climate-proof urban infrastructure, like increasing the capacity of drains (Satterwaithe et al. 2007; World Bank 2011a). In this vein, Jakarta is moving to integrate climate change adaptation into its development plans, most recently integrating adaptation and mitigation policies into its 2030 spatial plan.28 Its flood prevention program includes the development

26. This case builds on Sierra (2011a). 27. Related government case studies prepared by the city can be found at http://siteresources .worldbank.org/INTURBANDEVELOPMENT/Resources/336387-1306291319853/CS _ Jakarta.pdf and www.scribd.com/doc/59919883/Jakarta-s-Adaptation-Strategy-Edit. 28. A summary of the Spatial Plan 2030 climate policies can be found in Table 1 of a World Bank case study on the intersection between climate change, disaster risk, and the urban poor (World Bank n.d.).

sustainable infrastructure for urban growth

287

of flood control infrastructure, the improvement of canal drainage systems, the conservation of water areas, the building of sea dikes in the north coastal region, the rehabilitation of mangrove forests, a resettlement policy along the rivers and channels, and the raising of critical road infrastructure (Texier 2008). Communities are also adapting through early warning systems, neighborhood-level canal dredging, and increased house elevations. Nongovernmental organizations are supporting these efforts with education and planning. Jakarta’s government plans to reinforce the use of existing social networks to complement physical interventions through a community empowerment program. Implementation of this plan will be challenging. Financing Challenges Developed countries have promised global adaptation finance as part of their pledge to provide $100 billion a year by 2020 to support both mitigation and adaptation. Grants for adaptation mainly support least developed nations and small island states and will likely provide little to emerging economies like Indonesia. As such, adaptation demands will put pressure on local and regional government budgets and move those governments to use other channels for funding, like multilateral development banks.29 The city’s initial $190 million program will finance dredging and rehabilitation of floodways, canals, and retention basins, along with development of a flood management information system. The program is being financed by the local, regional, and national governments ($50 million) along with a $140 million loan from the World Bank and a $500,000 grant from the government of the Netherlands.30 The project is helping implement a new system to transfer funds from the central to local governments as part of Indonesia’s fiscal decentralization program. Knowledge Gaps Availability of data for decision making is a key constraint. Spatial information on local impacts is poor and not well integrated with information on socioeconomic vulnerability (Firman et al. 2011). At the same time, there are a number of international efforts under way to fill knowledge gaps. Japan has supported studies of the river catchment areas and work on a sewerage master plan; Australia is supporting an urban resilience study; and the Dutch government is providing flood hazard mapping and development

29. Indeed, given the strong linkage between good development practice and the actions needed to build resilience, traditional development assistance programs are also beginning to incorporate climate adaptation considerations into strategy, programming, and project design. For the urban sector, examples include the review Cities and Climate Change: An Urgent Agenda (World Bank 2010a), which sets out the issues, approaches, and partnerships that will guide its work; flagship studies like the Asian Development Bank/JICA (Japan’s bilateral development agency)/ World Bank report Climate Risks and Adaptation in Asian Coastal Megacities (World Bank 2010b); and the United Nations Habitat report Cities and Climate Change: Policy Directions (United Nations Habitat 2011). 30. The project is described in World Bank (2011c). Amounts are rounded.

288

Katherine Sierra

of a master plan for coastal management and protection. Jakarta has signed an agreement with the city of Rotterdam for knowledge exchange on best practices, and a joint research program to develop climate adaptation tools for Jakarta is bringing together academic researchers from the Netherlands and Indonesia (Jakarta Climate Adaptation Tools n.d.). Jakarta has also been active in a number of international knowledge-sharing platforms. These include the C40; the Asian Cities Climate Change Resilience Network, which supports a network of cities in Asia (Rockefeller Foundation n.d.); and the Delta Alliance network, which aims to improve the resilience of the world’s deltas through knowledge sharing (Delta Alliance n.d.). Governance Challenges Jakarta’s own assessment of its implementation experience so far points to poor coordination between the city and provincial governments as a significant barrier. This is especially evident with respect to preserving natural resources, particularly water resource management and environmental protection. Another assessment points to inadequate commitment to the programs by local officials despite public statements (Firman et al. 2011), while the World Bank/government case study points to the complexity of integrating climate change, spatial planning, and poverty alleviation. Community engagement is critical, but still a work in progress. The use of community-led processes to support local adaptive responses, despite government plans, has been assessed as uneven.31 The government has also been criticized for inadequate engagement with the community in the preparation of the Spatial Plan 2030.32 These issues are common in many emerging and developing countries where the demands of implementing complex adaptation strategies strain already low institutional capacity. Building capacity and investing in the soft elements of the adaptation agenda will be as important as hard infrastructure investment. TRANSFORMING ENERGY SYSTEMS THROUGH INNOVATION IN AUSTIN

Austin’s Climate Protection Plan, adopted in 2007, aims to build a sustainable city by enacting “policies, procedures, timelines and targets as are necessary to make Austin the leading city in the nation in the effort to reduce and reverse the negative impacts of global warming” (Office of Sustainability 2012, 2). The Austin plan covers the energy, water, waste, transportation, land use, and food sectors. Its energy goals focus on greening the power supply and conserving energy. The city met its goal to have all city facilities powered by renewable energy

31. This was documented in a case study prepared by the World Bank in coordination with the government of Jakarta, which solicited feedback from stakeholders (World Bank n.d.). 32. For example, see the article by Sabarini (2010), which reported on feedback from NGOs concerned about limited public participation.

sustainable infrastructure for urban growth

289

by 2012. The plan also calls for 800 megawatts (MW) of new energy savings through energy efficiency and conservation by 2020. All city facilities, fleets, and operations should also be carbon neutral by that date. Austin Energy, the city’s utility, is also seeking to be a leader in the field of clean energy. It is the nation’s ninth-largest community-owned electric utility, and its slogan is “More Than Electricity.” It was mandated to achieve a 35 percent share of renewable energy in its portfolio by 2020, and it expects to meet this goal mainly through the purchase of wind-generated power, with current wind contracts for 200 MW expected to increase to 1,000 MW by 2020. Austin Energy also aims to increase the commercial solar component in its energy mix to 200 MW.33 It has doubled the share of renewables in its portfolio, from 5.1 percent in 2007 to 10.3 percent in 2011 (Austin Energy 2010). The utility has also implemented a number of energy efficiency programs, most recently taking advantage of social media by launching a Facebook application that lets consumers benchmark their own home energy use against that of similar homes (Rule 2012). Austin Energy began to experiment with distributed energy in 2004, with a program of up-front rebates for the installation of commercial and residential solar systems. In 2010, it introduced a performance-based incentive system that pays for each kilowatt-hour of electricity produced over a 10-year period. Customerowned solar generation totals 4 MW. Although Austin’s energy innovation mandates are particularly ambitious, the tools that the city and its utility are using are nonetheless fairly representative of action being taken by other cities in the United States, albeit not always with the same vigor. But Austin—along with San Francisco and other cities leading the way to clean energy—strives to be a technology innovation leader. To this end, the Pecan Street Project was launched in 2008 when a group of public and private stakeholders came together to consider opportunities for Austin to build on its leadership in the semiconductor industry to become a national leader in clean energy.34 These stakeholders went on to form a public-business research consortium to enable the Pecan Street Project to pursue opportunities to reduce carbon emissions (Austin Icons Revisit MCC 2012), jump-start widespread renewable generation, start companies, create spin-offs, and create new jobs (Pecan Street Project 2010). The consortium’s members include the city of Austin, Austin Energy, the University of Texas’s Austin Technology Incubator and Clean Energy Incubator, the Chamber of Commerce, the Environmental Defense Fund,

33. As a first installment, it has contracted to purchase the annual output of a 30 MW solar farm, one of the largest in the United States. 34. The Texas Workforce Commission estimated that in 2009 green industries in Austin (including clean energy technologies) employed 45,672 people. In 2010, there were 175 clean energy technology companies, according to the Chamber of Commerce (Office of Sustainability 2012).

290

Katherine Sierra

and 11 private corporate technology partners.35 The project’s key goals are to nurture the development of Austin’s smart grid and to accelerate the introduction of distributed energy systems. In helping implement the Pecan Street Project, Austin Energy has as a first stage invested in the backbone of a smart grid telecommunications network and its associated hardware (computers and meters) and software applications, connecting 1 million consumers and 43,000 businesses to power plants and back. The project has championed the emergence of Smart Grid 2.0, which would manage distributed energy generation, build and manage energy storage, power and communicate with smart consumer appliances, and charge plug-in hybrid and electric vehicles. The project aims to deploy 300 MW of distributed energy by 2020 while using the creation of the Austin Energy smart grid as a “ ‘test lab’ for the many companies that will create the clean technology that the future system will require” (Pecan Street Project 2010, 19). Testing New Business Models The Pecan Street Project’s working group calls for testing new Austin Energy business models that shift the utility from a volume-driven commodity provider to one that provides services that use the smart grid to integrate energy efficiency and distributed generation. One possibility the group is considering is to pursue a fee-based business model for the provision of services. Customers would become “energy partners” with Austin Energy, and for a fixed fee would make their rooftops available for solar equipment and would agree to demand management practices, such as limiting use during peak hours (Pecan Street Project 2010, 16). A new business model, however, must consider the interaction between a traditional energy services business that is based on increasing sales volume and a new partnership service provision model that seeks to conserve energy and therefore decrease energy sales. In Austin, maintaining the utility’s financial integrity is critical not simply for its own sustainability but also because the utility is a major source of revenue for the city. The Quest for Affordability and Cost Competitiveness Although Austin Energy is a partner in the Pecan Street Project, its own strategy continues to stress the achievement of its renewable goals through purchasing commercial-scale renewable wind energy (Austin Energy 2012). The utility is concerned about the potential higher costs of distributed energy, though it is not clear whether it is taking into account benefits like the ability to dynamically respond to minuteby-minute demand throughout the day. The utility’s concern with costs and affordability mirrors the findings of a recent Brookings–Hoover Institution report (Banks et al. 2011) on distributed power systems in the United States, which

35. The corporate partners are Applied Materials, Cisco, Dell, Freescale Semiconductor, GE Energy, Gridpoint, IBM, Intel, Microsoft, Oracle, and Sematech.

sustainable infrastructure for urban growth

291

concluded that in most places the levelized costs of distributed energy systems are currently not competitive with those of central station fossil fuel generation. It did note that in some regions, distributed systems are cost competitive, including medium- and community-scale wind generation. It concluded that traditional economic analysis may not fully capture the range of benefits from decentralized systems, such as greater reliability and environmental benefits. The study recommended sustained investment in research and development, concluding that distributed power systems have “the potential to make a significant positive contribution to the US power system” (Banks et al. 2011, xi).

Conclusions Today, in the face of critical global environmental issues, leading cities are seizing opportunities to meet rising demands for infrastructure while capitalizing on new technologies and approaches to reducing resource use. International and national policy environments that provide incentives, such as the pricing of natural resources like fossil fuels and water to reflect externalities, are needed, as are local policies that promote efficiency. Capital costs of some alternative technologies remain high, and a number of barriers and risks must be overcome. Also needed to bridge this gap is financial support from market mechanisms or from national or international subsidies in the absence of appropriate price signals. But public finance is stretched, so policy makers are looking to innovative financing tools to use scarce public funds to leverage private capital. Cities are also forging new partnerships with the public sector, civil society, business, academia, and researchers that can help them accelerate the implementation of sustainability strategies by sharing their knowledge and by creating an environment for innovation in sustainable solutions. Examples of green growth approaches to support sustainable infrastructure development are found throughout the developed, emerging, and developing economies. Countries like Mexico are showing how they can use national approaches, accompanied by innovative finance, to scale up sustainable transportation to provide more efficient services, improve the quality of life through improved air quality, and increase access for the poor. Whether the momentum for innovation will survive Mexico’s change in presidential administrations bears watching. Proponents hope that the grounding of policy changes in legislation, the creation of a stable funding source, and support from a broad base of cities will allow the bus rapid transit program to endure. Cities like Jakarta are implementing urgent and difficult programs to build resiliency to a changing climate. They are trying to balance a focus on urgently needed infrastructure investments with strategies to strengthen ecosystems and social capacity while building new institutions. Partnerships are forming to improve data, share knowledge, and build capacity. The international community needs to support these efforts more vigorously with international climate finance for adaptation.

292

Katherine Sierra

Finally, cities like Austin are setting ambitious targets for reducing their energy footprints. They look to innovate by taking advantage of the proximity of technology centers and partnerships with businesses and researchers to capitalize on advances in information and communications technologies. In doing so, they hope to improve competitiveness and create jobs while also learning lessons that can help create a national energy future with a smaller carbon footprint. But the Austin case also suggests the need for stronger incentives by reflecting the cost of carbon emissions through either a carbon tax or a cap-and-trade scheme, smart grid development, and modifications to energy pricing or reliable but transitional national subsidies.

references Austin Energy. 2012. Quarterly briefing (26 January). www.austinenergy.com /aboutpercent20us/newsroom/Reports/QuarterlyReportJan2012.pdf. ———. 2010. Austin Energy resource, generation and climate protection plan to 2020 (22 April). Austin: Austin Energy. Austin icons revisit MCC. 2012. Pecan Street Inc. (8 March). www.pecanstreet.org /2012/03/austin-icons-revisit-mcc. Banks, J. P., J. Carl, K. Massy, P. Mokrian, J. Simjanovic, D. Slayton, A. G. Wagner, and L. V. Wood. 2011. Assessing the role of distributed power systems in the U.S. power sector. Washington, DC: Brookings Institution Energy Security Initiative and Hoover Institution Shultz-Stephenson Task Force on Energy Policy. Berlin, K., R. Hundt, M. Muro, and D. Saha. 2012. State clean energy finance banks: New investment facilities for clean energy deployment. Washington, DC: Brookings-Rockefeller Project on State and Metropolitan Innovation. Brown, J., and M. Jacobs. 2011. Leveraging private investment: The role of public sector climate finance. London: Overseas Development Institute. Carbon Disclosure Project. 2011. Cities 2011: Global report on C40 cities. Prepared by KPMG. London: Carbon Disclosure Project. Center for American Progress and the Global Climate Network. 2010. Leveraging private finance for clean energy. Washington, DC: Center for American Progress and Global Climate Network. Clapp, C., A. Leseur, O. Sartor, G. Briner, and J. Corfee-Morlot. 2010. Cities and carbon market finance: Taking stock of cities’ experience with clean development mechanism (CDM) and joint implementation (JI). OECD Environmental Working Paper No. 29. Paris: Organisation for Economic Co-operation and Development. Clean Technology Fund. 2009a. Investment plan for Egypt, January 29, 2009. Washington, DC: Clean Technology Fund. ———. 2009b. Investment plan for Mexico, January 16, 2009. Washington, DC: Clean Technology Fund. ———. 2009c. Investment plan for the Philippines, December 2009. Washington, DC: Clean Technology Fund. ———. 2009d. Investment plan for Vietnam, November 2009. Washington, DC: Clean Technology Fund.

sustainable infrastructure for urban growth

293

———. 2010a. Investment plan for Colombia, April 2010. Washington, DC: Clean Technology Fund. ———. 2010b. Investment plan for Nigeria, November 2010. Washington, DC: Clean Technology Fund. ———. 2010c. Investment plan for Vietnam supplemental note, June 2010. Washington, DC: Clean Technology Fund. ———. 2011. Update of the investment plan for the Philippines, December 2011. Washington, DC: Clean Technology Fund. Delta Alliance. N.d. Welcome to the Delta Alliance website. www.delta-alliance.org. Deng, T., and J. D. Nelson. 2011. Recent developments in bus rapid transit: A review of the literature. Transport Reviews 31(1):69–96. Environmental Audit Committee, House of Commons. 2011. The Green Investment Bank: Second report of session 2010–11. London: Stationery Office Limited. www.parliament.uk/business/committees/committees-a-z/commons-select /environmental-audit-committee/inquiries/green-investment-bank. Firman, T., I. Surbakti, I. Idroes, and H. Simarmata. 2011. Potential climate-change related vulnerabilities in Jakarta: Challenges and current status. Habitat International 35(2):372–378. Fuchs, R., M. Conran, and E. Louis. 2011. Climate change and Asia’s coastal urban cities: Can they meet the challenge? Environment and Urbanization Asia 2(1):13–28. Global Environment Facility. 2009. Investing in sustainable urban transit: The GEF experience, 2009. Washington, DC: Global Environment Facility. Hultman, N., K. Sierra, and G. Carlock. 2011. Energy and green growth: Recasting the options, re-envisioning sustainability. Washington, DC: Brookings Institution. International Energy Agency. 2011. World energy outlook 2011. Paris: International Energy Agency. Jakarta climate adaptation tools. N.d. Delta Alliance. www.delta-alliance.org/projects /jakarta-climate-change-adaptation-tools. Jenkins, J., M. Muro, T. M. Nordhaus, M. Shellenberger, L. Tawney, and A. Trembath. 2012. Beyond boom and bust: Putting clean tech on a path to subsidy independence. Washington, DC: Brookings Institution. Kamal-Chaoui, L., and A. Robert. 2009. Competitive cities and climate change. OECD Regional Development Working Papers 2009/2. Paris: Organisation for Economic Co-operation and Development, Public Governance and Territorial Development Directorate. Muro, M., J. Rothwell, and D. Saha. 2011. Sizing the clean economy: A national and regional green jobs assessment. Washington, DC: Brookings Institution. Nakhooda, S., A. Caravani, A. Wenzel, and L. Schalateck. 2011. The evolving climate finance architecture. Climate Fundamentals Briefing 2. London: Overseas Development Institute and Washington, DC: Heinrich Böll Stiftung North America. Office of Sustainability, City of Austin. 2012. Climate action annual report, 2010–11. Austin: Office of Sustainability. Organisation for Economic Co-operation and Development. 2007. Infrastructure to 2030, vol. 2: Mapping policy for electricity, water and transport. Paris: Organisation for Economic Co-operation and Development. ———. 2011a. Cities and climate change, 2011. Paris: Organisation for Economic Cooperation and Development.

294

Katherine Sierra

———. 2011b. Towards green growth: A summary for policy makers and tools for delivering green growth. Paris: Organisation for Economic Co-operation and Development. ———. 2012. The Chicago Proposal for Financing Sustainable Cities, OECD Roundtable of Mayors and Ministers, Mobilizing Investments for Urban Sustainability, Job Creation, and Resilient Growth, 8–9 March, Chicago. Pecan Street Project. 2010. Working group recommendations. Austin: Pecan Street Project. Rockefeller Foundation. N.d. Asian Cities Climate Change Resilience Network (ACCCRN). www.rockefellerfoundation.org/what-we-do/current-work/developing -climate-change-resilience/asian-cities-climate-change-resilience. Rule, H. 2012. Facebook app lets Austin Utilities customers compare energy use. Austin Post-Bulletin (4 April). http://postbulletin.com/news/stories/display .php?id=1492061. Sabarini, P. 2010. Administration open to feedback on spatial plan. Jakarta Post (12 January). www.thejakartapost.com/news/2010/01/12/administration-open -feedback-spatial-plan.html. Satterthwaite, D., P. Saleemul, H. Reid, and P. R. Lankao. 2007. Adapting to climate change in urban areas: The possibilities and constraints in low- and middle-income nations. International Institute for Environment and Development Human Settlements Discussion Paper Series. Sierra, K. 2011a. Adaptation to climate change in developing country urban deltas: Issues and approaches. Lincoln Institute Working Paper. Cambridge, MA: Lincoln Institute of Land Policy. ———. 2011b. The green climate fund: Options for mobilizing the private sector. London: Climate and Development Knowledge Network. Soehodo, S. 2011. Adaptation and vulnerability of Jakarta capital city. Presentation by deputy governor of Jakarta capital city at C40 São Paulo Summit, 30 May–3 June. http://c40citieslive.squarespace.com/storage/Jakarta%20Adaptation.pdf. Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Avery, M. Tignor, and H. L. Miller, eds. 2007. Climate change 2007: The physical science basis—Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. New York: Cambridge University Press. Sydney. 2011. State of the city 2011: Sustainable Sydney 2030. Sydney: Government of Sydney. Sykes, R., C. Chan, R. Encarnacion, and K. Richards. 2010. How should water utilities prepare for climate change? In Climate change and water: International perspectives on mitigation and adaptation, ed. C. Howe, J. B. Smith, and J. Henderson. London: IWA Publishing and Denver: American Water Works Association. Texier, P. 2008. Floods in Jakarta: When the extreme reveals daily structural constraints and mismanagement. Disaster Prevention and Management 17(3):358–372. United Nations. 1987. Our common future: Report of the United Nations World Commission on Environment and Development. New York: United Nations. ———. 2010. Work stream 7 paper: Public interventions to stimulate private investment in adaptation and mitigation. New York: United Nations High Level Advisory Group on Climate Finance. ———. 2012. World urbanization prospects: The 2011 revision. New York: United Nations.

sustainable infrastructure for urban growth

295

United Nations Environment Program. 2011. Cities: Investing in energy and resource efficiency. Nairobi: United Nations Environment Program. United Nations Habitat. 2011. Global report on human settlements 2011: Cities and climate change—Policies directions. Nairobi: UN Habitat. Wallis, M., M. Ambrose, and C. Chan. 2009. Climate change: Charting a water course in an uncertain future. In Climate change and water: International perspectives on mitigation and adaptation, ed. C. Howe, J. B. Smith, and J. Henderson. London: IWA Publishing and Denver: American Water Works Association. World Bank. 2010a. Cities and climate change: An urgent agenda. Washington, DC: World Bank. ———. 2010b. Climate risks and adaptation in Asian coastal megacities: A synthesis report. Washington, DC: World Bank, Asian Development Bank, and Japan International Cooperation Agency. ———. 2010c. Mexico urban transportation transformation program project appraisal document. www.wds.worldbank.org/external/default/WDSContentServer /WDSP/IB/2010/03/08/000333037_20100308232051/Rendered/PDF/515820 PAD0P107101Official0Use0Only1.pdf. ———. 2010d. World development report 2010: Development and climate change. Washington, DC: World Bank. ———. 2011a. The economics of adaptation to climate change: Synthesis report. Washington, DC: World Bank. ———. 2011b. State and trends of the carbon market: 2011. Washington, DC: World Bank. ———. 2011c. Indonesia: Jakarta urgent flood mitigation project (Jakarta Emergency Dredging Initiative) project appraisal document. 27 December. ———. 2012. Inclusive green growth: The pathway to sustainable development. Washington, DC: World Bank. ———. N.d. Climate change case study: Jakarta—Climate change, disaster risk management and the urban poor. Washington, DC: World Bank. http:// siteresources.worldbank.org/INTURBANDEVELOPMENT/Resources/336387 -1296405826983/Ellis-Lee.pdf.

11 Understanding Urban Infrastructure-Related Greenhouse Gas Emissions and Key Mitigation Strategies Anu Ramaswami

F

or the first time in human history, more than half of the world’s inhabitants are now living in cities and urban areas. Between 2007 and 2050, the world’s population will increase from 6.6 to 9.4 billion people, 70 percent of whom will be living in urban settings (United Nations 2007). Because cities house such a large proportion of the world’s people, they are major contributors to global greenhouse gas (GHG) emissions. The International Energy Agency estimates that more than 70 percent of global GHG emissions are attributed to energy use in cities (International Energy Agency 2008). Recognizing this fact, more than 1,200 cities worldwide have taken on the challenge of measuring and mitigating GHG emissions associated with their communities, including more than 1,000 cities that have signed on to the U.S. Mayor’s Climate Protection Agreement.

Cities and Global Greenhouse Gas Emissions MEASUREMENT CHALLENGES

Greenhouse gas accounting at the city scale is confounded by the relatively small spatial size of cities compared to nations, as a result of which two issues arise (Ramaswami et al. 2011). First, essential infrastructures that provide energy, water, wastewater treatment, commuter and airline transport, and other basic services to urban residents often cross city boundaries. Hence, the energy use to provide these infrastructure services often occurs outside the boundary of the cities 296

greenhouse gas emissions and key mitigation strategies

297

using them. This results in the phenomenon of energy being embodied in the transboundary provision of various infrastructure services to cities. For example, energy used to produce and pump water to cities over long distances would be termed the energy embodied in water supply. Beyond infrastructures, significant trade of other goods and services that likewise occurs across city boundaries is also associated with embodied energy and GHG emissions. Consequently, human activity in cities—occurring in residential, commercial, and industrial sectors—stimulates both direct in-boundary GHG emissions (occurring within the geopolitical boundary of the community) and transboundary emissions (occurring outside). Thus, strictly limiting the measurement of energy use and GHG emissions to those occurring within a city’s boundary can provide an incorrect and even misleading picture. In some cases, a purely geographic measurement approach may create unintended incentives to simply move GHG emissions outside the boundary. As we discuss new technologies, urban design strategies, and policy levers for low-carbon city development, it is important to develop robust methods for measuring GHG emissions associated with cities, addressing both in-boundary and transboundary emissions. INFRASTRUCTURE SUPPLY CHAIN GHG EMISSION FOOTPRINTS OF CITIES

Ramaswami et al. (2008), Hillman and Ramaswami (2010), and Chavez and Ramaswami (2012) have articulated a transboundary infrastructure footprint (TBIF) to explicitly incorporate the impact of in-boundary and transboundary infrastructures on the GHG emissions associated with cities. Infrastructures are defined broadly as those that provide basic needs of water, energy, food, mobility and connectivity, shelter (building materials), sanitation/waste disposal, and public spaces in cities, and are also important for economic activity in all cities (Chavez and Ramaswami 2012). The TBIF methodology combines urban material and energy flows associated with these key urban infrastructure services in a city with the life cycle energy needed to provide these services, obtained from life cycle assessment. The energy and materials use in a city associated with the key infrastructure sectors in a TBIF can be detailed as follows: U U

Energy use in buildings and facilities (residential, commercial, and industrial) and the provision of such energy by electric power plants and natural gas plants, which may be located outside the city boundary. Fuel use for transportation activities by the community (gasoline, diesel, and jet fuel). Because surface travel occurs regionally and air travel occurs nationally and globally, appropriate methods are needed to allocate road and air travel to communities based on the actual demand for transportation activities. — Increasingly, the literature is converging toward allocating surface

298

t

Anu Ramaswami

travel based on demand elicited from regional metropolitan transportation models, rather than merely counting all vehicle trips within the city boundary, which often include pass-through and highway trips that have no relationship to the city of interest. — Methods for allocating airline travel to communities are also emerging. A full allocation of airline trips to each community requires detailed data on whether the trip was initiated by residents, businesses, or visitors to the community and on the average distance of travel for each category of passenger. Such data are nearly impossible to obtain. Hence, a simpler estimate allocates jet fuel loaded at each airport (i.e., fuel needed for a one-way trip out of the airport) to surrounding cities using that airport. Such allocation can be done either using airline passenger surveys (see Chavez et al. 2012) or using regional surface transport models that provide data on the proportion of all road trips to and from the airport that arise from the surrounding cities, separating out employee trips (see Hillman, Janson, and Ramaswami 2011; Ramaswami et al. 2008). The embodied energy of various materials needed to support key infrastructure supply chains to the city: — The production and refining of transportation fuels to support the transportation activities allocated to cities, as described above. — Energy use to provide water supply, wastewater treatment, and waste management services to the community, even if such energy use occurs outside the city boundary. — Embodied energy to produce key urban built environment materials, such as cement, iron, and steel. — Additional energy embodied in long-distance freight transport to and from the community. — Energy embodied in the production of food to support the community’s residential and commercial sectors.

While the provision of food is not traditionally considered infrastructure, a food supply is essential to life in cities, just like the provision of water, wastewater, energy, and transportation services. Community-wide use of food includes food for the residents’ personal consumption and food for the hospitality/restaurant industry that serves visitors to the community. New methods are emerging that track not only food consumed at home using consumer expenditure surveys (as in Ramaswami et al. 2008), but also food sales data in restaurants that can help track food services to residents and visitors (Chavez and Ramaswami 2012). All these infrastructure services support the entire community—that is, local homes, businesses, and industries as well as visitors, and provide a productionbased view of the city. Indeed, the impact of visitors—can be important both in small resort cities like the ski resort towns of Vail and Aspen, Colorado, and in large tourist destinations like London, New York, and Paris.

greenhouse gas emissions and key mitigation strategies

299

The TBIF method accounts for key infrastructures serving the city as a whole and is thus analogous to national production-based GHG accounting with the challenge of infrastructure fragmentation by city boundaries being overcome by the transboundary approach. The production-based TBIF accounting method has been related mathematically to consumption-based GHG accounting that incorporates the full supply chains of all goods and services used by resident households and final economic consumption sectors in a city, including global trade (see Chavez and Ramaswami 2012). The consumption account, however, excludes energy used by local businesses that serve visitors or that produce goods and services that are exported. A dual approach is therefore recommended with TBIF to support city infrastructure planning and a separate consumption-based accounting to inform household consumption behaviors. This dual approach has now been adopted in city-scale GHG reporting protocols developed recently by ICLEI-USA and also by the British Standards Institute (PAS 2070; BSI 2012). These protocols enable standardization and institutionalization of the TBIF method described in this chapter. AVOIDING DOUBLE COUNTING IN TBIF

The TBIF for Denver, Colorado, which incorporates all the infrastructure components discussed here, is shown in figure 11.1. Care must be taken in adding the various infrastructure sectors to avoid double counting GHG emissions. (Appropriate methods are described in greater detail in the referenced manuscripts [see Hillman and Ramaswami 2010; Ramaswami et al. 2008].) Future work may also try to separate out the energy use in the buildings energy sector for providing information-communication services, and to track the supply chains that support this infrastructure. At present, however, the emphasis is on tracking the key infrastructure services shown in figure 11.1—the provision of energy, mobility/transportation, water supply, waste treatment, food, fuel, and building materials (e.g., cement) in cities. Indeed, the community-scale GHG accounting protocol of the Local Governments for Sustainability now includes consideration of all the above infrastructure sectors (ICLEI-USA 2012). Some of the services, such as water supply, wastewater treatment, and the embodied energy of construction materials such as cement, contribute less than 2 percent of the total infrastructure footprint (see figure 11.1). Potential future additions to infrastructure sectors in the TBIF, such as the supply chains supporting the information-communication sector, can be compared against these numbers to judge their relative importance. CONVERGENCE

Developed for Denver in 2008, the TBIF method has since been tested in eight U.S. cities and shows good convergence between the transboundary GHG emission footprints computed at the city scale versus national benchmarks of 25 (mt [metric tons]-CO2e) per capita per year, suggesting that the transboundary challenge in addressing city-scale GHGs may have been overcome by these transboundary infrastructure supply chain inclusions (see Hillman and Ramaswami

300

Anu Ramaswami

Figure 11.1 Transboundary Infrastructure GHG Emissions Footprint (TBIF) for Denver

Cement 2%

Water/WW 0.4% Long-distance Waste freight 0.7% 2% Food 13%

Fuel processing 6%

Building electricity (residential, commercial, industrial) 36%

Air travel 6% Commercial vehicles 3% Trucks and SUVs 10% Transboundary infrastructure GHGs

Cars 6%

Residential natural gas 5%

Commercial/industrial natural gas 10%

Note: Embodied energy and transboundary GHG emissions are hatched. Source: Adapted from Ramaswami et al. (2008).

2010). Appropriate metrics to represent the resulting GHG emissions are currently being studied. It is proposed that the TBIF GHG emissions be expressed per unit productivity of the community (based on gross domestic product [GDP]), Figure 11.1 while consumption-based GHG associated with homes be represented per resiLincoln_Ingram_Infrastructure dent (per capita). A meta-analysis across 40 U.S. counties confirms that TBIF GHG emissions expressed per unit GDP track well with measures of local urban efficiency, while consumption-based GHG emissions per capita track most closely with household expenditures (Chavez 2012). POLICY RELEVANCE AND CONTRIBUTION OF INFRASTRUCTURE SECTORS

The TBIF represents together all the key infrastructure sectors and their supply chains, which support the activities of business, industry, and households in a community. As a result, cross-infrastructure substitutions to reduce GHG emis-

greenhouse gas emissions and key mitigation strategies

301

sions are made visible—e.g., the substitution of airline travel in the transportation sector with the typically lower energy usage for teleconferencing in the buildings sector. Because supply chains of infrastructure are incorporated in the TBIF, the method is also effective in informing cross-scale strategies for GHG mitigation (e.g., by greening the supply chain of concrete through suitable substitutes for cement). The TBIF also supports effective, sustainable urban planning that addresses infrastructure needs to support the future growth of the city as a whole—its homes, visitors, businesses, and industries considered together. Using the information in the TBIF, planners and policy makers can examine the different infrastructure sectors that contribute to GHG emissions. While cities differ vastly in how each sector contributes to GHG emissions, in general a study of eight U.S. cities shows that large cities tend to exhibit these general characteristics (Hillman and Ramaswami 2010): t t

t

t

Energy use in buildings and facilities contributes a large proportion (greater than 40 percent) of TBIF GHG emissions. As our electricity supply includes more renewables, this contribution is likely to decrease. Transportation sector emissions follow closely (between 30 and 40 percent of the overall TBIF emissions) when all modes of travel are included and fuel refining is incorporated. Emission reductions in this sector are most challenging and likely to decrease slowly with smart growth and transitoriented policies and with the penetration of new vehicle-fuel technologies. The GHGs embodied in food use in the community are significant and exceed 10 percent of the TBIF GHG emissions. Cities can lower these emissions by altering the nature of demand for food in communities—for example, by encouraging a healthy diet with less red meat. Other infrastructure sectors such as water, wastewater treatment, waste disposal, and the GHG embodied in urban materials, together, total between 5 and 10 percent, depending on the characteristics of different cities. Cities in the developing world have higher GHG contributions from these sectors due to the release of untreated wastewater as well as a high level of construction activities using cement (e.g., see the discussion of GHG emissions in Delhi in Chavez et al. 2012).

The TBIF thus presents a unique view of cities connecting community-wide activities that use infrastructure services with the life cycle energy and materials needed to provide these services. Strategies that reduce or alter the use of energy, water, infrastructure-related materials, and food in cities, as well as those that encourage the cleaner production of these material-energy flows or that promote symbiotic exchanges between municipal and industry sectors (e.g., Van Berkel 2010), can reduce the transboundary infrastructure GHG footprint of cities.

302

Anu Ramaswami

Buildings and Land Use Patterns Buildings and land use patterns significantly affect the use of energy and materials in urban infrastructure. Energy use in buildings is shown to be highly correlated first with weather, represented by heating and cooling degree days, i.e., the number of days in a year and the extent to which these days exceed a comfortable ambient temperature of 65 degrees Fahrenheit. Cities in cold regions will have a large number for heating degree days, while those in very hot regions have large cooling degree days. Energy use in buildings—which is needed for both heating and cooling—is shown to be strongly correlated with the heating and cooling degree days representing the regional climate. Indeed, the average energy use in homes in different U.S. cities is observed to be close to the state average reported in the residential energy consumption surveys, and representative of climate conditions (Hillman and Ramaswami 2010). When comparing cities, building occupants and the square footage of homes are also emerging as an important factor, along with the age of homes and the incomes of the home dwellers. Improved data sets are emerging that help uncover these various influences. However, there are many confounding studies on the effect of densification on energy use in buildings. While, in general, more compact buildings with shared walls reduce both direct and indirect energy use (embodied in materials), a decreasing trend in building occupancy levels in more compact buildings can offset these gains. When the income levels and consumption behaviors of occupants are further considered, the results can be mixed. Energy use in transportation has shown different types of correlations with population density. In a study of global cities (Newman and Kenworthy 1999), this relationship follows the classic exponentially decreasing curve when comparing global cities with large orders of magnitude differences in population density. The cities included Denver, Los Angeles, and New York City (in the United States) versus Geneva, Bangkok, and other international cities. However, it should be emphasized that for several cities, such as New York City, the contribution of mass transit and rail was not included. In a study of road travel GHG emissions across 40 U.S. cities, per capita surface transportation emissions showed little or no correlation with population density; the population density across U.S. cities does not range as widely as across global cities. Indeed, when regional commuter travel was apportioned based on origin-destination allocation, increases in activity density (i.e., the sum of homes plus businesses per acre) correlated with an increase in per capita transportation sector emissions, reflecting the importance of workplaces and businesses in generating travel demand (Hillman et al. 2011). For five of the eight larger U.S. cities studied—Boulder, Denver, Portland, Seattle, and Austin—the average per capita vehicle miles traveled (VMT) was very similar and hovered around 22 to 25 VMT per person per day, consistent with national benchmarks (Hillman and Ramaswami 2010). For these cities, despite differences in bicycle or transit ridership in the central city, regional patterns of travel changed very little. Since transportation

greenhouse gas emissions and key mitigation strategies

303

activities are generally regional, the impact of small local changes (e.g., bicycling or walking) is not very apparent at the scale of regional transport. These examples illustrate that the relationships between land use and travel behaviors are complex (National Research Council 2009). Organizations like the Federal Highway Administration have posited a threshold density of 7 to 10 dwelling units per acre, above which increases in density start yielding decreases in travel demand, with transit, car sharing, and other alternative modes becoming viable. A vast majority of U.S. cities and counties exhibit population densities well below this threshold, with exceptions such as New York City. The relationship between travel demand and land use density is made more complex by many different phenomena, including how travel demand is measured and potential self-selection biases for reduced motorized travel among urban versus suburban residents. Furthermore, as the well-known adage notes, correlation does not imply causality. Thus, strategies that only focus on densification rarely succeed in reducing per capita VMT. In addition to density, other variables such as diversity, design, access to transit, and regional access to jobs have significant impacts (National Research Council 2009). The elasticity of travel demand with respect to each of these variables, independently, is quite small; however, in the long term, a doubling of density with improvements in design, diversity, and distance to transit and regional accessibility to jobs is estimated to yield about a 25 percent reduction in per capita VMT in U.S. cities (National Research Council 2009). This does not mean, however, that land use change and improvements in building energy efficiency are not effective. The relative inelastic nature of travel demand with respect to land use variables merely indicates that there is significant momentum (inertia) and path dependency in urban systems. Thus, changing current GHG emission trajectories in the transportation sector will take a long time. Also needed are innovative community programs supporting behavioral change and strategic policy interventions, which are described next.

Key Mitigation Strategies Ramaswami et al. (2012) conducted a first-order analysis of key mitigation strategies in the buildings and transportation sector in typical U.S. cities, addressing the potential for near-term GHG mitigation over four to five years. Greenhouse gas mitigation is computed against a backdrop of an increasing trend in per capita electricity use as well as increased population, and key strategies are explored that will “bend the curve” of increasing community-wide TBIF-GHG emissions in the next five years. Strategies were organized according to the primary agency among three broad categories of social actors: individual users (e.g., individual homes and businesses), infrastructure designer-operators, and policy actors (government officials, lawmakers, nongovernmental actors, media, scientists, and others involved

304

Anu Ramaswami

Figure 11.2 Typical Strategies for GHG Mitigation Initiated by the Three Actor Categories

Voluntary programs: Rebates, incentives, awards

Infrastructure users (U)

Sustainable consumption Urban metabolism and GHG emission footprints

Feedback: Price, bills, real-time

Policy actors (P)

Regulatory programs: Building codes, ordinances, carbon tax, opt-in bond programs, renewable portfolio standards

Infrastructure designeroperators (D)

Sustainable production

Source: Ramaswami et al. (2012). Reprinted by permission of Environmental Science and Technology. Copyright © 2012 American Chemical Society.

in the policy process). Typical strategies for GHG footprint mitigation are shown in figure 11.2 and include the following (Ramaswami et al. 2012): .

t

t t

Figure 11.2 Voluntary adoption of energy conservation behaviors and efficiency measures by individual users (U), often incentivized Lincoln_Ingram_Infrastructure by rebates, awards, and other incentives provided by policy actors at local, state, or federal government agencies (U, P). These measures may include both energy conservation and efficiency in homes, as well as efforts to change travel behaviors. Voluntary actions among infrastructure designer-operators, such as adoption of green building practices, increased use of renewables in the electric grid, and adoption of higher-efficiency vehicles in bus fleet upgrades (D). Regulatory approaches are defined as those that need a mayoral decree or a vote by city council or other legislative bodies. Regulatory approaches institutionalize voluntary strategies aimed at sustainable consumption or sustainable production. Not all regulatory approaches have to be mandates. The following are examples of innovative regulations developed at the city scale: — Time-of-sale ordinances. Residential and commercial energy conservation ordinances in effect in Berkeley and San Francisco since the 1980s

greenhouse gas emissions and key mitigation strategies

305

require that all homes and commercial properties be renovated to basic energy efficiency standards at the time of sale (e.g., minimum Department of Energy–specified attic insulation, weather stripping, pipe insulation around hot and cold water pipes). — Climate Smart Loan Program. The city of Boulder’s Climate Smart Loan Program provides loans for more expensive energy efficiency improvements (e.g., windows, solar panels, solar hot water heaters). The loan is repaid through special tax assessments associated with that specific property even if it is subsequently sold. This overcomes one of the main barriers to investing in large home energy projects with payback periods longer than seven years, the average period of home ownership in the United States. — Date-certain and Smart Regulations. Date-certain regulations require that buildings be retrofitted to basic energy efficiency standards by a fixed date. Boulder’s Smart Regulations (as referred to in Boulder) require that all rental property be upgraded by 2014 to basic energy efficiency standards; properties will be reviewed at the time of renewal of rental licenses. — Behavioral feedback. Cities and utilities are also experimenting with different forms of energy feedback devices. A few utilities have implemented price feedback via monthly energy bills; this strategy is showing energy savings of 2 percent on average across the community (O-Power). Instantaneous behavioral feedback can be achieved using real-time energy meters that display energy use continuously. These meters have been shown to stimulate 6 to 15 percent savings in electricity use in pilot studies. A few utilities are contemplating requiring such low-cost devices in all homes, such as SoCal Edison in California. In addition to the city-scale regulations described above, several state-scale regulations can also reduce GHG emissions from buildings’ energy use. For example, Colorado’s Renewable Portfolio Standard currently requires 30 percent of the state’s electricity-generation portfolio to consist of renewable resources by the year 2020. The Clean Air, Clean Jobs Bill (HB 1361) requires that aging coal plants be phased out and replaced with cleaner-burning natural gas power plants that emit about half the CO2 for generating electricity. The federal renewable fuels policy shapes the penetration of biofuels in the fuel mix.

Strategy Analysis and Policy Implications The quantitative study of Ramaswami et al. (2012) reveals the following key findings comparing the effectiveness of voluntary, market-based, and regulatory strategies for GHG mitigation in the buildings and transportation sectors.

306

Anu Ramaswami

REDUCING GHG EMISSIONS IN THE BUILDINGS SECTOR

To reduce GHG emissions related to energy use in the buildings sector, three broad types of strategies are commonly suggested and adopted by many cities: U

U

U

Voluntary upgrades for home energy efficiency. Voluntary participation rates of home dwellers in typical home efficiency upgrade programs offered by cities are quite low, decreasing from a high of 50 percent for free giveaways of CFLs to less than 4 percent for modestly higher-cost items like attic insulation. Typical voluntary retrofit programs that hand out free CFLs or employ traditional random door-to-door neighborhood campaigns appear to be ineffective. Given the low participation rates, it is proposed that cities either explore innovative new ways of engaging the community in energy conservation or consider strategic city-scale regulations, such as the date-certain and time-of-sale ordinances that have been successful in some pioneering cities, as described in the previous section. Voluntary changes in green energy purchases. More consumers are making green energy purchases, not only in Denver, but throughout the United States. One electric utility reported 15 percent green electricity purchases. A stronger marketing campaign in this area may have significant impact in most U.S. cities, where green energy purchases are often less than 1 percent of the total use of electricity. Behavioral change with feedback devices. Instantaneous feedback devices (real-time energy information displays) and even monthly feedback via energy bills can have a significant impact on behavior, reducing electricity use by as much as 15 percent in numerous pilot studies. Behavioral change with feedback can be strategically combined with rebate programs that promote efficiency measures to yield higher-impact “hybrid” program designs. Mandates that require all homes or all new homes to have low-cost energy feedback devices can significantly increase the penetration of the devices for maximum impact.

Only a handful of cities are exploring regulatory strategies for GHG mitigation at the city scale, such as the time-of-sale ordinances and rental property regulations described in the previous section. A quantitative assessment of the impact of the different types of building sector GHG mitigation strategies in figure 11.3 shows that cities must have a strategic portfolio mix of voluntary programs along with a select few high-impact key regulatory or policy strategies. Both city-scale regulations that shape demand (see wedge 4) as well as state-scale regulations that shape energy supply, such as a renewable electricity portfolio (wedge 5), can have significant impact on GHG mitigation, as shown in the figure. Without regulatory strategies, little GHG mitigation will be realized, while a robust mix of different strategies can be effective in significantly reducing GHG emissions.

greenhouse gas emissions and key mitigation strategies

307

Figure 11.3 Relative Impacts of Key Pathways for Near-Term GHG Mitigation for Cities 7,700,000

Typical city programs: voluntary home retrofits Behavioral feedback

7,500,000 7,300,000

hases

BAU

mt-CO 2 e

7,100,000

n energy purc Voluntary gree City regulations

State programs: Renewable portfolio standard (RPS), demand-side management (DSM), and shifts to natural gas

6,900,000

1 2 3 4 5

6,700,000 6,500,000 6,300,000

1990 level

6,100,000 2007

2012

Source: Adapted from Ramaswami et al. (2012).

REDUCING GHG EMISSIONS THE TRANSPORTATION SECTOR FigureIN11.3

Mitigating GHG emissions from transportation activities is more challenging in Lincoln_Ingram_Infrastructure the short term and is likely to remain so in the longer term. The impact of new vehicle-fuel technologies has been evaluated by many researchers (e.g., see Argonne National Lab’s GREET1 model). A life cycle–based wells-to-wheels perspective— that includes improvement in not only tail-pipe emissions, but also energy expended in fuel production—is essential to identify system-wide reductions in GHG emissions. While some new vehicle-fuel technologies are promising (e.g., plug-in hybrids using renewable electricity), their production and penetration rates in society are still so small that they will have little impact in the near term on GHG emissions from the transportation sector.

1. Greenhouse gases, Regulated Emissions, and Energy use in Transportation.

308

Anu Ramaswami

Table 11.1 Impact of Different Strategies on Near-Term Transport Sector GHG Mitigation Over Five Years Strategy Type

Description and Assumptions

Voluntary changes in vehicles or travel modes among individuals/businesses

50% fleet upgrades for police, bus, and school fleets (only 2% VMT impacted) Offer community-wide individualized travel marketing program, assuming 10% Denver residents working outside Denver participate Double the number of Denver employees in employer-based commuter programs (carpool, vanpool, telecommuting, bus pass, etc.) Quadruple bike travel (Denver Bike Share Program)

Voluntary changes in travel services demanded, using innovative technologies/markets

Adoption of telepresence among 3.3% of air travelers Airline offsets purchased by 5% of air travelers

Regulatory or policy strategies

Smart growth planning: 75% of new population in Denver (or its equivalent) living in higher-density (double-density) areas, with all other favorable factors (access, design, diversity, distance to transit) Low-rolling-resistance tires mandated for all; tire change assumed every 5–6 years Pay-as-you-drive (PAYD) auto insurance offered community-wide, equivalent to a 40% gasoline tax

Percent (%) Reduction in Transport GHG (est. over 5 years) 0.2

0.6

0.6 0.01 0.7 1

0.9 1.8 2.3

GHG ! greenhouse gas; VMT ! vehicle miles traveled Source: Ramaswami et al. (2012).

For the near-term analysis, the following broad strategy types had the most impact, all of which were very small. (See details and computations provided in Ramaswami et al. 2012.) Cities are trying several voluntary strategies, such as those listed below: U

U

Fleet upgrades. School buses, transit buses, and government fleets contribute only 1.2 percent of VMT in the Denver region, and hence even when 50 percent of the vehicles are upgraded, the impact is small (less than a 1 percent reduction in GHGs), as shown in table 11.1. Employer-based commuter programs. The impact of employer-based incentives has been quantified in many U.S. metro area case studies, with

greenhouse gas emissions and key mitigation strategies

U

U

309

a reported average savings of 0.5 mt-CO2e annually per employee participating in the National Best Workplace for Commuters Program. A similar range of GHG mitigation (0.5 to 1 mt-CO2e per commuter) emerges in detailed studies of the Denver Regional Council of Governments’ RideArrangers program, in which 1,429 employers offered car pools, van pools, transit, and telecommuting during the 2009 program launch year. Doubling the number of employees participating in employer-based programs by 2012, which experts considered to be an aggressive goal, yielded 60,000 additional participants over five years and was estimated to mitigate 0.6 percent of the transportation-related GHG emissions compared to the base year. Individualized travel marketing (ITM). Additional mode shifts toward nonautomobile travel (e.g., transit) can be promoted through ITM programs that provide personalized information about existing mass transit routes and safe bike paths to promote switching to these alternative modes. Results from such interventions designed to change travel behavior are often in the gray literature and need to be verified. A recent review reports VMT reductions of 2 to 12 percent as a result of ITM programs across U.S. cities (Dill and Mohr 2010). Such a program implemented in Denver would target those who work outside the city (since Denver workers are covered in the RideArrangers program). If we assume 10 percent of these workers will respond, the GHG mitigation is of the order of only 0.6 percent in the best-case scenario. Quantifying the impact of shifts toward transit can be difficult. The above represents a best-case scenario where increased transit ridership is achieved with existing levels of service. In practice, the efficiency of transit depends on the ability to achieve a high loading of people in buses and trains, which in turn depends on population density and other factors, such as the state of the economy. Under current ridership conditions, energy use per person per mile traveled by bus is only slightly better than by automobile (BTS 2006). Shared automobile trips may yield a larger improvement than transit ridership. Furthermore, life cycle analysis can reveal useful insights about transit. For example, savings in transportation energy in an elevated bus rapid transit system in Xiamen City, China, were partially offset by the increase in energy needed to operate the elevators and other building-related aspects of the system (Cui et al. 2011). Bicycle programs. Bike share and other popular programs may also yield much less GHG mitigation than commonly understood. Indeed, studies of bicycle travel from several U.S. cities report average bicycle trip distances of about two miles (Krizek, Handy, and Piatkowski 2010). Returning to the Denver case study, if bicycle mode share in the city increases from 2 percent in 2009 to 10 percent in 2018, near the potential maximum observed in U.S. cities, it would result in a quadrupling of the current 100,000 annual bike trips logged in Denver. However, the resulting automobile displacement is computed to be very small, at about 1 million VMT

310

Anu Ramaswami

annually, compared to total motorized VMT exceeding 5 billion per year (Ramaswami et al. 2008). This is a combination of relatively short bicycle trips (2 miles on average) and the fact that only 22 to 66 percent of bicycle trips displaced automobile trips. Thus, the impact of bike share is very small (less than 0.01 percent), as seen in table 11.1. Thus, not all popular voluntary programs have high GHG impact, although employer-based and individual travel marketing may hold some promise. In contrast, as shown in table 11.1, innovative market-based strategies that model only a small percentage of airline travelers replacing air travel with teleconferencing or purchasing travel offsets can have the same magnitude of impact. In terms of policies, smart growth planning and innovative state-scale policies such as pay-as-you-drive insurance can also have a larger impact than the purely voluntary programs—the latter because it impacts all vehicles simultaneously. The analysis summarized in table 11.1 shows that rates of adoption of new technologies and policies are very important in shaping GHG mitigation in the transportation sector. A combination of technological change (vehicle-fuel technologies as well as information technologies that promote telecommuting and teleconferencing) along with land use policies are needed to reduce transportation sector GHG emissions in the long term. Careful field studies and life cycle analyses are needed to ensure that GHG savings in the transportation sector are not lost by related increases in energy use in other sectors (e.g., in providing telework, teleconference, or transit services).

Conclusions Based on the results shown in figure 11.3 and table 11.1, this chapter recommends that cities develop and analyze their transboundary infrastructure GHG emission footprints and conduct a simple strategy analysis addressing a key variable: the participation rates of people expected in various programs. Based on such data and with a portfolio mix of voluntary and regulatory strategies, cities can achieve significant reductions in GHG emissions in the buildings sector, while only small GHG reductions can be expected in the transportation sector. In general, this study shows that cities could target, in the near term, at best a 1 percent reduction per year in GHG emissions associated with buildings and transportation and will find this to be a challenging but achievable goal.

Afterword The TBIF method is now being standardized for use by a large number of cities. It is represented as basic plus community-wide supply chains in ICLEI-USA’s recently released protocol for reporting community-scale GHG emissions (ICLEI-USA 2012), available to more than 600 U.S. cities. The TBIF methodol-

greenhouse gas emissions and key mitigation strategies

311

ogy—represented as Direct-Plus-Supply Chain—is also incorporated in a publicly available standard (PAS 2070) for GHG accounting for cities being developed by the British Standards Institute.

references British Standards Institute (BSI). 2012. Publicly Available Standard (PAS) 2070. https://ecommittees.bsi-global.com/bsi/controller?livelinkDataID=51931145. Bureau of Transportation Statistics (BTS). 2006. National Transportation Statistics 2006, Table 1-32. Bureau of Transportation Statistics. http://apps.bts.gov /publications/national_transportation_statistics/2010/html/table_01_32.html. Chavez, A. 2012. Comparing city-scale greenhouse gas (GHG) emission accounting methods: Implementation, approximations, and policy relevance. Ph.D. diss., University of Colorado at Denver. Chavez, A., and A. Ramaswami. 2012. Articulating a transboundary infrastructure supply chain greenhouse gas emission footprint for cities: Mathematical relationships and policy relevance. Energy Policy. http://dx.doi.org/10.1016/j.enpol.2012.10.037. Chavez, A., A. Ramaswami, N. Dwarakanath, R. Ranjan, and E. Kumar. 2012. Implementing transboundary infrastructure-based greenhouse gas accounting for Delhi, India: Data availability and methods. Journal of Industrial Ecology 16(6):814–828. Cui, S., F. Meng, W. Wang, and J. Lin. 2011. GHG accounting for public transport in Xiamen City, China. Carbon Management 2(4):383–395. Dill, J., and C. Mohr. 2010. Long-term evaluation of individualized marketing programs for travel demand management. Portland State University, July. Hillman, T., B. Janson, and A. Ramaswami. 2011. Spatial allocation of transportation greenhouse gas emissions at the city scale. ASCE Journal of Transportation Engineering 137(6):416–425. Hillman, T., and A. Ramaswami. 2010. Greenhouse gas emission footprints and energy use metrics for eight US cities. Environmental Science and Technology 44(6):1902–1910. ICLEI-USA. 2012. U.S. community protocol for accounting and reporting of greenhouse gas emissions. Oakland, CA: ICLEI–Local Governments for Sustainability USA. www.icleiusa.org/tools/ghg-protocol/community-protocol/us-community-protocol -for-accounting-and-reporting-of-greenhouse-gas-emissions. International Energy Agency. 2008. World energy outlook. Paris: International Energy Agency. Krizek, K., S. Handy, and D. Piatkowski. 2010. Methods to analyze the substitution effects of walking and cycling and reductions in carbon dioxide. Paper presented at the annual conference of the Association of Collegiate Schools of Planning, Minneapolis (7–10 October). National Research Council. 2009. Driving and the built environment: The effects of compact development on motorized travel, energy use, and CO2 emissions. Special Report 298. Washington, DC: Transportation Research Board. Newman, P., and J. Kenworthy. 1999. Sustainability and cities: Overcoming automobile dependence. Washington, DC: Island Press. Ramaswami, A., M. Bernard, A. Chavez, T. Hillman, M. Whitaker, G. Thomas, and M. Marshall. 2012. Quantifying carbon mitigation wedges in US cities: Near-term

312

Anu Ramaswami

strategy analysis and critical review. Environmental Science and Technology 46(7): 3629–3642. Ramaswami, A., A. Chavez, E. Ewing-Thiel, and K. Reeve. 2011. Two approaches to greenhouse gas emissions accounting at the city-scale. Environmental Science and Technology 45(10):4205–4206. Ramaswami, A., T. Hillman, B. Janson, M. Reiner, and G. Thomas. 2008. A demandcentered hybrid life cycle methodology for city-scale greenhouse gas inventories. Environmental Science and Technology 42(17):6456–6461. United Nations. 2007. World urbanization prospects: The 2007 revision. New York: United Nations. Van Berkel, R. 2010. Quantifying sustainability benefits of industrial symbioses. Journal of Industrial Ecology 14(3):371–373.

commentary W. Ross Morrow Anu Ramaswami’s chapter, and the broader literature associated with it, concerns independent action by city authorities to reduce greenhouse gas (GHG) emissions associated with activities in the city. Cities are, and will continue to be, a key driver both as an immediate source of emissions and as a source of demand for goods and services that cause GHG emissions. However, emissions due to activities within a city boundary do not necessarily occur within that boundary or within any particular metro area boundary. These emissions must nonetheless be included if the city’s independent policy action is to result in reductions in GHG emissions. The issue of transboundary emissions is essentially not mathematical, but rather sociopolitical. As Ramaswami states, “A purely geographic measurement approach may create unintended incentives to simply move GHG emissions outside the boundary.” That is, what is important is the independent reactions of businesses and households to policy, and these actions are influenced by the ways in which behaviors are measured, regulated, and incentivized. Appropriate measurement of city GHG emissions for policy application has a “self-similarity” property familiar to the complex systems literature: measurement of a city’s emissions in the context of state emissions is similar to the measurement of state emissions within the nation, or national emissions in the context of global emissions. Here is an analogy directly related to the work that Ramaswami describes: the United States currently measures its own GHG emissions as a function of geographic boundaries; it does not include emissions occurring in other parts of the world associated with the demand for goods and services consumed within the United States. The remainder of this commentary addresses in more detail the two key themes in Ramaswami’s chapter: measurement and policy.

Measurement For the last few years, Ramaswami and her colleagues have successfully introduced and applied concepts for activity-based measurement of emissions associated with essential infrastructure services on which cities rely, but that may create emissions outside a city’s boundary. The central figure is the transboundary infrastructure footprint (TBIF), which allocates to the city the emissions associated with services located outside its boundaries that are essential to the city’s functioning. Not including emissions created outside city boundaries for activities within city boundaries certainly risks undercounting GHG emissions in the absence of a national policy that covers all aspects of the economy regardless of location. The risk presented by including such activities in estimates is just the opposite: 313

314

W. Ross Morrow

overcounting. Ramaswami and her colleagues have considered this issue in detail in references given in her chapter. Broadly speaking, it seems plausible that the activity-based allocation of GHG emissions (or other environmental stressors) to specific cities is feasible, reasonably practical, and important for effective regulation of GHG emissions by largely noncooperative entities like U.S. cities. To be effective within a noncooperative network, city-based policy approaches must consider the effects of actions taken by other cities. Using TBIF or another activity-based metric as a measure for motivating and measuring city policies to reduce GHG emissions is maximally beneficial only if the specific techniques for allocating activities from different geographical areas, city or otherwise, are standardized across cities implementing measurement. Speaking in purely mathematical terms, it is possible for collections of cities applying heterogeneous measurement techniques to result in either over- or undercounting of emissions. It would seem, however, that as long as rural1 emissions are significant sources of emissions associated with cities, as suggested in the research by Ramaswami and her colleagues, then emissions are undercounted if any city uses a geographic basis for measurement. Established institutions like the U.S. Conference of Mayors’ Climate Protection Center,2 which currently has 1,054 mayors joined under a single climate protection agreement, offer an opportunity to catalyze such standardization in measurement methodology. The agreement primarily identifies specific practices that should be prioritized or promoted, several of which Ramaswami identifies as being rather low-impact paths toward emissions reductions. Moreover, the current agreement only mentions a single optional inventory of city emissions, with no specific methodology required and no requirement to continually track emissions reductions toward goals defined by this inventory. Clear, standardized measurement using activity-based principles should also play a role in state and national policies that concern activities associated with cities. While cities certainly have a significant amount of local authority, specific policy actions will also overlap with state and national policies motivated to reduce GHG emissions. Current policies at higher levels of government are limited, relative to the scale of the climate change problem. In fact, under the Climate Protection Agreement, U.S. mayors agreed to urge state and federal governments to enact broad policies to reduce GHG emissions. When strong policies arrive, however, actions to reduce emissions over all scales in the hierarchical regulatory network that exists in the United States must be considered as a whole to ensure that social, economic, and environmental interests are balanced. Researchers in the physical sciences and engineering are becoming increasingly interested in such “multiscale” phenomena in which microscopic details of physical systems 1. Here, “rural” simply means locations not within the geographic boundaries of a city implementing measurement. This may, in fact, include rather developed areas. 2. The Climate Protection Center’s website is www.usmayors.org/climateprotection/revised.

commentary

315

are important for resolving macroscopic properties. One analysis paradigm in engineering, analytical target cascading, uses a hierarchical enterprise paradigm to enable multidisciplinary system optimization (Cooper et al. 2006). While it is unlikely that mathematical optimization technology can literally be applied to inform complex policy design at the national, state, and city levels, scientific insights from these domains may offer compelling metaphors for qualitative investigation of the balance required for effective overlapping policy actions at the national, state, and city levels. One technical concern about the language that Ramaswami uses lies in the concept of scale conversion. Here and elsewhere, she claims that the “convergence” of TBIF-based emissions measurements for eight cities to national average emissions suggests that TBIF-based measurement overcomes the transboundary measurement challenge that cities face. This should be a property not of city measurements themselves, but of the average of a group of city measurements. That is, there is no obvious reason each city should have emissions that reflect the national average, even if measured by activities; the combined effects of each city’s activities, measured with the TBIF, should reflect the national average as the number of cities measured increases as long as the emissions associated with “rural” activities largely service cities. A small variance in TBIF-based emissions measurement across several different metro areas is a distinct, and interesting, observation.

Policy Mathematically speaking, it does not matter which cities emissions are allocated to, as long as all emissions are accounted for and regulated. However, this “equivalence” of measurement does not necessarily help define appropriate policy to apply within the boundaries of cities that ultimately exist within a complex hierarchy of regulatory authority and may be heterogeneous in their desire to put in place policies that support reductions in GHG emissions. TBIF represents an improvement for policy for two reasons. The first concerns measurement alone. Suppose, for the moment, that there is no national policy. If some cities do not account for and regulate their emissions and each city that does account for and regulate emissions measures emissions via its geographic boundaries, then only some fraction of emissions will be accounted for and regulated. Even if every city accounts for and regulates its emissions on the basis of geographic boundaries, emissions from rural entities that drive activities associated with cities will not be accounted for or regulated. If, however, only some cities account for and regulate their emissions, but do so in a TBIF-based fashion, then emissions from other cities and from rural activities can be captured under regulation. Given that it is a (highly visible) minority of cities that are currently attempting to address their GHG emissions, this motivating factor for TBIF-based measurement is important. As soon as all entities are required to measure and reduce emissions, this issue is corrected.

316

W. Ross Morrow

The second reason TBIF represents a potential improvement for policy lies in socioeconomic equity and efficiency and applies even if all emitting entities are covered under some policy. If all cities account for and regulate their emissions and a national policy covers rural emissions not associated with cities, then all emissions can be addressed regardless of accounting based on emissions source or the end-use location of the associated good or service. However, the specific policy measures employed may not be economically efficient or socially equitable if policies are disconnected from the socioeconomic activities that drive emissions. Several useful examples are given in the chapter, including the greening of supply chains and increased use of telepresence in business. Another example might be refining capacity: the Gulf Coast contains a significant proportion of the petroleum- and chemicals-refining capacity for the United States, a capacity that largely exists to serve other major demand centers with transportation fuels and other goods. Direct regulation of metro areas with unusually high direct emissions would seem to disproportionately impact those areas relative to the areas that they serve. This risks market failures similar to the well-known principalagent problem. The use of TBIF or other more holistic emissions metrics would thus appear to be an improvement over purely geographic emissions accounting and regulation schemes from the perspective of effective and efficient policy for GHG emissions reduction in cities. However, TBIF-based measurement appears to require that cities accept higher levels of emissions associated with their activities (Hillman and Ramaswami 2010). Purely geographic measurement would seem to have two particularly attractive features for city officials. First, geographic measurement would localize policy activity on the actions that take place only within the city boundaries, offering boundaries for emissions inventory and reduction consistent with other components of city authority. Second, because geographic measurement has the potential unintended side effect of emissions relocation instead of reduction, it offers city officials the potential for relatively easy progress toward publicly stated goals. Accepting TBIF will thus require the “political will” to first accept responsibility for higher proportions of emissions than would likely be measured under a purely geographic measurement approach and then to design and implement policy measures that promote real changes in emissions rather than simply changes in geographically limited measures of emissions. To cities with a large “trade surplus” in goods and services associated with emissions, TBIF accounting may offer an advantage worth capitalizing on. For many cities, the political will to implement holistic measurement may have to be imposed through institutional arrangements or higher-level (state and federal) policy. As Ramaswami effectively points out, more creative efforts on the part of city officials appear to be required if cities are to participate meaningfully in GHG emissions reductions. Voluntary efforts, in particular, are shown to have extremely limited reductions in GHG emissions associated with city activities. While all improvements are valuable, the conclusion that emissions reductions only on the order of 1 percent per year are possible with current approaches

commentary

317

poses a challenge to decision makers who are genuinely interested in confronting the climate challenge. Agreements such as the U.S. Mayors’ Climate Protection Agreement, while motivating, lack explicit benchmarks for emissions performance and do not require continued measurement of policy performance. Until policy is closely linked to an initial inventory, specific quantitative targets, and subsequent emissions measurements, progress toward climate change goals will be limited.

Conclusions Ramaswami and her colleagues have, over the past five years, presented a useful body of research concerning activity-based measurement of GHG emissions for cities. Such measurements are motivated by independent actions currently being taken by cities to improve their GHG emissions profile in the absence of state and federal policies. Ramaswami’s chapter succinctly reviews the key challenge facing measurement for city policy: the existence of transboundary emissions produced outside geographic city boundaries that should nonetheless be addressed. The chapter also highlights the difficulties facing existing policy approaches at the city level. This commentary has tried to highlight several observations brought out by this research. First, the inclusion of transboundary emissions is necessary from a behavioral perspective, concerning the incentives and regulations that can most effectively and efficiently achieve emissions reduction, rather than being purely an accounting concern. Second, standardization of measurement methods is important for honestly addressing emissions from cities in an environment without consistent policies that reach across city boundaries. Such standardization could be executed within formal institutions that already link city officials or within the legislative language employed at higher levels (state or federal) of the policy hierarchy. Third, the results obtained by Ramaswami and her colleagues suggest that, for many cities, accepting activity-based metrics of emissions will more clearly illustrate the difficulty of emissions reduction. This confronts existing incentives for officials in many levels of government to realize short-run results to policy decisions that can be effectively communicated. Many current policy approaches, while often highly visible, may have limited impact on emissions calculated in a holistic fashion.

references Cooper, A. B., P. Georgiopoulos, H. M. Kim, and P. Y. Papalambros. 2006. Analytical target setting: An enterprise context in optimal product design. ASME Journal of Mechanical Design 128:4–13. Hillman, T., and A. Ramaswami. 2010. Greenhouse gas emission footprints and energy use metrics for eight US cities. Environmental Science and Technology 44:1902–1910.

12 Strengthening Urban Industry: The Importance of Infrastructure and Location Nancey Green Leigh

T

his chapter focuses on policies that can strengthen urban manufacturing and associated distribution/logistics activity in the context of shifting trends in industrial infrastructure, conversion pressures on urban industrial land, and attempts to limit low-density residential development. Within urban policy development and planning practice, a little-acknowledged conflict exists: efforts to increase population density and promote more effective use of grey infrastructure to support residential density contribute to industrial sprawl and the need to extend grey infrastructure further out in metropolitan areas. Grey infrastructure consists of human-made systems that support population and economic activity, such as roads, bridges, rail, water, and sewer systems. The continued expansion of grey infrastructure contributes to the growing carbon footprint of metropolitan areas and to associated climate change implications. It also adds to the long-unresolved burden of maintaining the nation’s infrastructure. Urban agglomeration theory, beginning with Alfred Marshall (1961), has long made the case for firm proximity (i.e., industrial density or clustering) as a fundamental feature of innovation and competitiveness. In resolving the conflict between strengthening urban manufacturing and increasing population density (along with associated commercial and service activity), policy makers require a better understanding of which industries are most competitive in central cities, and within these industries, which functions (research and development, production, distribution, remanufacturing, recycling) are most compatible with the surrounding area’s activities. This understanding is needed for creating a robust approach to providing urban infrastructure that supports industry in a manner 318

strengthening urban industry

319

compatible with desired residential, commercial, and service (private and public) activity. The argument for explicit attention to strengthening urban industry is based on five premises. First, urban industry is essential for two key reasons: to increase U.S. exports and elevate U.S. leadership in advanced manufacturing, which will help the country maintain its global economic position, and to address chronic problems of central city unemployment and poverty, the incidence of which is higher around older industrial areas. Second, the changing landscape of urban industry has shifted the location and purposes of buildings such as warehouses and distribution centers. Third, underinvestment in infrastructure and poor maintenance constrains urban industry and the development of the information economy. Fourth, the public sector is critical in developing policy that promotes urban brownfield development rather than suburban greenfield construction. A distorted land market in metropolitan areas disadvantages efforts to retain an adequate supply of industrial land. Distortions are driven by government subsidization of suburban and exurban development. This subsidization occurs at all levels of government and takes the form of property tax and other incentives and the subsidization of road construction and other infrastructure. Finally, climate change is real, and efforts to strengthen urban industry and maintain central city industrial land play a significant role in climate change mitigation by creating less pressure for expansion and by encouraging lower-impact development.

The Movement to Strengthen Urban Industry and Economic Development NATIONAL LEVEL

The manufacturing sector creates higher value than any other major sector in the economy. For example, for every dollar of manufacturing output, another $1.35 in economic activity is created, compared to $0.95 for transportation, $0.88 for information, and $0.63 for finance (President’s Council of Advisors on Science and Technology 2012). However, the U.S. position as the leading producer of manufactured goods is widely seen to be eroding: The loss of U.S. leadership in manufacturing is not limited to low-wage jobs in low-tech industries, nor is it limited to our status relative to lowwage nations. The hard truth is that the United States is lagging behind in innovation in the manufacturing sector relative to high-wage nations such as Germany and Japan, and the United States has relinquished leadership in some medium- and high-tech industries that employ a large proportion of highly skilled workers. In addition, the United States has been losing significant elements of the research and development (R & D) activity linked to manufacturing to other nations, as well as its ability to compete in the manufacturing of many products that were invented and innovated here—from laptop computers to flat panel displays to lithium ion batteries. (President’s Council of Advisors on Science and Technology 2012, 2–3)

320

Nancey Green Leigh

National-level policy is being directed toward strengthening the U.S. position as a creator of advanced manufacturing because it is viewed as critical for maintaining the nation’s leading economic position in the global market. Ezell and Atkinson identify five reasons why manufacturing is essential to a healthy economy (2011, 2): 1. It will be extremely difficult for the United States to balance its trade account without a healthy manufacturing sector. 2. Manufacturing is a key driver of overall job growth and an important source of middle-class jobs for individuals at many skill levels. 3. Manufacturing is vital to U.S. national security. 4. Manufacturing is the principal source of R & D and innovation activity. 5. The manufacturing and services sectors are inseparable and complementary. Ezell and Atkinson’s call for a national manufacturing strategy stems from three concerns. First, U.S. manufacturers face unfair competition because other countries have explicit manufacturing strategies. Second, market failures and externalities associated with manufacturing cause underperformance. Finally, once a manufacturing sector is lost to international competition, it is very unlikely it can be regained. In a review of 2009 and 2010 presidential and congressional initiatives and legislation, the U.S. Congress Joint Economic Committee (2010) reported more than a dozen other national economic recovery efforts supporting manufacturing exports and sustainable manufacturing, including the U.S. Manufacturing Enhancement Act, which was signed into law in August 2010. In June 2011, the committee followed up their review, published in Understanding the Economy: Promising Signs of Recovery in Manufacturing, with a hearing titled “Manufacturing in the USA: Why We Need a National Manufacturing Strategy.” Instead of a manufacturing policy that focuses on certain industry sectors, the panelists supported national manufacturing policies focusing on total exports and innovation and on specific application of knowledge and technology in advanced and sustainable manufacturing processes and products.1 As one of the responses to the Great Recession, a strong interest has developed in strengthening manufacturing. Recent national policy efforts focus on

1. In July 2012, the President’s Council of Advisors on Science and Technology issued “Report to the President on Capturing Domestic Competitive Advantage in Advanced Manufacturing.” Advanced manufacturing is defined as a “family of activities that (a) depend on the use and coordination of information, automation, computation, software, sensing, and networking, and/or (b) make use of cutting edge materials and emerging capabilities enabled by the physical and biological sciences, for example nanotechnology, chemistry, and biology. This involves both new ways to manufacture existing products, and especially the manufacture of new products emerging from new advanced technologies” (2012, ii).

strengthening urban industry

321

innovation in manufacturing (especially advanced manufacturing), promoting exports, and supporting manufacturing’s transformation to cleaner, more sustainable production. In June 2011, President Obama announced the Advanced Manufacturing Partnership of industry, government, and universities to invent and deploy new manufacturing technology, processes, and products. Earlier, the president promoted the National Export Initiative in his first two State of the Union addresses. The National Export Initiative established a goal of doubling U.S. exports by 2014. To meet this goal, a new federal program, the Renewable Energy and Energy Efficiency Export Initiative, began in late 2010 to focus on increasing U.S. capacity in clean energy manufacturing, promoting exports of renewable energy and energy-efficient manufactured goods and services, and encouraging waste reduction (Trade Promotion Coordinating Committee 2010). While it was the financial institutions and devastated residential and commercial markets that created the Great Recession and its lingering impacts, exports—particularly manufacturing exports that are dependent on an adequate supply of industrial land and infrastructure—are essential to a full recovery and continued prosperity. As Istrate and Marchio observe: In a slow recovery, exports are essential to job creation and the reorientation of the U.S. economy towards productive economic growth. Metropolitan areas are a vital part of this proposition. In 2010, exports were a major driver of the U.S. recovery and the largest metropolitan areas produced the majority of the nation’s exports. While the overall economy was still losing jobs, the rapid growth of U.S. export sales translated into 600,000 additional jobs in the first year of recovery. These are jobs not only in the industries producing the exported goods and services, but also in the suppliers to the exporting industries, and in the case of merchandise exports, in the transportation and wholesale trade industries. Manufacturing drove the rapid recovery of U.S. export sales. (2012, 21) METROPOLITAN LEVEL

A 2010 Brookings study highlights several advantages of metropolitan areas for export activity: a large and diverse pool of workers, strong logistics networks to move cargo, greater capacity (universities, investment in research and development, and basic, science, and venture capital) to support innovative products that can become exports, and proximity to other export-oriented firms. The study calls for greater support to increase the export activity of the nation’s metropolitan areas, and it specifically states: “Local metropolitan leaders should be concerned with increasing the export intensity of existing companies rather than simply recruiting new ones” (Istrate, Rothwell, and Katz 2010, 7). Urban areas actually do play an essential role in the manufacturing economy, as a recent Brookings study on the location of manufacturing illustrates (Helper, Krueger, and Wial 2012). The study found that 80 percent of all manufacturing jobs, including 95 percent of the highest technology jobs, were located in metro areas in 2010. Further, while manufacturing plants are typically thought to be

322

Nancey Green Leigh

large in size with hundreds of employees, the average manufacturing plant in the nation’s metro areas had only 57 employees. This raises the question of whether individual manufacturing plants require less space than is commonly perceived. Data for 2000 to 2010 show that the long-term trend of manufacturing jobs shifting away from metro areas and central metro counties continued. In particular, the location of high-tech firms shifted from central to outlying counties in metro areas between 1980 and 2010. The study’s authors state that this trend “should be an important policy concern” given that “firms in higher-density environments are more productive [and therefore] decentralization of manufacturing clusters could undermine the competitiveness of U.S. manufacturing” (Helper, Krueger, and Wial 2012, 31). The 366 metro areas in the United States generated 84 percent of manufacturing exports in 2010, but most of this activity was concentrated in the top 100 metro areas (based on population). The top 10 manufacturing exports in 2010 were (in rank order): transportation equipment; chemicals; machinery; computer and electronic products; petroleum and coal products; food; primary metal; medical equipment, sporting goods, and miscellaneous; fabricated metal products; and electrical equipment. The top 100 metro areas generated 75 percent of manufacturing exports, 75 percent of national gross domestic product, and two-thirds of all jobs (Istrate and Marchio 2012). Researchers recognize that most U.S. export activity originates in metropolitan areas and that manufacturing constitutes the majority of metropolitan export activity (Atkinson and Gottlieb 2001; Berube 2007; Istrate and Marchio 2012). Helper, Krueger, and Wial (2012) found that around two-thirds of the nation’s metro areas exhibit manufacturing clustering in one of six areas: computers and electronics, transportation equipment (trains, planes, autos, and ships), low-wage manufacturing (textiles, apparel, and furniture), chemicals, machinery, and food. They suggest that the continuation of recent gains in manufacturing employment will be “highly shaped by the local dynamics of regional supply chains and industry clusters” (Helper, Krueger, and Wial 2012, 35). Dense economic activity is vital, but they caution that market forces will not produce the amount of clustering activity the United States needs to remain competitive. They observe, for example, that firms underinvest in research and development and worker training because they take advantage of other firms’ investments. Thus, they call for “geographic high road” policies, such as more worker training and R & D, instead of the dominant policies of tax abatements and location subsidies that favor nonmetro and outlying metro counties. These “low road” policies are predicated on the assumption that what makes a location desirable is low wages, “even though such wages typically account for far less than 20 percent of a manufacturer’s total costs” (35). The national-level focus on strengthening manufacturing and exports shores up the efforts undertaken by a small group of forward-thinking U.S. cities to preserve industrial land in their urban cores. These efforts began well before the Great Recession. These cities were bucking the trend of the majority that seemingly accepted the view of a postindustrial economy, one in which supply-side and

strengthening urban industry

323

demand-side economic development strategies focused on real estate, retail, services, and housing development, largely framed within the smart growth movement. The acceptance of this viewpoint led to substantial urban industrial land rezoning in some cities and pressured industrial businesses into leaving urban industrial districts by raising rents and land values. The economic development approaches of the public and private sectors in these cities ignored industrial uses. Essentially, in buying into the postindustrial worldview, many cities allowed themselves to become deindustrialized. The highly influential smart growth movement played a significant role in this process. In “Smart Growth’s Blindside,” Leigh and Hoelzel (2012) critique the movement for its failure to “recognize connections between urban industrial land and the activities it supports with smart growth goals of limiting sprawl and revitalizing central cities” (87). They review the recent local industrial policies of 14 cities2 and 10 influential practice-oriented smart growth publications3 with local economic development components. In the 14 cities initiating local industrial policies, a significant amount of industrial land was converted to other uses as city planners pursued smart growth. (Table 12.1 provides details for eight of these cities.) Leigh and Hoelzel compare elements of the cities’ adopted local industrial policies with commonly accepted smart growth principles (table 12.2) to illustrate the contradictions between smart growth policies and efforts to protect and revitalize urban industrial areas. Further, Leigh and Hoelzel’s analysis of the smart growth literature found little to no acknowledgment of the need to coordinate urban industrial development practices with other mainstay smart growth activities. As a consequence, industrial land failed to be protected from conversion pressures. Leigh and Hoelzel conclude that approaches are needed that explicitly safeguard productive urban industrial land and discourage industrial sprawl. Further, policy makers need to avoid treating (whether consciously or not) smart growth and sustainable urban industrial development as conflicting goals. In addition to the need for industrial land to fulfill the key role that cities have to play in growing manufacturing and exports, industrial land is needed for critical urban services infrastructure, warehouses for goods coming in and out of the city, and private and public sector industry maintenance and repair activity. Taken together with manufacturing, these activities have more recently come to be labeled production, distribution, and repair (PDR) activities. Dempwolf 2. Atlanta, Baltimore, Boston, Chicago, Los Angeles, Minneapolis–St. Paul, New York, Oakland, Philadelphia, Portland, San Francisco, San Jose, Seattle, and Washington, DC. 3. These include American Planning Association (2003, 2009), Congress for the New Urbanism and U.S. Environmental Protection Agency (2002), International City/County Management Association and Smart Growth Network (2002, 2003), International Economic Development Council (2006), National Association of Local Government Environmental Professionals and Smart Growth Leadership Institute (2004), Smart Growth Leadership Institute (2007), Smart Growth Network, International City/County Management Association, and U.S. Environmental Protection Agency (2006, 2009).

324

Nancey Green Leigh

Table 12.1 Loss of Industrial Land to Rezoning in Select U.S. Cities City Atlanta, GA Boston, MAb Minneapolis–St. Paul, MNc New York, NYd Philadelphia, PAe Portland, ORf San Francisco, CAg San Jose, CAh a

Industrial Land Lost (Acres)

% Lost

Years

800 960 1,812 1,797 1,645 489 1,276 1,470

12 38 18 14 8 2 46 9

2004–2009 1962–2001 1990–2005 2002–2007 1990–2008 1991–2001 1990–2008 1990–2009

Leigh et al. (2009). Boston Redevelopment Authority (2001). c CDC Associates (2008). d Pratt Center (2009). e City of Philadelphia (2011). f City of Portland (2003). g San Francisco (2008). h City of San Jose (2009). a b

(2009) attributes this renaming of industrial land to efforts to avoid negative reaction to industrial activity as well as efforts to emphasize how industrial land use is connected to the overall land use system. The label also highlights possibilities for finer-grained planning approaches that are “more contextual and integrative” (15). The central cities of metropolitan regions are also interested in preserving their industrial lands as a means to reduce high poverty and unemployment rates. Despite declining unionization levels, which historically have provided high wages and benefit levels, and despite a recent trend whereby manufacturing productivity increases have not been reflected in employee compensation, manufacturing still pays significantly higher wages for blue-collar workers lacking a college education than other sectors, even when controlling for worker and job characteristics (Helper, Krueger, and Wial 2012). Further, the older working-class neighborhoods that are adjacent to many inner-city industrial areas often have the greatest concentrations of poverty. For example, 20 percent of Atlanta’s population but 33 percent of its poverty population live within a one-mile buffer from the center of each of the city’s three primary industrial areas (Leigh et al. 2009). While historical zoning practice established the separation of industrial uses from other uses for health and human safety, most of today’s industry does not require such separation. Urban industrial land can be made more attractive for manufacturing and related uses through innovative urban design and the accep-

strengthening urban industry

325

Table 12.2 Local Industrial Issues, Policies, Smart Growth Planning Summary of Urban Industrial Development Issues and Priorities in 14 Local Industrial Policies

Summary of Smart Growth Issues and Priorities Affecting Urban Industrial Development

Land Use Planning Issues and Priorities

V Loss of industrial land and ad hoc zoning conversions threatening productive industrial areas. V Market-driven overpricing of industrial land and competition from other land use alternatives. V Encroachment and compatibility of uses within and surrounding industrial areas.

V V V

Rezone land for functionality and compatible mixes of use. Facilitate transit-oriented development (TOD) and greater access to jobs. Foster compact and dense infill development.

Local Economic Development Planning Issues and Priorities

V Lack of available productive industrial land for advanced manufacturing and sustainable industrial businesses. V Link workforce training to high-quality, local industrial jobs. V Foster supportive and innovative business climates for industry.

V V V

Balance jobs and housing. Reduce job sprawl and job-resident spatial mismatch. Improve employment diversity, quality, and wages in urban job centers.

Source: Leigh and Hoelzel (2012).

tance of mixed land uses. Urban design can address the need for quality space that is buffered from and complementary to nearby residential uses (including appropriate infrastructure for trucks, autos, bicycles, and pedestrians). This can be accomplished through form-based design codes like the transects or overlay districts that are being-used in the Little River/Little Haiti industrial district of Miami (Miami21 2012). Mixing industrial activity with appropriate uses that yield higher property taxes can make it more profitable to retrofit industrial properties and can reduce the pressure for the wholesale rezoning of industrial land. The city of San Jose adopted a new development policy for its northern industrial area in January 2012. The area, known as Rincon de los Esteros, the Innovation Triangle, or the Golden Triangle, has been the location of some of the city’s well-known high-technology companies. It developed in a uniform and low-intensity manner since the 1980s as a result of city policy that was focused on regional traffic concerns and resulted in a floor area ratio of 0.35. Rincon de los Esteros “is characterized architecturally by low to mid-rise office buildings one or twostory light manufacturing and research & development facilities, surface parking lots and generous amounts of landscaping. . . . The block pattern is large

326

Nancey Green Leigh

and irregular and access into North San Jose is provided mostly from a limited number of regional freeways or expressways” (City of San Jose 2012, 3). The updated policy provides for an additional capacity for 20 million square feet of development, and specific sites can be converted from industrial to high-density residential sites based on specific compatibility criteria for industrial activity. The policy adds two land use changes to the area. The first is an industrial Core Area designation to support a “driving industry” corporate center along a primary corridor. The city seeks to allow and encourage more intense development for “driving industry” businesses along the North First Street Corridor. Driving industry businesses are businesses that sell goods and/or services outside of the region, bringing in significant revenues that help drive the San Jose economy. The City envisions a very active corridor of mid-rise (4–12 story) industrial office buildings . . . [that] will foster a concentration of high-tech businesses located so as to make best use of existing infrastructure resources. (City of San Jose 2012, 5)

The resulting floor area ratio is expected to be 1.2. The second land use change is a Transit/Employment Residential District Overlay to expand supporting residential and commercial uses in the industrial area. This land use change provides for the development of up to 32,000 new residential units . . . through the conversion of . . . existing industrial lands within a proposed Transit/Employment Residential District Overlay area. New residential units would also be allowed through mixed-use development within the Core Area. . . . This residential development is intended to provide housing in close proximity to jobs to allow employees the opportunity to reduce their commute travel times, to make increased use of transit facilities, and to reduce overall traffic congestion. (City of San Jose 2012, 6)

The policy specifies needed improvements to transportation infrastructure to support the new high-intensity development.

Changing Landscapes of Urban and Suburban Industry Prior to suburbanization, manufacturing in cities tended to be located in multistory buildings, and warehouse facilities were located along rail spurs. Due to historical pre-auto building density patterns, these properties were more likely to be bounded on all sides, making it impossible to expand them (Fitzgerald and Leigh 2002). Thus, carving out the large industrial sites that would be required of a major distribution center would likely require assembling multiple parcels. Manufacturing suburbanization after World War II coincided with shifts to mass production layouts in one-story large-footprint buildings that were surrounded by parking lots and adjacent to major road networks with easy on and

strengthening urban industry

327

off access. The predominant production mode was a manufacture-to-stock or just-in-case system generating large product inventories that, in turn, required large warehouses. But the return to manufacture-to-order or just-in-time modes of production associated with lean production can counter the perceived obsolescence of central city industrial facilities that have smaller footprints and multiple stories (Leigh 1996). New emphases on research and development activity, as well as incubators for advanced manufacturing, often coming out of urban universities, can also spur demand for central city industrial facilities. A recent special report by The Economist entitled “The Third Industrial Revolution” predicts “there will be millions of small and medium-sized firms that will benefit from new materials, cheaper robots, smarter software, an abundance of online services and 3D printers that can economically produce things in small numbers” (Economist 2012, 20). To realize the significant opportunities for innovation from these new technologies requires proximity between R & D activity and manufacturing. These trends suggest that urban industrial land can be reused in the advanced economy for industrial activity. When industrial properties need retrofitting, however, reuse presents great challenges. The low rental rates of these properties make it difficult to pay for renovations. If the properties are also brownfields in need of environmental remediation, they are “upside down” even before renovation or modernization activity takes place. (The cost of brownfield versus greenfield development will be explored in a later section of this chapter.) Beyond the demand for industrial land created by trends such as smart growth or “back to the city” movements, a key reason for the conversion of industrial land has been high vacancy rates. Shifts in the building and infrastructure requirements for manufacturing and distribution are a major factor contributing to the high vacancy rates. Much of the older industrial urban land is considered functionally obsolete. Thus, little has been built in older urban areas for either manufacturing or warehouse activity. Faced with high industrial vacancy rates and pressure to revitalize underperforming property, it is not surprising that the public and private sector have been distracted from considering the implications of declining urban industrial land supplies. Older buildings, whether manufacturing or warehouse, in urban or older suburban locations, may require repositioning to avoid vacancy. For example, Larry Callahan, CEO of Patillo Industrial Real Estate, a firm that has operated in the southeastern United States for over 60 years, suggests they can be useful for subassembly operations that produce components for larger manufacturing operations, incubators, repair operations, and showroom or display activity. City warehouses have become functionally obsolete for several reasons. The shift in freight transport from rail to air and trucking has decreased the demand for centralized urban warehouse space along rail spurs. The shift to containerization has transformed the way goods are moved to market, along with much of the industrial infrastructure that supports distribution activity. The railroad companies that transport containers to and from port cities and between major

328

Nancey Green Leigh

regions of the country operate in long-haul mode. It is not cost effective for them to off-load one container at a rail spur in an urban center or even a suburban node. Instead, they move between the ports and multimodal centers. At the multimodal centers, they off-load high volumes of containers, which are then put on trucks and driven to cities. This shift has made older warehouses with smaller loading bays unusable for many industrial businesses, as has the use of larger tractor-trailers (increasing in length from a standard of 40 feet to 53 feet), which have difficulty navigating through central-city streets. Trucking deregulation resulted in transportation costs being minimized when goods were shipped from a national or regional location that minimized distance to all customers. The centralization of warehousing, along with advances in storage and disbursement of products, required much larger warehouse spaces with high ceilings for automated racking systems. The primary new industrial building construction of the last couple of decades has been large warehouses known as distribution centers that take up much larger footprints than typically found in center cities. Callahan observes that the size of warehouse buildings grew from 100,000 square feet (SF) in the mid-1980s to 800,000 SF in the late 1990s, and some are over 1 million SF today (pers. comm.). An 800,000 SF building occupies 18.3 acres. As a state-of-the-art facility, it requires 30-foot ceiling heights (older space often has half that height) and a truck court with 200 feet depth on both sides of the building to accommodate 120 truck bays. This allows for product to be brought in one side of the building, processed within the building, and then taken out for distribution on the other side. Altogether, the facility consumes 60 or more acres and must be located on land near a freeway exit. Callahan suggests that distribution centers are being built in exurban environments because of the large footprint they require, not because it is the best logistical location (pers. comm.). However, a logistics expert suggests that these decentralized locations may become a significant issue if, as he expects, the need to contain rising fuel costs supersedes the need for cheap land (McCurry 2012). Essentially, the distribution/warehouse building segment of the industrial real estate market has evolved into a tiered system. At the top end are very large distribution centers located in exurban locations. Demand exists for smaller buildings for sectors that need to provide quick delivery, such as medical products, or that provide support for smaller operations, such as handling product returns. The growing interest in materials reuse, remanufacture, and recycling is also creating demand for less expensive warehouse space in cities. There is still demand for warehouse space that supports the efficient distribution of products specifically created for inner-city markets. Firms that produce products that are more easily transported on smaller trucks (due to small size or smaller volumes) can also use inner-city warehouse space (Fitzgerald and Leigh 2002). Additionally, industrial space that is obsolete for warehousing could be converted to manufacturing.

strengthening urban industry

329

Infrastructure Challenges for Urban Industry KEY TYPES

Traditional infrastructure requirements for industry have expanded with the development of the information economy. While specific requirements vary across industry sectors, in general, the primary determinants of infrastructure requirements fall into five categories: 1. Characteristics of the industrial site: elevation, soil type, storm drainage system. 2. Transportation: access; links to and types of highways; distance from mass transit, airports, and ports to industrial site; bus and rail service availability. 3. Water and sewer: size of water and sewer mains, capacity of treatment plants. 4. Energy: natural gas and electric power quality. 5. Telecommunications: distance to central office, switch (is fiber available?), Points of Presence (distance of point where long-distance carrier hands off service to subscriber). A fundamental issue for reusing industrial land for new industrial use lies in the issue of financing retrofits as well as maintaining infrastructure. Although it is not within the scope of this chapter to provide case study data on these costs, the next section’s accounting of overall finance issues for the maintenance of infrastructure provides some perspective on the challenges. COST OF PROVISION AND MAINTENANCE

In a series of reports issued under the main title “Failure to Act,” the American Society of Civil Engineers (2011a, 2011b, 2011c) explored the economic impact on industry of chronic underinvestment in three of the five key areas of infrastructure: surface transportation, water and sewer, and electricity. Surface Transportation The report on surface transportation infrastructure (American Society of Civil Engineers 2011b) noted that in 2010 the United States ranked 19th out of the top 20 countries for quality of roads, and 18th out of the top 20 for quality of railroads. In particular, the report quantified the negative impacts to the United States’ ability to export if the surface transportation deficiencies are not corrected: By 2040 the cost of infrastructure deficiencies is expected to result in the U.S. losing more than $72 billion in foreign exports in comparison with the level of exports from a transportation-sufficient U.S. economy. These exports are lost due to lost productivity and the higher costs of American goods and services, relative to competing product prices from around the globe. (2)

330

Nancey Green Leigh

The poor state of surface transportation is expected to cause businesses to divert increasing portions of their earned income to pay for transportation delays and vehicle repairs, income that could be invested in innovation and expansion. The surface transportation infrastructure in urban areas is in worse shape than in rural areas, and thus urban areas—major cities in particular—will experience the greatest negative impacts from the systemic failure to invest in infrastructure maintenance and new infrastructure. The American Society of Civil Engineers (ASCE) report provides calculations of costs to correct the deficiencies at the present time and into the future. Failure to do so poses particular problems for efforts to strengthen urban manufacturing. Water and Sewer Each year, new water lines are constructed to connect more distant dwellers to centralized systems, continuing to add users to aging systems. Although new pipes are being added to expand service areas, drinking-water systems degrade over time, with the useful life of component parts ranging from 15 to 95 years (American Society of Civil Engineers 2011c). Water is, of course, essential to human life and to the economy. As the second ASCE “Failure to Act” report observes, farms depend on irrigation to grow crops, while commercial businesses and government offices require clean water. Particularly relevant to urban manufacturing, industries such as food and chemical manufacturing as well as the power plants that supply electricity cannot operate without “clean water that is a component of finished products or that is used for industrial processes or cooling” (American Society of Civil Engineers 2011c, 1). But as the above quote implies, the underfunded and undermaintained U.S. water system is continually taxed by the decentralization of population and industry. The failure to invest in an adequate water and sewer system is estimated to cost businesses in the United States $147 billion between 2011 and 2020; if left unaddressed, the cost will be $1.487 trillion between 2020 and 2040. Electricity Of the three infrastructure types, electricity is distinguished by the fact that it is largely privately owned. However, it is publicly regulated. The infrastructure of the industry is divided into three segments that are connected and interdependent: (1) generation plants; (2) transmission lines; and (3) local distribution equipment (American Society of Civil Engineers 2011a). Also distinct from the other two infrastructure types, electricity infrastructure has experienced substantial investment in recent years. However, ASCE’s “Failure to Invest” report on this topic indicates there will still be a cumulative electricity infrastructure funding gap of $107 billion by 2020 that will cost businesses and households $197 billion (in 2010 dollars). In particular, substantial losses in the manufacturing sectors are projected from less reliable electricity service due to the failure to invest. By 2020, this could result in a $10 billion loss in exports; this loss will grow to $40 billion by 2040 if not addressed, countering national efforts to maintain global competitiveness.

strengthening urban industry

331

A fundamental question is whether our major cities have adequate electricity infrastructure to support the desired increases in manufacturing and export activity at the national and local levels. Congress has mandated congestion studies of the electricity grid. In 2009, four megaregions were identified as significant areas of concern. On the East Coast, this included New York down to Washington, DC. On the West Coast, the Seattle-Portland region, San Francisco region, and greater Los Angeles region were all identified (U.S. Department of Energy 2009). Areas of concern are more likely to experience higher electricity costs as users compete for inadequate supply as well as experience reduced grid reliability. These four megaregions encompass cities that Leigh and Hoelzel (2012) have identified as actively seeking to protect industrial lands. An additional reason to be concerned about the state of the electricity infrastructure across the country is because green infrastructure systems and predictions of a third industrial revolution that will advance economies and civilization are dependent on electricity. These will be discussed in the final section of this chapter.

Urban Brownfield Versus Suburban Greenfield Development Redeveloping brownfield industrial sites is another fundamental challenge that contributes to high urban industrial vacancy rates as well as the suburbanization of industry. Brownfield industrial sites are those that were previously developed and are known or suspected to have some form of environmental contamination. The U.S. Environmental Protection Agency and state-level environmental protection agencies regulate the cleanup of brownfield sites. Different standards of cleanup are applied, based on the proposed reuse of the site. For example, a brownfield redevelopment site that is being proposed for residential use will have higher standards of cleanup than one proposed for commercial or industrial use. The standard of cleanup will greatly affect the cost of redevelopment, as will the overall size of the site and the type and extent of the contamination. Cleaning up a brownfield site for industrial reuse is generally considered to be less expensive than for residential use. However, central city industrial land is already less desirable for much of the industrial real estate market and has low rental values due to obsolescent building structures, lower road network accessibility, aging infrastructure, and other factors. Thus, the costs (in dollars and time) of remediation create an additional disincentive to locate on previously developed urban industrial land. While government environmental regulation can be said to have instigated the brownfield problem, in response, all levels of government have created support for brownfield redevelopment. Over the last quarter of a century, governments have offered grant and loan programs, financial incentives, technical expertise, and regulatory clarity. At the same time, private sector development expertise and support (e.g., finance, insurance, cleanup technology) have evolved, making brownfield redevelopment more feasible and attractive (Leigh 2008, 2009).

332

Nancey Green Leigh

Published data comparing the actual costs of brownfield redevelopment and greenfield development for industrial use are scarce. De Sousa’s 2002 study of the Toronto metro area constructed prototypical development scenarios for brownfield and greenfield industrial sites from an examination of actual development projects. The brownfield site was located on city of Toronto port land that had mid-level contamination. The greenfield site was located in a business park in a pro-growth suburban community that had attracted significant development. The brownfield site’s estimated tax revenues were nearly three times that of the greenfield site, while development charges were only 16 percent and transportation externalities were less than 80 percent of those of the greenfield site (calculated from De Sousa 2002, table 9). De Sousa’s research focused on public sector benefits and costs. His earlier work (2000) focusing on the private sector found that “the perception that brownfield redevelopment is less cost-effective and entails greater risks than greenfield development, on the part of the private sector, is true for industrial projects” (1). In Portland, Oregon, a consortium of public agencies sponsored a study that compared the costs and issues associated with industrial sites in greenfields and brownfields (Port of Portland 2004). The study focused on four industrial uses considered appropriate for the Portland metropolitan area: high-tech manufacturing, industrial park, warehouse/distribution, and general manufacturing. The report detailed the uses and specifications for the sites under comparison as follows: U

U

U

U

High Tech Manufacturing includes high technology industries that are primarily related to manufacturing and processing. In this study, a 350,000 SF high-tech facility is tested that includes two 125,000 SF fabrication plants, one 40,000 SF central utility building, one 60,000 SF office building and 725 parking spaces. Industrial Park is a series of larger individual buildings whose uses could include light industrial manufacturing, distribution or industrial services. For this project, 630,000 SF of industrial park space, divided into multiple buildings, was tested on both sites. Warehouse/Distribution includes industries primarily engaged in the warehousing, storage and distribution of goods. For this project, 400,000 SF of distribution space in a single building with 200 parking spaces and 300 trailer spaces was tested on both sites. General Manufacturing includes industries utilizing manufacturing processes. For this project, three single-user general manufacturing facilities were tested on each site. These facilities totaled 450,000 SF in three buildings—a 100,000 SF user, a 150,000 SF user, and a 200,000 SF user—and 1,100 parking stalls to serve all three facilities. (2)

Greenfield and brownfield sites appropriate for the four uses were identified in the Portland metro area. The costs of development were classified into four categories: on-site construction costs, system development charges and credits,

strengthening urban industry

333

off-site construction costs, and environmental remediation costs for the brownfield sites. The results of the analysis, depicted in table 12.3, show that brownfield remediation costs were greater than greenfield infrastructure development costs when viewed from the perspective of a private developer doing a speculative development. For example, in the industrial park row of the table, brownfield remediation costs exceeding $8.7 million have to be incurred before redeveloping the potential site compared to $5.2 million that must be spent providing infrastructure to the potential greenfield site. (In the cases examined, the warehouse infrastructure costs for the potential greenfield were less than those for the potential brownfield site.) The study concluded: There is an economic challenge to maintaining industrial zoned brownfields as industrial properties after they are cleaned up. The remediation costs of bringing an “upside down” brownfield site “right side up” often cannot be recovered when the site can be developed only for industrial land values. Industrial land values in the Portland metropolitan area tend to range from $3.50 to $6.50 per square foot, the lowest value of any major land use. For comparison, office and residential land ranges from $7.50 to $10.00 per square foot, while commercially zoned land is valued at significantly higher levels. As remediation costs must be deducted from land value, industrially zoned property has the most limited ability to absorb clean-up costs while still maintaining a positive residual land value. (10)

Because the private sector has little incentive to redevelop industrial-zoned brownfields as industrial properties, the role of the public sector is critical for strengthening urban manufacturing and export activity. The Portland study reiterates calls elsewhere for the public sector to help the private sector by reducing Table 12.3 Comparing the Costs Associated with Greenfield and Brownfield Development in Portland, OR Use Industrial Park General Manufacturing High-tech Manufacturing Warehouse/ Distribution

Brownfield Remediation Costs

Overall Cost Differential

Total

PSFa Bldg.

Total

PSF Bldg.

Total

PSF Bldg.

$8,748,863

$13.89

($5,181,167)

($8.22)

$1,319,162

$2.09

$22,980,475

$51.07

($1,323,000)

($2.94)

$21,581,081

$47.96

$28,027,465

$80.08

($1,428,500)

($4.08)

$27,030,361

$77.23

$7,821,799

$19.55

$444,500

$1.11

$8,553,079

$21.38

PSF per square foot. Source: Port of Portland (2004).

a

Greenfield Differential Costs

334

Nancey Green Leigh

the cost of capital and assisting with the initial characterization of contaminated sites.

Transforming Infrastructure for Greener, Lower-Carbon, Evolving Industries Industry is the most intensive infrastructure user of the economy’s sectors. Industry uses all forms of surface, air, and water transportation. It is an intensive user of land and built structures (the latter of which are some of the largest constructed). And it is an intensive user of utilities: power, water, sewer, and telecommunications. The suburbanization and exurbanization of industry has directly contributed to the expansion of grey infrastructure. While the creation of this industrial sprawl is presented as an inevitability of the evolving competitive economy, it is worth considering whether this is really true. Are there alternative industrial development patterns and infrastructure systems that can foster greater sustainability and still be competitive in advancing development? At the level of the industrial site and from the perspective of the firm, there may be cost efficiencies in the green retrofitting of buildings and site-specific infrastructure. Callahan of Patillo Industrial Real Estate states that lighting should always be the first retrofit to make as new fixtures provide better illumination and reduce electricity costs; therefore, it pays for itself. Plumbing often has to be retrofitted because of changes in code requirements, but doing so saves on water and sewer fees. Callahan notes that over the last 10 years, roofs have been increasingly retrofitted with white thermoplastic olefin (TPO) tiles that improve energy efficiency and last longer than conventional roofing (pers. comm.). Installing energy-saving green roofs help lower the urban heat island effect. The best-known industrial green roof, and also the largest in the world, is on the Ford Rouge truck manufacturing plant in Dearborn, Michigan. The 454,000 SF, or 10.4 acre, roof is planted with a drought-resistant groundcover known as sedum that weighs less than 15 pounds per square foot. The sedum traps airborne dust and dirt, absorbs carbon dioxide, and creates oxygen, all of which improve air quality. However, the roof’s primary function is to collect and filter rainfall as a key part of a natural storm water management system (The Henry Ford n.d.). Ford is an example of a private company that voluntarily adopted a lowimpact infrastructure system. Low-impact development is typically focused on water and sewer infrastructure. Local governments seeking to lower the cost of providing infrastructure are beginning to mandate low-impact development practices for industrial and other forms of developments. Los Angeles implemented its Low Impact Development Ordinance in May 2012 because it is seen as a cost-efficient means of managing storm water and decreasing water pollution. The ordinance applies to new industrial development and any existing industrial development where more than 500 square feet of hardscape is added. The ordi-

strengthening urban industry

335

nance requires that rainwater from a rainstorm of three-quarters of an inch or more must be captured, infiltrated, or used on site (City of Los Angeles n.d.). NEW INDUSTRY FROM MORE SUSTAINABLE INFRASTRUCTURE

Jeremy Rifkin (2012), the principal architect of the EU’s Third Industrial Revolution economic sustainability plan, argues that as infrastructure transforms to support lower carbon emissions, new industries may emerge. Rifkin identifies what he calls five pillars of the infrastructure of a third industrial revolution: 1. Shifting to renewable energy. 2. Transforming the building stock of every continent into micro–power plants to collect renewable energies on-site. 3. Deploying hydrogen and other storage technologies in every building and throughout the infrastructure to store intermittent energies. 4. Using Internet technology to transform the power grid of every continent into an energy-sharing intergrid that acts just like the Internet. When millions of buildings are generating a small amount of energy locally, on-site, they can sell surplus back to the grid and share electricity with their continental neighbors. 5. Transitioning the transport fleet to electric plug-in and fuel cell vehicles that can buy and sell electricity on a smart, continental, interactive power grid. From these pillars, Rifkin predicts, will grow an “energy internet” in which green energy will be produced in homes, offices, and factories and shared the same way that information is now shared online. Related to Rifkin’s vision and advising of the EU is Coutard and Rutherford’s (2011) observation that “we are witnessing an unprecedented critique of the extensive networked infrastructures built over the past 150 years for the provision of essential services such as water, sanitation, electricity, and heating” (106). In response to this critique, alternative, smaller-scale technological systems are developing that many view as more sustainable. The authors label this movement “post-networked urbanism.” Table 12.4 presents Coutard and Rutherford’s characterizations of features of large centralized networks of infrastructure and those of the alternative “sustainable techno-ecocycle approach” to infrastructure. The sustainable techno-ecocycle approach emphasizes local provision of infrastructure: that is, autonomy from large-scale infrastructure such as electricity and water and sewer. Coutard and Rutherford suggest four forms of organization for the sustainable techno-ecocycle approach: (1) off-grid, whereby homes, businesses, and even municipalities provide their own utility infrastructure needs; (2) feedinto-grid, whereby locally generated electricity must be purchased by the large centralized power generator; (3) loop-closing, whereby locally produced water and energy is also treated, recycled, and reused; and (4) “beyond net” (beyond

336

Nancey Green Leigh

Table 12.4 Contrasted (Opposed) Paradigms: Large Technical Networked System Versus Sustainable “Techno-ecocycle” Large Centralized Network

Sustainable Techno-ecocycle

Linear metabolism: tapping, supply, disposal Decoupling between local resource availability and use Territorial solidarity Technical systems Flows, imperviousness, kinetics Hydraulics-based model Supply-side model Economics of expansion and growth (scale, scope, club) Unbounded, ever-growing consumption Sector-based, sequential management Irreversibility, obduracy, “momentum” Carbon dependent

Circular metabolism: recycling, reuse, retrieval (Re)coupling between local resource availability and use Territorial autonomy Ecological systems Stocks, porosity, stasis Resource-based model Demand-side model Economics of preservation Bounded consumption Cross-sector, integrated management Reversibility, adaptability Carbon neutral

Source: Coutard and Rutherford (2011), table 8.1. Reprinted by permission of the publisher.

the network), whereby users have their own wells and septic/treatment systems. Post-networked infrastructure systems may divert financial resources from incumbent systems and may lead to increasing reluctance to finance large systems. This could set up a vicious cycle of declining support, quality, and reliability of large infrastructure systems that, in turn, raises concerns for “the financial sustainability of the incumbent infrastructure, which in most cases will remain of crucial importance, if only as a last-resort supply system, if we are to avoid a ‘tragedy of the infrastructural common’ ” (122).

Conclusions The efforts to strengthen urban manufacturing and associated distribution/ logistics activity coincide with shifting trends in industrial infrastructure and conversion pressures on urban industrial land. These efforts also coincide with the predicted looming crisis in infrastructure and with predictions of revolutionary shifts in how manufacturing occurs and how infrastructure systems are deployed. Thus, the relationship between urban development patterns, industrial land, and strengthening urban manufacturing is complex and evolving. However, urban policy development and planning practice efforts to decipher this relationship

strengthening urban industry

337

must acknowledge the conflict that exists today: present-day efforts to increase population density and promote more effective use of infrastructure to support that density contribute to industrial sprawl by displacing industry and extending grey infrastructure further out in metropolitan areas. We must begin with this little-acknowledged fact to ensure that future transitions in industrial activity and the provision of infrastructure are effective in containing sprawl and lowering urban carbon footprints. In the United States, due to the failure to finance maintenance needs, older infrastructure systems are deteriorating and failing. States and localities pay more than 90 percent of government non-defense capital outlays. Even before the Great Recession shrank the public revenue sources for making capital improvements, experts observed that elected officials were reluctant to either raise taxes or create new user fees to pay for infrastructure maintenance. With high unemployment rates and declining income for most U.S. households, that reluctance is likely to be even stronger. In the near to medium term, urban manufacturing and the economy overall are faced with looming problems for productivity and competiveness due to the failure to invest in all forms of critical infrastructure. Further, the economy continues to experience a very weak recovery from the Great Recession. Job creation numbers continue to fall significantly below expectations. While overall unemployment was around 8 percent in late 2012, unemployment for construction workers exceeded 14 percent. Construction workers would primarily be engaged in rebuilding the nation’s infrastructure. While state and local governments typically pay more than 90 percent of the cost of this infrastructure, their ability to do so has been severely hampered by declining tax revenues. A federal government Keynesian response would seriously increase public investment in national building programs, make a significant contribution to solving the national infrastructure crisis, and help strengthen urban manufacturing. Although not proposing a substitute for a stronger federal government response to the infrastructure crisis, this chapter has highlighted local government efforts that make better use of existing infrastructure and foster new forms of lower-impact infrastructure. These efforts show promise for the retention and intensification of urban manufacturing and other industrial activity, along with supportive residential and commercial uses. This chapter began by noting the premises on which is based the need to give explicit attention to strengthening urban industry and retaining urban industrial land. It should be clear that these premises flow from the reality that the urban land market suffers from many distortions. Government subsidizes suburban and exurban development by providing reduced property tax and other incentives and by subsidizing new roads and other new infrastructure. The historic bias of funding new infrastructure rather than repairing and maintaining infrastructure is undisputed. Additionally, the urban land market does not reflect the full cost of converting agricultural or greenfield land to developed land.

338

Nancey Green Leigh

Hence, compensating efforts to strengthen urban industry and maintain central city industrial land are needed. These efforts offer multiple benefits. They will help mitigate climate change by reducing urban expansion pressures and by incorporating low-impact infrastructure standards that can reduce long-run costs of infrastructure provision. They can help address the chronic problems of unemployment and poverty that frequently prevail near central city industrial areas. And finally, strengthening urban industry is essential for raising the United States’ level of exports and its leadership in advanced manufacturing in the global economy.

references American Planning Association. 2003. Smart growth audits. PAS No. 512. Prepared by J. Weitz and L. Waldner. ———. 2009. Smart codes: Model land-development regulations. PAS No. 556. Prepared by M. Morris. American Society of Civil Engineers. 2011a. Failure to act: The economic impact of current investment trends in electric infrastructure. Washington, DC: American Society of Civil Engineers. ———. 2011b. Failure to act: The economic impact of current investment trends in surface transportation infrastructure. Washington, DC: American Society of Civil Engineers. ———. 2011c. Failure to act: The economic impact of current investment trends in water and wastewater treatment infrastructure. Washington, DC: American Society of Civil Engineers. Atkinson, R. D., and P. D. Gottlieb. 2001. The metropolitan new economy index. Washington, DC: Public Policy Institute. Berube, A. 2007. Metro nation: How U.S. metropolitan areas fuel American prosperity. Washington, DC: Brookings Institution, Metropolitan Policy Program. www.brookings.edu/~/media/research/files/reports/2007/11 /06%20metronation%20berube/metronationbp. Boston Redevelopment Authority. 2001. Economic development initiative: Industrial fact sheet. www.bostonredevelopmentauthority.org/. CDC Associates. 2008. Making it green in Minneapolis–St. Paul. Mayor’s Initiative on Green Manufacturing. www.bluegreenalliance.org/news/publications/document /0003.4.pdf. City of Los Angeles. N.d. Low impact development. www.lastormwater.org/wp -content/files_mf/lid_howdoesitapplytoyou.pdf. City of Philadelphia. 2011. Draft Philadelphia 2035: Comprehensive plan. www.phila .gov/cityplanning/pdfs/ CitywidePlan_Draft_02152011_opt_A.pdf (February). City of Portland. 2003. Portland harbor industrial lands study: Part two: Interviews and analysis. Portland Bureau of Planning. www.planning.ci.portland.or.us (February). City of San Jose. 2009. City of San Jose general plan amendments affecting the industrial land supply: 1990–2009. www.sanjoseca.gov/ planning/gp/special_study /Industrial_Conversions_Since_1990.pdf.

strengthening urban industry

339

———. 2012. North San Jose area development policy. www.sanjoseca.gov/planning /nsj/docs/NSJ_Policy_Feb2012rev.pdf. Congress for the New Urbanism and U.S. Environmental Protection Agency. 2002. Smart scorecard for development projects. Prepared by W. Fleissig and V. Jacobsen. Coutard, O., and J. Rutherford. 2011. The rise of post–networked cities in Europe? In Cities and low carbon transitions, ed. H. Bulkeley, V. C. Broto, M. Hodson, and S. Marvin. New York: Routledge. Dempwolf, C. S. 2009. An evaluation of recent industrial land use studies: Do theory and history matter in practice? www.arch.umd.edu/downloads/pdfs/research/PG _industrial_land/scottdempwolf.pdf. De Sousa, C. 2000. Brownfield redevelopment versus greenfield development: A private sector perspective on the costs and risks associated with brownfield redevelopment in the Greater Toronto Area. Journal of Environmental Planning and Management 43(6):831–853. ———. 2002. Measuring the public costs and benefits of brownfield versus greenfield development in the Greater Toronto Area. Environment and Planning B: Planning and Design 29:251–280. Economist. 2012. Manufacturing: The third industrial revolution. www.economist .com/node/21553017 (21 April). Ezell, S. J., and R. D. Atkinson. 2011. The case for a national manufacturing strategy. Washington, DC: Information Technology and Innovation Foundation. www.itif .org/publications/case-national-manufacturing-strategy (April). Fitzgerald, J., and N. G. Leigh. 2002. Economic revitalization: Cases and strategies for city and suburb. Thousand Oaks, CA: Sage. Helper, S., T. Krueger, and H. Wial. 2012. Locating American manufacturing: Trends in the geography of production. Washington, DC: Metropolitan Policy Program at Brookings. www.brookings.edu/~/media/research/files/reports/2012/5 /09%20locating%20american%20manufacturing%20wialh/0509_locating _american_manufacturing_report.pdf. International City/County Management Association and Smart Growth Network. 2002. Getting to smart growth I: 100 policies for implementation. ———. 2003. Getting to smart growth II: 100 more policies for implementation. International Economic Development Council. 2006. Economic development and smart growth. Istrate, E., and N. Marchio. 2012. Export nation 2012: How U.S. metropolitan areas are driving national growth. Washington, DC: Brookings Institution, Metropolitan Policy Program. Istrate, E., J. Rothwell, and B. Katz. 2010. Export nation: How U.S. metros lead national export growth and boost competitiveness. Washington, DC: Brookings Institution, Metropolitan Policy Program. www.brookings.edu. Leigh, N. G. 1996. Fixed structures in transition: The changing demand for office and industrial infrastructure. In The transition to flexibility, ed. D. C. Knudsen, 137–154. Norwell, MA: Kluwer Press. ———. 2008. Oral and written testimony on re-authorization of the national brownfields legislation, before the U.S. House Subcommittee on the Environment, 14 February.

340

Nancey Green Leigh

———. 2009. Brownfield redevelopment: The practice of local government. Washington, DC: ICMA Press. Leigh, N. G., and N. Hoelzel. 2012. Smart growth’s blind side: Sustainable cities need productive urban industrial land. Journal of the American Planning Association 78(1):87–103. Leigh, N. G., K. Driemier, N. Hoelzel, R. Jain, J. Mansbach, E. Morrow, C. Moseley, S. Stevens, and E. Zayas. 2009. A plan for industrial land and sustainable industry in the City of Atlanta: Final report. Prepared for City of Atlanta and Atlanta Development Authority. Atlanta: Georgia Tech Research Corporation. http://smartech .gatech.edu/xmlui/handle/1853/35791. Marshall, A. 1961. Principles of economics: An introductory volume, 9th ed. London: Macmillan. McCurry, J. W. 2012. Fuel costs now drive D.C. site decisions. Site Selection. www .siteselection.com/issues/2012/jan/logistics-hubs.cfm (January). Miami21. 2012. Amended zoning code. www.miami21.org/final_code_April2012.asp. National Association of Local Government Environmental Professionals and Smart Growth Leadership Institute. 2004. Smart growth is smart business: Boosting the bottom line and community prosperity. Port of Portland. 2004. Brownfield/greenfield development cost comparison study. www.portofportland.com/PDFPOP/Trade_Trans_Studies_Brnfld_Stdy_Exec_Smry .pdf. Pratt Center for Community Development. 2009. Issue brief—Protecting New York’s threatened manufacturing space. New York. http://prattcenter.net/issue-brief /protecting-new-yorks-threatened-manufacturing-space (April). President’s Council of Advisors on Science and Technology. 2012. Report to the President on capturing domestic competitive advantage in advanced manufacturing. Washington, DC. www.whitehouse.gov/sites/default/files/microsites/ostp/pcast _amp_steering_committee_report_final_july_17_2012.pdf. Rifkin, J. 2012. The third industrial revolution. Making It (14 February). www .makingitmagazine.net/?p=4514. San Francisco. 2008. Central waterfront area plan. www.sf-planning.org/ (December). Smart Growth Leadership Institute. 2007. Implementation tools. Smart Growth Network, International City/County Management Association, and U.S. Environmental Protection Agency. 2006. This is smart growth. The Henry Ford. N.d. The living roof. The Henry Ford. www.thehenryford.org/rouge /leedlivingroof.aspx. Trade Promotion Coordinating Committee. 2010. Renewable energy and energy efficiency export initiative. www.export.gov/reee/eg_main_023036.asp. U.S. Congress Joint Economic Committee. 2010. Understanding the economy: Promising signs of recovery in manufacturing. http://jec.senate.gov (August). U.S. Department of Energy. 2009. 2009 electric transmission congestion study. http:// energy.gov/oe/downloads/2009-electric-transmission-congestion-study. U.S. Environmental Protection Agency. 2009. Smart growth guidelines for sustainable design and development. Prepared by Jonathan Rose Companies and Roberts and Todd.

commentary Alain Bertaud Nancey Green Leigh’s chapter is a passionate advocacy for the conservation of industrial land in urban areas and for the creation of new industrial areas closer to the city core. Leigh attributes the decline of U.S. manufacturing primarily to a generalized land market failure and to delusive planning ideologies like smart growth. Leigh also blames rigid zoning regulations and deficient infrastructure for the decrease in manufacturing in the United States. Although it includes many references to the economic and planning literature, the chapter is definitely contrarian compared to the current trends in land use economics and in planning practice. Ironically, the Asian cities to which much of the outsourced U.S. manufacturing has relocated have been practicing for the last 20 years the type of policy that Leigh deplores: the removal of traditional industrial sites from city centers into faraway suburbs or into smaller towns in order to shift pollution and improve industrial productivity. The impressive supply chains that provide a decisive comparative advantage to Chinese cities’ factories have not been built on obsolete centrally located industrial land—the central planners in Chinese cities created an abundance of it—but in new large, low-density industrial areas in distant suburbs or in smaller towns. The industrial clusters dispersed along the Pearl River Delta or along the Tianjin-Tanggu corridor have exactly the type of spatial structure that Leigh blames for the demise of American industries. The chapter recommends government supply-side subsidies to rejuvenate manufacturing in the United States, in spite of what is described as weak demand for industrial land: high vacancy rates and low rent levels in the existing industrial estates that have survived relocation. The chapter’s arguments can be divided into three main topics: (1) urban industrial land is important for macroeconomic development; (2) land market’s failure, smart growth doctrines, and inadequate zoning regulations are responsible for the decrease of U.S. urban industrial land; and (3) deficient U.S. infrastructure reduces manufacturing productivity and competitiveness.

Urban Industrial Land Leigh’s lead argument is that the U.S. economic recovery depends on preserving and reinforcing the manufacturing sector because of its ability to create jobs, to boost exports, and to provide high wages for a large number of unskilled workers. In addition, government intervention to maintain industrial areas in or close to urban cores would have a significant impact in decreasing global warming. The chapter implies that local land use issues have macro implications on exports, poverty, and climate change. These issues obviously cannot be dealt with 341

342

Alain Bertaud

efficiently at the local level and therefore should be elevated to the national policy level. Given that land use in the United States is largely a state and local government responsibility, it is not clear what leverage the federal government would have to constrain local market forces and local land use regulatory failures. If the arguments that Leigh develops are correct, then the federal government could intervene for the conservation of industrial land in the same way that it intervenes for the preservation of wetlands or for the protection of endangered species. The argument that the decline in U.S. manufacturing is mostly due to land market failures and in particular misconceived smart growth policies seems a little far-fetched. The decline in manufacturing jobs as part of total employment has been a secular trend whose causes seem to be mainly due to an increase in productivity (Malpezzi 2011). Could maintaining industrial land in inner-city areas have an impact on poverty? Leigh’s argument that “the central cities of metropolitan regions are also interested in preserving their industrial lands as a means to reduce high poverty and unemployment rates” and that “the older working-class neighborhoods that are adjacent to many inner-city industrial areas often have the greatest concentrations of poverty” implies that the maintenance or provision of industrial land in inner cities would help create jobs for the urban poor. This supply-side argument is difficult to believe in view of the weak demand trends shown in figure C12.1. Having given up on markets, Leigh seems to imply that maintaining industrial parcels close to city centers through government fiat could have the added benefit of reducing global warming. She writes, “Hence, compensating efforts to strengthen urban industry and maintain central city industrial land are needed. These efforts offer multiple benefits. They will help mitigate climate change by reducing urban expansion pressures and by incorporating low-impact infrastructure standards, which can reduce long-run costs of infrastructure provision.” It is true that locating industries close to city centers might reduce trip length for workers, suppliers, and customers. However, this can be done only by displacing denser commercial and residential areas, therefore lengthening the trips of many more users. The argument that maintaining industrial areas close to city centers would reduce greenhouse gas emissions for transportation does not hold, unless of course the industrial areas are old-fashioned sweatshops with very high worker densities. The type of land use that Leigh advocates is strongly reminiscent of the industrial belts created in the former USSR around the Moscow and St. Petersburg city cores while dense residential housing projects were pushed into the periphery. The possible increase in global warming caused by misconstrued local urban development policies discriminating against industrial areas is addressed in the following section.

commentary

343

Figure 12C.1 U.S. Manufacturing Employment Trends, 1939–2011 20,000

45 40 35

15,000

25

Percentage

Workers

30

20

10,000

15 10 5,000 1939 1941 1943 1945 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011

5

Thousands of workers

Percentage of total employment

Source: U.S. Bureau of Labor Statistics; Malpezzi (2011).

Land Market’s Failure, Smart Growth Doctrines, and Inadequate Zoning Regulations Figure 12C.1 The chapter argues that industrial land use is decreasing in urban areas because of Lincoln_Ingram_Infrastructure market failure, the smart growth movement, and inadequate zoning categories. According to Leigh, the main impacts of market failure have consisted of the abusive conversion of industrial land into other land use and the displacement of industrial zones from central areas toward the periphery. However, the evidence given for market failure is not convincing. For instance, Leigh acknowledges that “a key reason for the conversion of industrial land has been high vacancy rates.” Another reason has been that “when industrial properties need retrofitting . . . the low rental rates of these properties make it difficult to pay for renovations.”

344

Alain Bertaud

High vacancy rates and low rents seem to show a healthy market reaction to decrease in demand rather than a sign of market failure. The chapter usefully describes the current trend in modern industrial layout as very land intensive. It is not surprising, then, that industrial land-intensive activities—few workers per hectare, low floor area ratio—are pushed by market forces to the far periphery of cities to be replaced by activities that are better able to substitute capital for land—high floor area ratio—and have a higher density of people or workers per hectare. The industrial sweatshops of the past could maintain themselves in or close to city centers precisely because of the high density of workers per unit of land and therefore the low area of land per worker. This is apparently no longer the case for modern factories, where robots are more numerous than workers. The smart growth movement has never been very coherently applied to cities, and while smart growth planners may wish for high densities, these cannot be achieved in the absence of market demand. Leigh notes that “supply-side and demand-side economic development strategies focused on real estate, retail, services, and housing development [were] largely framed within the smart growth movement. The acceptance of this viewpoint led to substantial urban industrial land rezoning in some cities and pressured industrial businesses into leaving urban industrial districts by raising rents and land values.” The smart growth movement does not have the power to raise either rents or land value; however, the market does have this power, provided there is an alternative competing higher and best use. It is also difficult to blame a local authority for changing zoning regulations to facilitate the replacement of an obsolete industrial area with an alternative use that will provide more tax revenue. The welfare effect of such a change for the whole urban community is evident. Post–World War II manufacturing suburbanization was characterized by large-footprint warehouses with low worker densities. Leigh foresees a change where some new industries will have a smaller footprint: “New emphases on research and development activity, as well as incubators for advanced manufacturing, often coming out of urban universities, can also spur demand for central city industrial facilities.” Leigh argues that most of the negative externalities traditionally linked to industrial use do not exist anymore, at least for some types of industries: “While historical zoning practice established the separation of industrial uses from other uses for health and human safety, most of today’s industry does not require such separation.” It is true that most current zoning regulations are terribly unimaginative and that the industrial label immediately raises red flags in public hearings about possible zoning changes. It is quite possible that new potential Silicon Valleys are being nipped in the bud by obsolete zoning regulations, to the detriment of the U.S. economy. It would have been interesting to see this topic further developed in the chapter. If this argument is true, then we have a clear regulatory failure. However, the location of these new industries, with very few negative environ-

commentary

345

mental externalities, would be better allocated by the market once the regulatory barriers have been removed.

Deficient U.S. Infrastructure Leigh argues convincingly that deficiencies in current U.S. infrastructure decrease the productivity and financial viability of manufacturing. The Global Competitiveness Report, 2011–2012 (World Economic Forum 2011) ranked U.S. infrastructure 24th although the United States ranked fifth in global competitiveness. “In a series of reports issued under the main title ‘Failure to Act,’ the American Society of Civil Engineers . . . explored the economic impact on industry of chronic underinvestment in three of the five key areas of infrastructure: surface transportation, water and sewer, and electricity,” Leigh writes. “The report on surface transportation infrastructure . . . noted that in 2010 the United States ranked 19th out of the top 20 countries for quality of roads, and 18th out of the top 20 for quality of railroads. In particular, the report quantified the negative impacts to the United States’ ability to export if the surface transportation deficiencies are not corrected.” This is a strong point that could be further developed and might have a bearing on the location of manufacturing in the long run. Leigh’s chapter, in its passionate advocacy for the conservation of industrial land use, proposes supply-side stimulants, while simultaneously and convincingly exposing the current weakness of U.S. manufacturing’s demand side. This weakness is confirmed by secular trends in the percentage of total employment. The chapter has two strong points: t

t

Exposing the rigidity of current urban zoning in accommodating new forms of modern manufacturing activities that do not have the negative environmental externalities linked with traditional manufacturing. After passing the zoning hurdle, it is quite possible that these new kinder, gentler types of industrial activities might be able to compete on the market for central locations. Pointing to the negative impact on manufacturing productivity of letting the U.S. infrastructure further deteriorate.

However, I am puzzled by the author’s readiness to abandon market land price signals as a major mechanism in organizing the spatial structure of cities. Moving low-density employment areas to the fringe of cities does not augment sprawl; maintaining them in core areas does. I am also puzzled by Leigh’s advocating that local governments should subsidize land cleanup cost and even land cost to maintain manufacturing activities in areas where other better and higher uses would occur without subsidies. (“Because the private sector has little incentive to redevelop industrial-zoned brownfields as industrial properties, the role

346

Alain Bertaud

of the public sector is critical for strengthening urban manufacturing and export activity.”)

references Malpezzi, S. 2011. A primer on real estate and the aggregate economy: Know your macro indicators. Madison: Wisconsin School of Business. World Economic Forum. 2011. The global competitiveness report, 2011–2012. Geneva: World Economic Forum.

13 What Is the Value of Infrastructure Maintenance? A Survey Felix Rioja Good roads, canals, and navigable rivers, by diminishing the expense of carriage, put the remote parts of the country more nearly upon a level with those in the neighboring town. They are upon that account the greatest of all improvements. adam smith, The Wealth of Nations (1776)1

P

ublic infrastructure has been established as the foundation for the productive activities of a country. Road and rail networks, water systems, power generating and distribution systems, and telecommunications are essential inputs for an economy’s production of goods and services. Of course, it matters not only how much public infrastructure a country has, but what condition the infrastructure is in. Infrastructure wears out with time and use, so proper and timely maintenance must be periodically conducted. According to the U.S. Congressional Budget Office (CBO), operations and maintenance expenditures are those that are “generally required to provide the services needed for infrastructure to function and that are often necessary for the repair and safe operation of existing infrastructure” (Congressional Budget Office 2007). Neglecting proper maintenance leads to a decline in infrastructure’s condition: pothole-filled roads, loss of irrigation water, power outages, dropped phone calls, and so on. In the

I would like to thank Karin Brandt, Jacob Greenstein, Gregory Ingram, Vito Tanzi, Waheed Uddin, and the conference participants at the 7th Annual Land Policy Conference: Infrastructure and Land Policies, sponsored by Lincoln Institute of Land Policy, June 4–5, 2012, for helpful comments and suggestions. I am also grateful to Maria Bernedo and Fernando RiosAvila for their research assistance.

347

348

Felix Rioja

short run, infrastructure in bad condition imposes costs on users. In the long run, failure to maintain infrastructure in a timely fashion leads to greater costs of rebuilding. Consider the following examples of costs imposed on users. First, as estimated by the American Society of Civil Engineers (ASCE) in 2009, bad road conditions in the United States impose a cost on motorists of $67 billion annually. The ASCE study assigned grades to various types of infrastructures. Five sectors received a grade of D minus: drinking water, inland waterways, levees, roads, and wastewater. Second, to see how private producers can be affected by maintenance neglect, consider the following example from Zambia reported by Heggie (1995). In 1992, the Federation of Zambian Road Hauliers commissioned a study on the effects of bad road conditions on a vehicle’s operating costs. One set of vehicles (a truck and a tractor-trailer) traveled for one year along a road in good condition. Another set of vehicles traveled along a pothole-filled road in bad condition. The vehicles were delivering products to the market or bringing needed materials for production. The study compared the costs of repairing shocks, springs, brake shoes, clutch, and so on. The additional costs are detailed on table 13.1. At the end of the year, the company had spent about $14,000 more on repairing the vehicles using the pothole-filled road. Table 13.2 presents data on the condition of infrastructure around the world. The first data column shows the transmission and distribution losses of electrical power as a percentage of total output. Countries are grouped according to the World Bank’s (2008, 2011) classification system. In the Low Income countries group, 22 percent of power is lost. Middle Income countries fare a little better, with 11 percent of power lost. Both groups, however, have higher losses than the High Income countries, where the power loss is only on average 6 percent. Table 13.1 Additional Operating Costs Due to Using a Road in Bad Condition Item Tires and tubes Clutch and pressure plate Wheel bearings Set of brake shoes Set of springs Spring hangers and bushes Welding and steering assembly Shock absorbers Total extra costs

Quantity

Extra Annual Cost (US$)

10 1 4 1 1 4 — 4

5,952 1,071 803 1,050 1,667 452 2,826 510 14,331

Source: Adapted from Heggie (1995). Heggie reports on a study of the Federation of Zambian Road Hauliers Ltd., which describes the additional costs to a set of vehicles used to transport goods and materials along a pothole-filled road for one year.

what is the value of infrastructure maintenance?

349

Table 13.2 The Condition of Infrastructure Country Group

Low Income Middle Income Lower Middle Upper Middle Low and Middle Income East Asia and Pacific Europe and Central Asia Latin America and Caribbean Middle East and North Africa South Asia Sub-Saharan Africa High Income European area

Electrical Power

Telephones

Roads

Transmission and Distribution Losses (% of Output)

Faults per 100 Mainlines

Percentage Paved

22 11 9 13 12 7 12 16 17 24 9 6 6

8 22 8

10 24

8.3

21 54 49 45 62 86 22 79 54 19 81 87

Source: World Bank (2008, 2011). Electrical power data are for 2005. Telephone data are for 2006. Road data are for 2009.

A similar picture emerges from loss indicators in telephone communication, as measured by the number of faults per 100 mainlines. The Lower Middle Income group and the Middle Eastern and North African countries have on average 22 to 24 faults per 100 mainlines. Conversely, Upper Middle Income countries only have about eight faults, which is two-thirds less. Table 13.2 also presents the percentage of paved roads in these country groups, although this is not explicitly a measure of infrastructure condition. An article by Reinikka and Svensson (2002) further illuminates how private firms react when the infrastructure is in poor condition. The authors studied data at the firm level in Uganda. They found that, due to disruptions in infrastructure services (which they proxy by an unreliable and inadequate electric power supply), the firms themselves attempt to fill this void by investing in power generators, waste disposal equipment, and so on. As a result of these extra expenditures, however, the firms reduce their investment in noninfrastructure capital needed for their productive activities, and this reduction in investment leads to a decrease in productive capacity. The long-term consequence of neglect is the high cost of major reconstruction of infrastructure. According to the World Development Report (WDR) (World

350

Felix Rioja

Bank 1994b), in sub-Saharan Africa about $13 billion worth of roads built in the 1970s and 1980s had eroded due to deficient maintenance. In Latin America the report estimated that every $1 not spent on maintenance will eventually cost $3 to $4 in premature reconstruction. A well-maintained road should last 10 to 15 years before it needs to be resurfaced. The lack of maintenance can cause severe deterioration requiring resurfacing in as little as five years. In the case of power lines, the expenditure of $1 million to reduce power line losses could save $12 million in generating capacity (World Bank 1994b). Often new construction takes place while the maintenance of existing infrastructure is neglected. Between 1979 and 1984, for example, 6,000 kilometers (km) of paved road were built in Brazil. During this same period, another 6,000 km of roads declined from “fair” to “poor” condition, and 2,000 km went from “good” to “poor” due to maintenance neglect (Harral 1988). In the United States, the National Surface Transportation Policy and Revenue Study Commission (2008) recommended spending between $225 and $340 billion annually to maintain the surface transportation network. Put in perspective, these figures are between one-third and one-half of the size of the American Recovery and Reinvestment Act (the fiscal stimulus) of 2009. The condition of infrastructure is very much related to the achievement of the Millennium Development Goals set by the United Nations (Estache 2004). For example, infrastructure in good condition can play a key role in reducing poverty and improving access to education and to health care facilities. According to Estache and Fay (2007), maintenance needs have been estimated to be between 1.5 percent and 3.3 percent of gross domestic product (GDP) for developing countries, yet most developing countries spend much less than this. Not only is actual maintenance of infrastructure neglected, but also the study of maintenance as a topic has been somewhat neglected, perhaps because data on infrastructure maintenance are hard to come by, while data on new construction are readily available. This chapter reviews several issues associated with infrastructure maintenance. It does not attempt to be encyclopedic on all maintenance-related issues, but simply tries to survey some of the salient issues. Four main questions are discussed. First, what theoretical framework explains the channels through which maintenance is effective, and what determines its optimal level? Second, what do the available data on maintenance tell us about its economic rate of return? Third, what has the empirical evidence found on the effects of maintenance on productivity and growth? Fourth, how have countries funded maintenance expenditures? Finally, an interesting case study from Peru is discussed, which provides a creative approach to the maintenance of rural roads.

Theoretical Framework The study of the role of infrastructure in growth models goes back to at least Arrow and Kurz (1970). Barro (1990) wrote one of the seminal papers introduc-

what is the value of infrastructure maintenance?

351

ing productive public expenditures into a growth model in which public capital is an essential input in the production function. About the same time, Aschauer (1989) found large returns from investment in infrastructure. This survey will follow Rioja’s (2003a) framework, which introduces infrastructure maintenance into a neoclassical growth model and applies it to a developing country scenario. By introducing maintenance and exploring its role, the model extends previous theoretical work on public capital by Barro (1990) and Glomm and Ravikumar (1994), among others.1 The economy has many firms that operate in a competitive market and produce a final good, yt, according to the following production function: yt
¦ (KGt , kt , lt ). where kt is private capital, lt is labor, and KGt is the stock of public infrastructure, which is a government-provided input.2 The production function f exhibits constant returns to scale to private inputs. Specifically, the calibration part of Rioja
 (1) kt lt . The elasticity of (2003a) uses a standard Cobb-Douglas form as yt  KGt output to public infrastructure is measured by the parameter , which has been estimated with various data, methods, and countries. (For a summary and metaanalysis, see Bom and Ligthart 2009.) The government taxes firm output at rate !t, so firms maximize net-of-tax profit, t  (1 -
t ) ¦ (KGt ,kt ,lt ) - wt lt - rrkt. The government collects tax revenue equal to !tyt, and in this simplified framework all of the revenue is spent on maintenance of infrastructure mt, so that !tyt " mt. Maintenance expenditures, in turn, affect the depreciation rate of infrastructure, which is endogenous:


Gt (mt , kt ). Hence, this depreciation rate, #Gt, depends on how much is spent on maintaining the infrastructure (mt) and on how much the infrastructure is used, which is proxied by kt.3 The depreciation rate is reduced with more maintenance expenditures, and it increases with more use.4

1. Papers by McGrattan and Schmitz (1999), Schmalensee (1974), and Feldstein and Rothschild (1974) studied how private firms choose optimal maintenance of their capital stock. 2. Feehan (1998) states that a consensus in the public inputs literature has emerged to model productive public inputs (e.g., infrastructure) as factor augmenting. 3. This specification is similar to McGrattan and Schmitz’s (1999). Alternatively, in Kalaitzidakis and Kalyvitis (2004), usage is proxied by output so their depreciation rate is a function of maintenance as a share of GDP, #Gt(m/y). 4. Another way that maintenance can affect this model is developed in Rioja (2003b), in which it also affects the effective stock of infrastructure.

352

Felix Rioja

Given that Rioja’s (2003a) model is focused on developing countries, new infrastructure construction is financed entirely by foreign donations or aid in amount Dt. Aid and concessional loans are in the real world the most common ways of financing new public projects.5 Public infrastructure then accumulates as follows: KGt1
 Dt  (1  Gt (mt , kt ))KGt. Donations from foreign countries finance new public infrastructure investments, so they augment the next period’s infrastructure stock. Reducing maintenance expenditures, however, increases the depreciation rate and reduces the next period’s stock of infrastructure. Households in the model maximize utility subject to budget constraints. For briefness of exposition, the household problem is not specified again (see Rioja 2003a). The government seeks to determine its optimal policy, that is, to choose the tax rate that maximizes social welfare. Hence, the government must choose the optimal level of maintenance to maximize well-being. The details are specified in Rioja (2003a). The optimal tax rate, which is also the optimal share of GDP that should be devoted to maintenance, is found to depend on the following: D ö æ l *t = g ç
, other parameters, kt , KGt , t ÷ . è KGt ø Two points about this optimal solution are worth emphasizing. First, the higher the ratio of donations to existing infrastructure (Dt /KGt), the lower the optimal rate of maintenance will be. The intuition is that if international donors raise the amount they spend on building new infrastructure, the home government finds it optimal to reduce its maintenance expenditures for existing infrastructure. The government in the model cares about the total stock of infrastructure that raises output and, hence, consumption and the welfare of the population. As donors fund more and highly visible projects, they add to the infrastructure stock, and it is optimal to reduce maintenance expenditures, which have to be raised by painful taxation. This point is certainly born out in the Brazilian example discussed earlier (Harral 1988), where 6,000 km of new paved roads were built, while at the same time another 6,000 km declined from “fair” to “poor” condition and 2,000 km went from “good” to “poor” condition. Second, the solution for optimal maintenance share implies that the more productive infrastructure is (i.e., the higher !), the more the government should spend on maintaining it.

5. Even when some of the funds to finance new infrastructure are borrowed from industrial countries’ governments or international organizations, some of this debt may effectively be forgiven later or never fully repaid. Of course, the model can be generalized in several ways with countries borrowing or using tax revenue to finance infrastructure.

what is the value of infrastructure maintenance?

353

Subsequent work has extended this framework in several directions. Kalaitzidakis and Kalyvitis (2004) present an endogenous growth model in which the government has to finance both maintenance and new infrastructure. They also study how maintenance may affect the impact of infrastructure on growth. While Barro (1990) had derived that the optimal share of GDP to be devoted to infrastructure should be equal to the elasticity of infrastructure in production, Kalaitzidakis and Kalyvitis (2004) find that, once the model accounts for maintenance, the optimal share of GDP devoted to overall public investment (infrastructure and maintenance) should be larger than this elasticity. In addition, they derive a nonlinear relationship between the ratio of maintenance to new investment and economic growth. Figure 13.1 illustrates this nonlinear relationship. Growth increases when the maintenance to new investment ratio increases up to an optimum, but growth decreases thereafter. Another recent contribution by Agénor (2009) further extends the framework of analysis of this issue by noting that maintenance could also affect the durability of private capital. The longevity of a firm’s machines, like the Zambian company’s trucks described in the introduction, can be affected when infrastructure is in bad condition. In summary, theoretical frameworks have modeled maintenance as affecting primarily the depreciation rate of public infrastructure. Neglecting maintenance Figure 13.1 Relationship Between Growth and the Ratio of Maintenance to New Investment Lack of maintenance investment

Lack of new infrastructure investment

Output growth rate

Maximum output growth rate

Maintenance investment New infrastructure investment Source: Kalaitzidakis and Kalyvitis (2005).

Figure 13.1 Lincoln_Ingram_Infrastructure

354

Felix Rioja

increases the depreciation rate and, hence, reduces the overall flow of services provided by the infrastructure network.

Maintenance Data ECONOMIC RATE OF RETURN

According to the WDR (World Bank 1994b), returns on maintenance on road projects were almost twice as much as those on projects that involved mainly new construction. This is a benchmark comparison that should be kept in mind when reviewing the returns on various maintenance projects. Most of the estimates of the rate of return for maintenance come from infrastructure projects that involved the World Bank and other multilateral institutions. In particular, the economic rate of return (ERR) is most commonly reported by these studies, a few of which are listed in table 13.3. The Congressional Budget Office (1988) also provided some estimates of the ERR for highway maintenance in the United States. The CBO calculates a range of 25 to 38 percent. Several different strategies were evaluated. One strategy was to sustain existing levels of maintenance expenditures. This strategy had the largest ERR. On the other end, a strategy that fixed all deficiencies involved the highest cost, which would exceed the reductions in transportation costs. An intermediate strategy of raising expenditures to achieve minimum standards yielded an ERR of about 29 percent. The remainder of ERR estimates in table 13.3 are from World Bank projects. For example, in the 1990s the World Bank helped finance the maintenance of roads in Bolivia (World Bank 2001). Maintenance was performed on 781 km of gravel roads and 765 km of paved roads in various areas of the country. The costs per km of maintenance were $43,000 for gravel roads and $97,660 for paved roads. The average ERR for the maintenance was 34 percent for gravel and 36 percent for paved. Another maintenance project financed by the World Bank in Jamaica consisted of (1) asphaltic concrete applied to existing pavement on about 160 miles of roads; (2) asphaltic sealing and resealing with double surface treatment of some 360 roads; and (3) drainage improvement and surface patching for (1) and (2). The periodic maintenance was justified on the basis of lower vehicle operating costs and time savings for road users resulting from improved surfaces and higher average speeds. The final project report (World Bank 1992) indicates that the ERR for the completed portion of the project was 42 percent. Table 13.3 lists several other similar road maintenance projects and their ERRs, which vary according to country but are on average fairly high. With a rough average of 30 percent, the maintenance of roads has a very high return. Of course, one question that arises is this: if maintenance has such a high ERR, why do countries not prioritize these expenditures? Typically, financing may be the issue. Governments have an easier time accessing lending (typically at concessional low rates) or grants from international institutions for the construction of new infrastructure than for maintenance. Unlike some of the projects

what is the value of infrastructure maintenance?

355

Table 13.3 Economic Rate of Return for Maintenance Projects Country

Source

Project Description

United States Bolivia

Congressional Budget Office (1988) World Bank (2001)

Gambia

World Bank (1994a)

Jamaica

World Bank (1992)

Burkina Faso

World Bank (1993)

Tunisia

World Bank (1990)

Nepal

World Bank (1986)

Study calculated the return of various federal highway maintenance projects. Maintenance plan for 1992–1995 that included periodic maintenance of 781 km of paved roads and 765 km of gravel roads. Road maintenance activities mainly comprised regraveling, resealing, and routine maintenance of national and local road networks and low-cost paving of gravel road sections. Project comprised periodic road maintenance that included asphaltic concrete applied to existing pavement on 160 miles of road and asphaltic sealing and resealing with double surface treatment of 360 roads. Only 70% of the project was completed. Four-year program (1982–1985) of road maintenance that consisted of graveling roads and routine maintenance. The Program for Highway Maintenance included expanding road maintenance activities and improving the efficiency of the maintenance organization. The Road Maintenance Project consisted of purchasing maintenance equipment, training mechanics, and providing technical assistance for maintenance programming and operations.

Economic Rate of Return (%) 25–38 34.4–35.9

23

42

50 (graveling) 100 (routine maintenance) 70

32.9

listed in table 13.3, which had financing from the World Bank, in most cases once the infrastructure is built, the government must raise the maintenance cost from its general revenue sources. Many of the resulting competing spending priorities cannot be delayed as they carry political costs. Maintenance can be delayed, however, so in times of budget tightening, infrastructure and maintenance may bear most of the burden, as described by Easterly, Irwin, and Servén (2008). More discussion on the funding of maintenance, in particular for roads, is presented in the following section.

356

Felix Rioja

EMPIRICAL EVIDENCE

Data on maintenance have not been collected in many countries or are not available due to issues identified by McGrattan and Schmitz (1999). There is no market transaction for much maintenance expenditure, so collecting data requires the surveying of public organizations. Furthermore, data may be recorded in different accounts in different countries, if they are recorded at all. One of the few known systematic collections of maintenance data is the Canadian Survey of Capital and Repair Expenditures. This survey has data on maintenance spending for government organizations and private firms. Kalaitzidakis and Kalyvitis (2005) provided the first study of the role of maintenance using these data. From 1956 to 1993, Canada’s total public infrastructure-related spending (new construction plus maintenance) was 7.4 percent of its GDP. Maintenance spending accounted for 1.5 percent of GDP, so it was about 21 percent of total public spending on infrastructure. Kalaitzidakis and Kalyvitis empirically tested the effect of maintenance on Canadian growth. Their findings are somewhat counter to expectations. Their results show that Canada would actually have benefited from a reduction in public infrastructure-related expenditures. These findings imply that at 7.4 percent of GDP, Canada’s public spending was too high. The authors calculate the optimal share at 6.7 to 7 percent of GDP, which would have yielded higher growth in Canada. Furthermore, the findings imply that most of the proposed reduction should come from a decrease in maintenance expenditures. Maintenance was 21 percent of total infrastructure-related public spending during the period studied. The authors find that this share should be reduced to about 14 percent. Kalaitzidakis and Kalyvitis’s findings can be also interpreted in the context of figure 13.1. Canada’s ratio of maintenance to new investment was to the right of the optimal share. Hence, reducing this share would yield higher growth. While the results for Canada are interesting, they are not necessarily representative. Many developing countries likely underinvest in new infrastructure and especially in maintenance. Furthermore, a study by Brox (2008) shows that infrastructure-related spending in Canada has recently fallen to half of its level in the 1960s and that maintenance has been neglected, leading to the crumbling of some infrastructure.6 Brox estimates that Canada would need about $72 billion for new projects and $123 billion for maintenance of existing infrastructure. Kalaitzidakis and Kalyvitis (2005) attempt to estimate the effects of infrastructure and of operations and maintenance (O&M) expenditures on total factor productivity (TFP) by accounting for potential spillover effects that infrastructure in one state may have on neighboring states. The spillover variables are weighted averages of infrastructure and O&M spending in neighboring

6. A notable example cited by Brox (2008) is the collapse of an overpass on the Boulevard de la Concorde in Laval in 2006.

what is the value of infrastructure maintenance?

357

states.7 Capturing the spillover effects involves estimating nonlinearities, which the authors do by using semiparametric estimation techniques. They find that O&M expenditures in one state have positive effects on a neighboring state’s TFP and that these effects are larger than the impact of within-state O&M expenditures. On average, a 1 percent increase in O&M spending by one state raises output in a neighboring state by about 0.4 percent. In addition, they find that the spillover effect of O&M is about eight times higher than that of capital expenditures. While these estimates are mostly confirmed in their sensitivity analysis, it is unclear why O&M has a much larger spillover effect than that of construction. In a related study, Kalyvitis and Vella (2011) look at the effects of public expenditures by different levels of government in the United States. The data, published by the Congressional Budget Office (2007), include infrastructurerelated spending at the federal, state, and local levels. The three levels of government share responsibilities in infrastructure provision and maintenance. Table 13.4 presents a summary of the data. In the United States, about 2.6 percent of GDP is devoted to infrastructure expenditures (with almost half of that going to transport infrastructure). Of the total devoted to infrastructure, O&M accounts for about 49 percent. As the table shows, state and local governments spend a larger share than does the federal government. One question the authors consider concerns where the marginal dollar of spending is most productive: maintenance or new investment? In addition, they Table 13.4 Capital and Operations and Maintenance Expenditures on Public Infrastructure in the United States, 1956–2004 Expenditure Capital and O&M (% GDP) O&M (% of total expenditure) Capital and O&M, transportation (% GDP) O&M, transportation (% of total transportation expenditure) Capital and O&M, water (% GDP) O&M, water (% of total water expenditure)

Federal

State and Local

Total

0.7 25.0 0.5

1.9 58.4 1.3

2.6 48.8 1.8

23.8 0.2 30.0

58.1 0.6 59.0

47.9 0.8 50.9

O&M = operations and maintenance; GDP = gross domestic product Source: Adapted from Kalyvitis and Vella (2011) using Congressional Budget Office (2007) data. Averages are presented for all variables.

7. These weights are designed to capture the degree of interdependence of each neighboring state. Hence, states are weighted by the flow of goods across states and alternatively by the relative size of their economic activity.

358

Felix Rioja

attempt to determine which level of government should spend the marginal dollar. According to public finance theory, some public goods may be more efficiently provided at the state or local level. The empirical results show that the productivity growth rate would not be very much affected by changing U.S. aggregate infrastructure spending or by shifting the allocation of funds between new investment and O&M. These results would imply, then, that U.S. expenditures have been close to optimal between 1956 and 2006. The disaggregated results by level of government, however, show that increasing infrastructure spending by states and localities, and especially increasing O&M, would have positive effects. Given that the national-level allocations are close to optimal, this means that any increases in the public expenditures at the federal level would then have a detrimental impact on productivity growth. A potential explanation of these findings is offered by Holtz-Eakin (1994), who proposes that because the federal government provided incentives to states and localities by offering matching grants for new construction of infrastructure, states and localities did not spend enough on maintenance.

The Funding of Maintenance Expenditures for Roads As described in the previous section, the rates of return for maintenance are typically very high, but adequately financing maintenance has presented challenges over the years. This section focuses on the financing of roads, which are one of the largest components of infrastructure. Various financing schemes have been tried over the years. One early approach to financing road maintenance, which the World Bank (2007a) calls “the budget approach,” involved drawing funds from the country’s general budget. The view was that since infrastructure was publicly owned, then the funding to maintain it should come from the government budget. One problem with this approach was that the fungibility of revenue sources implied that the priority placed on maintenance by the government du jour could vary from year to year. Often, funding was directed to many more pressing political priorities. The drawbacks of the budget approach were identified back in the 1970s and were described by the World Bank (1979). As a potential solution, the World Bank and the International Monetary Fund supported the creation of separate agencies, known as “road funds,” which many developing countries subsequently established. According to Gwilliam and Shalizi (1999), “first-generation road funds” were established in the 1960s and 1970s in Africa, Asia, and Latin America. These funds were “extrabudgetary arrangements through which an earmarked stream of tax revenues was put at the disposal of a road department or agency” (183). Typically, fuel taxes were the source of these tax revenues. However, first-generation road funds had several weaknesses and did not work well. One weakness, according to Heggie (2000), was the lack of oversight over these agencies and the lack of transparency in their activities. Cabinet ministers could borrow from the fund to fill an immediate need like paying the salaries of government workers. Government corruption

what is the value of infrastructure maintenance?

359

was found to be associated with lower operations and maintenance expenditures (Tanzi and Davoodi 1998; see also Tanzi 2005). Often, these agencies were not technically audited to check if the funds had been properly disbursed on approved projects. Another weakness was that some revenues were raised from activities unrelated to road use. In addition, the earmarking of revenues took away revenues from other sectors, creating inefficiencies. Having identified these and other weaknesses in the first generation of road funds, a second generation began to be established in the 1990s. According to Gwilliam and Shalizi (1999), these funds no longer use earmarked revenues; rather, “they are funded by levies or surcharges designated as ‘user charges’ and identified separately from general taxation” (161). Fuel taxes are still the main source of revenue for these funds. Typically, road funds are managed by a board composed of several members from the private and public sectors. This managing board identifies the projects needing maintenance and allocates funding. Then the board contracts out the necessary work in a transparent way. For example, Zambia’s road fund was established in 1994. The private sector is well represented on the board (7 out of 11 members), and the chairperson is also a private sector actor (Gwilliam and Kumar 2002). The makeup of the board varies in different countries, and in some places the minister of works becomes the chairperson. While many developing countries have established road funds, most western European countries use the budget approach, in which maintenance is a line item in the general budget (World Bank 2007a). This approach appears to work well in countries with strong institutional frameworks. However, a few industrialized countries, like the United States, Japan, and New Zealand, have had road funds for a long period. For example, the U.S. Highway Trust Fund, established in the 1950s, originally funded just highway construction and maintenance, but since 1983 it has also funded mass transit projects. How well have second-generation funds performed? The Independent Evaluation Group (IEG), an independent group within the World Bank, examined several reports produced by World Bank staff and summarized their findings (World Bank 2007a). The maintenance-related outcomes evaluated include the level of funding, the quality of the road work, and the efficiency of operations.8 According to the IEG, there has generally been an increase in the percentage of roads in good condition from year to year compared to before the establishment of a road fund. For example, in Zambia, the annual change in the percentage of roads in good condition was 4 percent, which is roughly the average for the countries considered. In terms of the level of funding, most countries for which data were available had more funds available for maintenance after they established a secondgeneration road fund. Also, the share of maintenance that was contracted out

8. Some of the studies and countries surveyed include Gwilliam and Kumar (2002)—seven African countries; Zietlow (2004)—six Latin American countries; and Benmaamar (2006)—27 African countries.

360

Felix Rioja

increased to as high as 90 percent in Zambia and Ghana. This resulted in an improvement in operational efficiency as the cost of maintenance per kilometer decreased by about 15 percent.9 Second-generation funds have not solved the maintenance issue, however. According to a survey administered by the IEG to World Bank transportation staff (World Bank 2007a), only about 40 to 50 percent of estimated maintenance needs were being covered. The remaining available funds were allocated to other uses, only some of which were related to transportation. Another issue identified in the IEG survey was that spending was focused on maintaining main roads at the expense of secondary roads. In summary, while second-generation funds have improved maintenance in many countries, much work remains to be done, as the state of roads, and infrastructure in general, is deficient.

Case Study: Peruvian Rural Roads Project One of the most interesting and successful case studies of infrastructure maintenance is the Peruvian Rural Roads Project described in Fay and Morrison (2007) and the World Bank (2007b). In Peru about half of the population is considered “extremely poor” (living on less than $1 per day), and the largest share of the poor population lives in rural areas. The lack of roads and the poor condition of roads have played a role in the population’s endemic poverty, as people living in these rural communities cannot access markets, jobs, and services. In 1995 the Peruvian government embarked on an innovative approach to road management. The World Bank and the Inter-American Development Bank helped with the financing of this project. The approach involved letting the rural communities set priorities for the roads to be rehabilitated. Community meetings were organized and plans drafted at numerous gatherings. The management of rural roads was decentralized over time, giving provincial municipalities authority, budget responsibility, and technical expertise. About 38 provincial road institutes were created with many local authorities as members. These road institutes then contracted the rehabilitation and maintenance of the roads to newly formed micro-enterprises comprised of some of the poorest members of the communities. In all, more than 500 micro-enterprises were formed, employing 5,700 workers (30 percent of whom were female). The timeliness and efficiency of maintenance improved significantly, while at the same time increasing entrepreneurial capacity and reducing poverty. About 13,000 km of roads were receiving adequate and timely maintenance by 2005. As a first step,

9. In addition, some countries have had success using “performance-based contracts.” Contractors agree to maintain roads so that they meet certain specified physical conditions. Hence, contractors have incentives to maintain and repair roads in an adequate and timely fashion to receive payments. See Zietlow (2005) for a summary of the experience of industrial and developing countries using performance contracts.

what is the value of infrastructure maintenance?

361

these mostly gravel roads were rehabilitated. Then, the maintenance involved “simple works regularly performed throughout the year to maintain the drainage systems (ditches, culverts, vegetation) and the running surface (filling potholes and ruts, maintaining the surface camber), supplemented from time to time with spot interventions to restore passage” (World Bank 2007b, 7). Given the efficiencies gained with this program, according to Greenstein (2012), the costs of routine road maintenance decreased from $1,200 per km to $750 per km. One evaluation of this project by Escobal and Ponce (2003) focuses on the question of the project’s effect on welfare: how did better-maintained roads affect household per capita income and consumption? Escobal and Ponce used a “propensity score matching” methodology that allowed them to compare the income of a household located near a rehabilitated road with an estimate of the income that the household would have earned had the rehabilitation not occurred. They found that the income of households near roads that were maintained was indeed higher ex post. In particular, these households were able to increase nonfarm income as transportation to towns became easier. Another effect of the Peruvian Rural Roads Project has been its effect on democracy and civic participation. Because the program gave the communities decision-making power over projects, it improved inclusion and participation. According to Remy Simatovic (2008), the rehabilitation and periodic maintenance of roads decreased the rural population’s costs of getting to towns or voting stations. This increased the rural population’s participation in the political process, which reached the levels of participation common in urban areas. Furthermore, many individuals who became involved with the project as community board members or as part of the micro-enterprises were elected to municipal management positions in later years (Remy Simatovic 2008). In summary, the Peruvian Rural Roads case may constitute a benchmark for many developing countries’ rural road networks, as it succeeded in rehabilitating and maintaining roads by employing many poor local workers, allowing their communities to participate in decisions affecting them, and expanding their economic possibilities.

Conclusions This chapter has reviewed research on several aspects of infrastructure maintenance. First, it reviewed how maintenance can play a role within standard growth models by reducing the depreciation of public infrastructure. These models show that an optimal level of maintenance expenditures can increase a country’s growth rate. Second, a survey of the economic rates of return in a variety of countries and projects found that estimated rates of return for maintenance are uniformly high. Third, empirical studies have found that maintenance can have a significant and positive effect on productivity and economic growth. Given these positive findings, it remains somewhat of a puzzle why maintenance has been neglected in industrial countries as well as in developing

362

Felix Rioja

countries. One possible reason has been inadequate financing. Different countries have experimented with various ways to finance maintenance. With regard to roads in particular, the creation of second-generation road funds has met some success by establishing dedicated funding sources (e.g., fuel taxes) and letting independent boards decide on the proper allocation of expenditures, hence improving efficiency and transparency. While these second-generation road funds have been an improvement, in many countries the amount and allocation of funds is still below optimal. Often, considerations of political economy have played a role in the insufficient funding for maintenance. At times, governments have diverted some of the funds designated for maintenance or have favored new infrastructure investments over the maintenance of existing infrastructure. That is, politicians may perceive that they receive more support when they complete new infrastructure investments than when they repair an ailing network. It has been easier for the governments of developing countries to access donor aid and concessional lending for new infrastructure construction, though more recently multilateral institutions have factored future maintenance expenditures into the amount of their contributions. The Peruvian Rural Roads Project is a notable success story. The project was designed to involve communities and hire local workers, improving the economic condition of the affected population. Various evaluations of its outcomes have found that the timeliness and efficiency of maintenance increased as a result of the project. In addition, the project led to job creation, higher incomes, and increased participation in the political process by the affected communities. Adapting and replicating this approach may be a way to improve maintenance and economic outcomes in other countries.

references Agénor, P.-R. 2009. Infrastructure investment and maintenance expenditure: Optimal allocation rules in a growing economy. Journal of Public Economic Theory 11(2): 233–250. American Society of Civil Engineers. 2009. Report card for America’s infrastructure. www.infrastructurereportcard.org. Arrow, K. J., and M. Kurz. 1970. Public investment, the rate of return, and optimal fiscal policy. Baltimore: Johns Hopkins University Press. Aschauer, D. A. 1989. Is public expenditure productive? Journal of Monetary Economics 23:177–200. Barro, R. 1990. Government spending in a simple model of endogenous growth. Journal of Political Economy 98:103–125. Benmaamar, M. 2006. Financing of road maintenance works in sub-Saharan Africa: Reforms and progress towards second generation road funds. Discussion Paper No. 6. Sub-Saharan Africa Transport Policy Program. Washington, DC: World Bank. Bom, P. R. D., and J. E. Ligthart. 2009. How productive is public capital? A metaregression analysis. Working Paper No. 0912. International Center for Public Policy. Atlanta: Georgia State University, Andrew Young School of Policy Studies.

what is the value of infrastructure maintenance?

363

Brox, J. A. 2008. Infrastructure investment: The foundation of Canadian competitiveness. IRPP Policy Matters 9(2):1–48. Congressional Budget Office. 1988. New directions for the nation’s public works. Washington, DC: U.S. Government Printing Office. ———. 2007. Trends in public spending on transportation and water infrastructure, 1956 to 2004. Washington, DC: U.S. Government Printing Office. Easterly, W., T. Irwin, and L. Servén. 2008. Walking up the down escalator: Public investment and fiscal stability. World Bank Research Observer 23(1):37–56. Escobal, J., and C. Ponce. 2003. The benefits of rural roads: Enhancing income opportunities for the rural poor. Working Paper No. 40. Lima, Peru: Grupo de Analysis Para el Desarrollo (GRADE). Estache, A. 2004. Emerging infrastructure policy issues in developing countries: A survey of the recent economic literature. Policy Research Working Paper Series 3442. Washington, DC: World Bank. Estache, A., and M. Fay. 2007. Current debates on infrastructure policy. Policy Research Working Paper Series 4410. Washington, DC: World Bank. Fay, M., and M. Morrison. 2007. Infrastructure in Latin America and the Caribbean: Recent developments and key challenges. Washington, DC: World Bank. Feehan, J. P. 1998. Public investment: Optimal provision of Hicksian public inputs. Canadian Journal of Economics 31(3):693–707. Feldstein, M., and M. Rothschild. 1974. Towards an economic theory of replacement investment. Econometrica 42:393–423. Glomm, G., and B. Ravikumar. 1994. Public investment in infrastructure in a simple growth model. Journal of Economic Dynamics and Control 18:1173–1187. Greenstein, J. 2012. Comments on “What is the value of infrastructure maintenance? A survey.” Washington, DC: USAID. Gwilliam, K., and A. Kumar. 2002. Road funds revisited: A preliminary appraisal of the effectiveness of “second generation” road funds. TWU Series, TWU-47. Washington, DC: World Bank. Gwilliam, K., and Z. Shalizi. 1999. Road funds and user charges. World Bank Research Observer 14(2):159–185. Harral, C. 1988. Road deterioration in developing countries: Organization and management of road maintenance. World Bank Policy Study. Washington, DC: World Bank. Heggie, I. G. 1995. Management and financing of roads: An agenda for reform. World Bank Technical Paper No. 275, Africa Technical Series. Washington, DC: World Bank. ———. 2000. Road funds-1: What went wrong? World Highways 9(6):35–37. Holtz-Eakin, D. 1994. Public-sector capital and the productivity puzzle. Review of Economics and Statistics 76:12–21. Kalaitzidakis, P., and S. Kalyvitis. 2004. On the macroeconomic implications of maintenance in public capital. Journal of Public Economics 88:695–712. ———. 2005. Financing new public investment and/or maintenance in public capital for growth? The Canadian experience. Economic Inquiry 43:586–600. Kalyvitis, S., and E. Vella. 2011. Public capital maintenance, decentralization and U.S. productivity growth. Public Finance Review 39(6):784–809. McGrattan, E., and J. Schmitz. 1999. Maintenance and repair: Too big to ignore. Federal Reserve Bank of Minneapolis Quarterly Review 23:2–13.

364

Felix Rioja

National Surface Transportation Policy and Revenue Study Commission. 2008. Transportation for tomorrow: Report of the National Surface Transportation Policy and Revenue Study Commission. Washington, DC: U.S. Department of Transportation. www.transportationfortomorrow.org/final_repor/index.htm. Reinikka, R., and J. Svensson. 2002. Coping with poor public capital. Journal of Development Economics 69:51–69. Remy Simatovic, M. I. 2008. Impact of rural roads program on democracy and citizenship in rural areas of Peru. Working Paper 45124. Washington, DC: World Bank. Rioja, F. K. 2003a. Filling potholes: Macroeconomic effects of maintenance vs. new investment in public infrastructure. Journal of Public Economics 87:2281–2304. ———. The penalties of inefficient public infrastructure. Review of Development Economics 7(1):127–137. Schmalensee, R. 1974. Market structure, durability, and maintenance effort. Review of Economic Studies 41:277–287. Tanzi, V. 2005. Building regional infrastructure in Latin America. Working Paper SITI-10. Washington, DC: Inter-American Development Bank. Tanzi, V., and H. Davoodi. 1998. Corruption, public investment, and growth. In The welfare state, public investment and growth, ed. H. Shibata and T. Ihori, 41–60. Tokyo: Springer-Verlag. World Bank. 1979. The highway maintenance problem. Washington, DC: World Bank. ———. 1986. Project completion report: Nepal—Second highway project (Credit 730NEP). Report 6296. South Asia Regional Office. www-wds.worldbank.org /external/default/WDSContentServer/WDSP/IB/1986/06/20/000009265_3960924 205119/Rendered/PDF/multi_page.pdf. ———. 1990. Project completion report: Republic of Tunisia—Fourth highway project (Loan 1841-TUN). Report No. 8648. Infrastructure Operations Division. Country Department II. Europe, Middle East and North Africa Regional Office. www-wds .worldbank.org/external/default/WDSContentServer/WDSP/IB/1990/05/18 /000009265_3960924231425/Rendered/PDF/multi_page.pdf. ———. 1992. Project completion report: Jamaica—Highway maintenance project. Report No. 11128. Department III. Infrastructure Division. Latin America and the Caribbean Regional Office. www-wds.worldbank.org/external/default /WDSContentServer/WDSP/IB/1992/09/17/000009265_3960925164212/Rendered /PDF/multi_page.pdf. ———. 1993. Project completion report: Burkina Faso—Fourth highway project (Credit 1164-UV). Report No. 12001. Infrastructure Division, Sahelian Department. Africa Region. www-wds.worldbank.org/external/default/WDSContentServer /WDSP/IB/2008/11/27/000333038_20081127040944/Rendered/PDF/120010 PCR0REPL101official0use0only1.pdf. ———. 1994a. Project completion report: Republic of Gambia—Second highway maintenance project (Credit 1682-GM). Report No. 13490. Infrastructure Operation Division. Africa Region. http://www-wds.worldbank.org/external/default /WDSContentServer/WDSP/IB/1994/08/31/000009265_3961007043804 /Rendered/PDF/multi_page.pdf. ———. 1994b. World development report, 1994: Infrastructure for development. New York: Oxford University Press. ———. 2001. Implementation completion report (IDA-23950) on a credit in the amount of SDR 58.6 million to Bolivia for a second road maintenance project. Re-

what is the value of infrastructure maintenance?

365

port No. 23327. Finance, Private Sector and Infrastructure Department. Country Management Unit–LCC6C. Latin America and the Caribbean Regional Office. ———. 2007a. Evaluation of bank support for road funds. Background Paper for Evaluation of World Bank Assistance for the Transportation Sector, 1995–2005. Washington, DC: World Bank. ———. 2007b. Implementation completion and results report on a loan in the amount of US$50 million to the Republic of Peru for the Second Rural Roads Project. Report No. ICR0000366. www-wds.worldbank.org/external/default /WDSContentServer/WDSP/IB/2007/08/23/000310607_20070823104222 /Rendered/PDF/ICR0000366.pdf. ———. 2008. World development indicators. Washington, DC: World Bank. ———. 2011. World development indicators. Washington, DC: World Bank. Zietlow, G. 2004. Road funds in Latin America. Presentation at the University of Birmingham, Senior Roads Executive Programme (26–30 April). ———. 2005. Cutting costs and improving quality through performance-based road management and maintenance contracts: The Latin American and OECD experiences. Working Paper. Birmingham, U.K.: University of Birmingham.

commentary Waheed Uddin Felix Rioja backs up his survey of the literature on the importance of timely infrastructure maintenance with a case study from Peru. The performance of key infrastructure assets is generally poor in low-income countries. For example, electric power losses are 22 percent of output for low-income countries, 11 percent for middle-income countries, and 6 percent for high-income countries. Developed and high-income countries show a high correlation between gross domestic product per capita and paved road density/energy consumption per capita. Through specific examples of road maintenance, Rioja argues that well-maintained infrastructure increases a country’s productivity and economic growth. He points out that infrastructure has been neglected in developing and industrial countries alike, although I would argue that the maintenance and rehabilitation of highway and road networks in the United States have been given their due share of both federal and state transportation budgets. Rioja’s chapter focuses on four areas related to the role and value of maintenance: (1) theoretical frameworks that explain how maintenance affects productivity and determine its optimal level; (2) a survey of the economic rate of return for maintenance; (3) the empirical evidence on the effects of maintenance on productivity and growth; and (4) a review of the various ways countries have financed the maintenance of roads. Rioja concludes with a case study on a Peruvian program for maintaining rural roads that may provide a model for other countries. I agree with Rioja’s assertion about the role and value of infrastructure maintenance. The maintenance phase has the most significant influence on the life of road infrastructure (Hudson, Haas, and Uddin 1997), as shown in figure C13.1. In many developing countries and emerging economies, there is a clear deficit of infrastructure supply and demand. This implies that decision makers need to spend more of their agencies’ budgets on investing in new capital assets than on maintenance. This commentary focuses on four issues related to sustainable infrastructure systems, which are not discussed or are inadequately discussed in Rioja’s chapter. U

366

Infrastructure asset management system (AMS) implementation. AMS has been the key to successful highway maintenance programming and budget optimization in North America. It requires computerized asset inventory, periodic condition monitoring using performance measures and their thresholds for maintenance intervention, and performance modeling for life cycle analysis. However, the World Bank and other international infrastructure banks have not adequately implemented AMS until recently; instead, they provided loans to build highways in developing countries without evaluating other competing modes of transportation, such as rail

commentary

367

Figure C13.1 Influence of Life Cycle Phases on Infrastructure: Road Network Example 100 Increasing expenditures 50

50 Decreasing influence

0

Cumulative total cost (percentage)

Level of influence (percentage)

100

0

Planning Design Construction Maintenance 1

Short-term CO2 emissions

20

Long-term CO2 emissions Time (years)

Source: Hudson, Haas, and Uddin (1997).

U

U

assets. These highway loans also generally ignored female participation, adverse impacts on air quality, environmental sustainability, and financial ability to perform timely maintenance and preservation. Life cycle analysis (LCA) of costs and benefits for most effective mainteFigure 13C.1 nance programming. Maintenance-intervention scheduling should be based on maintenance-needsLincoln_Ingram_Infrastructure assessment using periodic condition-monitoring data combined with maintenance-intervention criteria for structural and functional adequacy. The use of LCA is imperative considering agency costs, user costs, and societal benefits. The LCA approach is significant for scheduling and prioritizing transportation budgets using effective maintenance and rehabilitation interventions. Environmental sustainability with respect to air pollution and carbon emission. In the United States, greenhouse gas (GHG) emissions from transportation represent 33 percent of total U.S. GHG emissions (Uddin 2012). About 81 percent of transportation-related emissions are produced by road vehicles (mainly automobiles and trucks). Air pollution due to vehicle emissions affects people with respiratory diseases and affects produc-

368

Waheed Uddin

Figure C13.2 Vehicle Operating Cost as a Function of Pavement Condition on Pavement Serviceability Index Scale of 0 (Failed) to 5 (Excellent)

Pavement serviceability index

5 4 3 2 1 0 0

10

20

30

40

Increase in vehicle operating cost (percentage) Source: Uddin (2006).

U

tivity, especially in the summer months. These public health costs amount to $0.15 per vehicle mile in addition to $0.15 to $0.50 per vehicle mile of 13C.2 vehicle operating costs (VOC).Figure Other significant user cost is crash-related mortality costs (UddinLincoln_Ingram_Infrastructure 2006). Poor road conditions increase VOC and crash risks (figures C13.2 and C13.3), as discussed by Uddin (2006). The burning of fossil fuel produces a large amount of CO2 emissions, which include 28 percent energy-related emissions by transport vehicles, trailing behind 34 percent emissions by electric power plants (Energy Information Administration 2007). This anthropogenic CO2 emission is a source of 80 percent of the global warming potential. Sustainable transportation policies for cities and urban areas. Cities are now contributing about 75 percent of all global anthropogenic GHG emissions (Uddin 2012). The built infrastructure of cities and intercity/ interstate travel lead to significant adverse impacts on air quality, GHG emissions, and global warming. Transportation sources contribute 28 percent of energy-related GHG emissions in the United States and 23 percent worldwide. A recent study (Uddin 2012) shows that traffic-related CO2 emissions in the United States are higher per capita in several rural and smaller cities than in large urban areas. The addition of highway lanes to add capacity and ease congestion, the inadequate use of mass transit, the growth of urban sprawl, the construction of more roads, and the prevalence of traditional stop-controlled intersections are primary contributors to vehicle-related CO2 and air quality degradation. Less work-related

commentary

369

Figure C13.3 VOC Components Tire 8.4% Vehicle repair and maintenance 30.5% Gasoline 34.7%

Oil 1.6% Depreciation 24.8% Source: Uddin (2006).

travel by cars and more use of mass transit buses and rail services can decrease CO2 emissions and improve environmental sustainability perforFigurebe13C.3 mance. These policy aspects must addressed by stakeholders in every country, including international lending institutions, in order to promote Lincoln_Ingram_Infrastructure environmentally sustainable transportation infrastructure.

references Energy Information Administration. 2007. Annual energy outlook. Washington, DC: Energy Information Administration, U.S. Department of Energy. Hudson, W. R., R. Haas, and W. Uddin. 1997. Infrastructure management. New York: McGraw-Hill. Uddin, W. 2006. Air quality management using modern remote sensing and spatial technologies and associated societal costs. International Journal of Environmental Research and Public Health 3(3):1–9. ———. 2012. Mobile and area sources of greenhouse gases and abatement strategies. In Handbook of climate change mitigation, ed. W. Chen, J. Seiner, T. Suzuki, and M. Lackner. New York: Springer.

14 How and Why Does the Quality of Infrastructure Service Delivery Vary? George R. G. Clarke

A

ccess to infrastructure and the quality of services are very poor in many developing countries. This is a problem because studies have found that poor-quality service and weak access to infrastructure can slow development and impede growth.1 Improving access and quality would therefore benefit many people in developing countries. However, when many sectors (power, water, telecommunications, and transportation) have problems—as they often do in developing countries—it is not clear how governments should focus their efforts. This chapter seeks to answer two questions. The first concerns how the availability, quality, and price of infrastructure vary across countries. To answer, we first look at the correlation between different measures of infrastructure services

The data used in this chapter come from various sources, including the International Telecommunication Union, the International Energy Agency, and the World Bank’s World Development Indicators, Doing Business Indicators, Logistic Performance Indicators, World Governance Indicators, and Enterprise Surveys (World Bank 2012). I would like to thank Ahmed M. Abdel Aziz, Karin Brandt, and Gregory Ingram for helpful discussions and comments on earlier drafts. Responsibility for all errors, omissions, and opinions rests solely with me. 1. See, for example, Straub (2008) for a recent review of the literature on infrastructure and growth. In a survey of 64 empirical papers, Straub concludes that close to two-thirds of the studies found a positive and significant link between various measures of infrastructure and economic growth. For the three sectors with the greatest number of studies (electricity, roads, and telecommunications), positive links were found in about 70 percent of the studies between physical measures of infrastructure and economic growth. See also World Bank (1994).

370

how and why does the quality of infrastructure service delivery vary?

371

across countries. This chapter updates information from the 1994 World Development Report: Infrastructure for Development (World Bank 1994), which made similar comparisons using data from the mid-1990s.2 Access is highly correlated within countries for different infrastructure services (water, electricity, mobile phones, fixed-line phones, roads, and rail). So, for example, more people have electricity in countries where access to fixed-line and mobile phone service is higher and where road and rail networks are more dense. This is partly because per capita income and population density strongly affect all measures of access. That is, access to most infrastructure services is higher in countries where income is higher and population density is higher.3 In contrast, service quality and price are not highly correlated within countries. Countries with poor service in one sector do not necessarily have poor service in other sectors. For example, countries in which power outages are common do not necessarily have more unpaved roads. Similarly, although prices are often positively correlated across services (e.g., when per-minute charges for mobile phones are high, power prices are also high), the correlations are lower than for access. The second question addressed here is how infrastructure affects the operations and growth of firms. The chapter shows that firm managers are more concerned about electricity than about transportation and that the strength of their concern is strongly related to the reliability of electricity service. In contrast, managers’ perceptions about transportation are not strongly associated with measures of availability, service quality, or price. The most robust correlations are related to the cost and time associated with importing materials. Managers said transportation was a more serious problem in countries where it costs more to import a 20-foot container, where it takes longer for goods to clear customs, and where connections to international trade routes are worse.

Data This section describes the main variables in the empirical analysis. For each type of infrastructure (transportation, electricity, telecommunications, and water), the measures are assigned into categories related to access, price, and quality of service. The data are country-level, cross-sectional, and mostly for 2009–2010. The data come from a variety of sources: the World Bank’s World Development Indicators database (World Bank 2011b), the International Telecommunication Union’s World Telecommunication Indicators database (International

2. World Bank (1994) preceded—and indeed encouraged—the large increase in private sector participation in infrastructure that occurred in the 1990s and early 2000s. For a review of the literature on privatization, including in the infrastructure sector, see Megginson (2005), Megginson and Netter (2001), and Shirley and Walsh (2000). 3. This is consistent with results from the early 1990s from World Bank (1994).

372

George R. G. Clarke

Telecommunication Union 2012), the World Bank’s Logistics Performance Indicators database (Arvis et al. 2012), the International Energy Agency’s Energy Prices and Taxes database, and the World Governance Indicators database (Kaufmann, Kraay, and Mastruzzi 2009). Additional data come from the World Bank’s Doing Business Indicators (World Bank 2011a) and Enterprise Surveys database (World Bank 2012). Data from these two sources apply to formal firms in the economy. The Doing Business Indicators make various assumptions about the type of enterprise involved. For the most part, however, the Doing Business Indicators are calculated for medium-size or large formal enterprises.4 Similarly, the Enterprise Survey only covers formal firms in manufacturing, retail trade, and services with at least five employees.5 Because of the World Bank’s focus on development, almost all Enterprise Survey data are for low- and middle-income countries. The indicators from these two sources might not represent the experiences and perceptions of informal microenterprises. Full descriptions and sources for each of the variables are included in appendix 14.1.

Correlations of Different Measures of Infrastructure Performance This section looks at the correlation between different measures of infrastructure access, price, and quality. As in the 1994 World Development Report, we are interested in the extent to which measures of performance are correlated at the country level. That is, do the same countries tend to have better infrastructure services over a range of performance measures? Although looking at simple correlations can be informative, it is possible that correlations between the different measures might reflect the effect of income, population density, or other macroeconomic variables on the availability, price, and quality of infrastructure. We therefore also look at the correlations after controlling for these differences, which is done by estimating an ordinary leastsquares (OLS) model allowing macroeconomic factors and institutional quality to affect the availability, price, and quality of infrastructure services. We then look at the correlation of the residuals to see how highly correlated the performance of different infrastructure services is after controlling for these macroeconomic and institutional differences.

4. The Doing Business website provides detailed descriptions of how the indexes are constructed. See www.doingbusiness.org. 5. The surveys covered all manufacturing sectors (group D based on ISIC 3.1), construction (group F), retail and wholesale services (subgroups 52 and 51 of group G), hotels and restaurants (group H), transport, storage, and communications (group I), and computer and related activities (subgroup 72 of group K). Only formal firms with at least five employees are included. See World Bank (2009) for more detail.

how and why does the quality of infrastructure service delivery vary?

373

DETERMINANTS OF INFRASTRUCTURE PERFORMANCE

To control for macroeconomic and institutional differences, the various measures of infrastructure are regressed on a set of macroeconomic control variables, and the residuals from each of the regressions are calculated.6 Infrastructure ! " # $ macroeconomic controls # % institutional quality # & The independent variables used here are similar to the variables used in Wallsten (2001), which looks at the telecommunications sector. Because the point is to identify things that might cause the high correlation between the various measures of access across sectors, the focus is on things that are likely to affect access in all sectors—not just telecommunications. Therefore, the institutional variables related to competition and regulation of telecommunications are not included. Area is included because population density is likely to be important for some forms of infrastructure. These are the variables used:7 U

U

U

U

Per capita income. For the most part, infrastructure services could be expected to be affected by per capita income. For infrastructure services that government agencies provide (e.g., roads), countries with higher per capita income should generally find it easier to finance infrastructure needs. Moreover, to the extent that infrastructure services are normal goods, demand should be higher in wealthier countries. Area. We expect service availability, quality, and prices to be affected by population density. Because area is included in log-form and the regression also includes population in log-form, this variable essentially allows us to control for population density.8 Population. If there are economies of scale in providing infrastructure services, then population size might affect infrastructure service. As discussed above, this variable also controls for population density given that (log of) area is also included among the regressors. Urban population. Especially in developing countries, infrastructure provision is limited in rural areas.9 Moreover, the cost of network expansion and maintenance is generally higher in rural areas. It can therefore be

6. See table 14.17 for data sources for macroeconomic variables. 7. Long-term and short-term debt has been omitted because including these variables significantly reduces sample size. 8. The null hypothesis that population density alone affects infrastructure services can be tested by testing whether $1 ! '$2, where $1 is the coefficient on (log of) population and $2 is the coefficient on (log of ) area. 9. See, for example, table 2.2 in Clarke and Wallsten (2003).

374

U

U

George R. G. Clarke

expected that coverage will be lower, quality will be lower, and prices will be higher in countries with large rural populations. Exports. Export orientation might also affect demand for infrastructure services. In particular, export-oriented firms might have greater demand for both transportation and communications infrastructure.10 One concern about this variable is the potential for endogeneity. That is, it is possible that the availability of infrastructure affects export performance, rather than the reverse. Corruption. Given state involvement in regulating, financing, and implementing infrastructure projects, corruption might affect the availability, quality, and cost of infrastructure services. That is, it is likely that the cost of public infrastructure will be higher in countries where corruption is a problem.11 In practice, this variable is likely to serve as an overall proxy for institutional development. As Langbein and Knack (2010) note, country-level measures of institutional development (e.g., related to the rule of law, regulatory quality, and corruption) tend to be very highly correlated.12 As a result, it is difficult to isolate the effects of corruption from the effects of other aspects of institutional development.

AVAILABILITY OF INFRASTRUCTURE SERVICES

This chapter looks at six measures of access: (1) the percentage of the population that has access to electricity; (2) the percentage of the population that has access to improved water; (3) mobile phone subscriptions per 100 inhabitants; (4) fixed-line phone subscriptions per 100 inhabitants; (5) rail density (kilometers per 100 square kilometers [km per 100 sq. km]); and (6) road density (km per 100 sq. km). Higher values mean greater access for all measures. These are meant to capture the extent to which the population has access to or uses infrastructure services. Later, the chapter looks at similar regressions for the price and quality of infrastructure services. For the most part, the measures of access and availability are strongly correlated with one another (table 14.1), and the average absolute correlation is 0.6.

10. For example, see Freund and Weinhold (2002, 2004) and Clarke and Wallsten (2006) for a discussion of the impact of Internet access on exporting. Similarly, Djankov, Freund, and Pham (2010) show that increasing the time to export has a large impact on trade. Each additional day that it takes to export a product, due to transportation and customs delays, reduces trade by more than 1 percent. 11. Consistent with this, Kenny (2007) notes that construction contractors are more likely to pay bribes than other firms, and they pay more as a percentage of sales when they do. 12. See also Kaufmann, Kraay, and Mastruzzi’s (2010) response to Langbein and Knack (2010).

how and why does the quality of infrastructure service delivery vary?

375

Table 14.1 Correlation of Access Indicators Before Controlling for Macroeconomic Regressors Access to Electricity Access to electricity Access to improved water Mobile phone subscriptions Fixed-line phone subscriptions Rail density Road density

Access to Improved Water

Mobile Phone Subscriptions

Fixed-Line Phone Subscriptions

Rail Density

Road Density

1.00 0.77*** (0.00) 0.74*** (0.00) 0.77*** (0.00) 0.41*** (0.02) 0.27 (0.13)

1.00 0.64*** (0.00) 0.66*** (0.00) 0.52*** (0.00) 0.58*** (0.00)

1.00 0.63*** (0.00) 0.51*** (0.00) 0.41*** (0.00)

1.00 0.57*** (0.00) 0.63*** (0.00)

1.00 0.88*** (0.00)

1.00

***, **, * statistically significant at 1%, 5%, and 10% significance levels. Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

So, for example, access to improved water sources, mobile and fixed-line phone subscriptions, and rail density are higher, on average, in countries where access to electricity is higher. Most of the simple correlations are greater than 0.5, and all except one are statistically significant. This suggests that similar factors affect access for all types of infrastructure. It seems plausible that the high correlation between the various measures of access could be due to some macroeconomic variable affecting all measures of access. One factor that might affect access across sectors is income. As shown in table 14.2, most measures of access increase as per capita income increases. Table 14.3 shows the results from regressions in which the dependent variables are measures related to access and availability of infrastructure. Consistent with table 14.2, access to infrastructure is higher when per capita income is higher. The coefficients are statistically significant and positive in all six regressions. The coefficients imply that a 1-percent increase in per capita income would increase the population with access to electricity by 0.3 percent, the population with access to improved water by 0.1 percent, the number of mobile phone subscriptions by 0.3 percent, and the number of fixed-line phone subscriptions by

376

George R. G. Clarke

Table 14.2 Different Measures of Access by Income Level Access to electricity Per capita electricity consumption Fixed-line phone subscriptions Mobile phone subscriptions Rail density Road density

Low

Lower Middle

Upper Middle

High

19.0 130.8 0.9 31.2 0.3 11.0

71.2 755.6 8.0 59.8 0.7 25.5

98.3 2,282.3 21.5 98.6 1.0 19.0

99.7 6,693.1 41.7 117.6 4.9 129.0

Note: See appendix 14.1 for full variable descriptions. Income levels are based on classifications from the World Bank. Low-income countries have per capita income of $1,005 or lower; lower-middle-income countries have per capital income between $1,006 and $3,975. Uppermiddle-income countries have per capita income between $3,976 and $12,275. High-income countries have per capita income of more than $12,275. Source: See appendix 14.1 for data sources.

0.4 percent.13 Rail and road density would increase by 2.6 percent and 0.2 percent, respectively. Access also appears to be affected by population density. The coefficients on area are negative and statistically significant in all six regressions. This indicates that access is generally lower in countries that have greater area. In contrast, the coefficient on population is positive in all regressions and is statistically significant in four of the six regressions. Because access tends to be lower in countries that are larger in area but higher in countries with larger populations, access appears to be negatively correlated with population density. These results are consistent with the idea that it is easier, and potentially cheaper, to expand access in densely populated countries.14 Per capita income, population density, and, to a lesser extent, urban population and institutional quality (as proxied by control of corruption) explain a significant part of cross-country differences in access to infrastructure. The R-squared terms for the regressions are between about 0.64 and 0.82. Controlling for per capita income, population density, and the other control variables reduces the correlation between the access indicators significantly

13. All elasticities are calculated at the mean values of the access indicators. 14. It is, however, important to note that we can reject the null hypothesis that the coefficients are equal in absolute value in all six regressions at a 5 percent significance level or higher. That is, area and population appear to affect the different services to different degrees for each type of service.

377

0.70

74 20.693*** (5.21) –4.464** (–2.55) 7.649*** (3.75) 0.347** (2.05) –0.146 (–1.42) –0.603 (–0.15) –191.906*** (–4.97)

160 23.400*** (7.18) –3.895*** (–2.63) 1.046 (0.58) 0.347** (2.38) 0.057 (0.66) –5.216* (–1.77) –112.639*** (–3.58) 0.65

0.64

Mobile Phone Subscriptions

0.75

160 7.918*** (6.93) –1.815*** (–3.50) 2.014*** (3.19) 0.037 (0.72) –0.051* (–1.69) 5.756*** (5.57) –60.876*** (–5.51)

Fixed-Line Phone Subscriptions

Telecommunications

140 9.760*** (6.89) –2.626*** (–4.17) 1.055 (1.40) 0.085 (1.35) –0.113*** (–2.90) –1.162 (–0.91) 15.916 (1.21)

Access to Improved Water (% of Population)

Access to Electricity (% of Population)

Note: T-statistics in parentheses. ***, **, * statistically significant at 1%, 5%, and 10% significance levels. Source: See appendix 14.1 for data sources.

R-squared

Observations Per capita gross national income (log) Area (log, sq. km) Population, total (log) Urban population (% of population) Exports of goods and services (% of gross domestic product) Control of corruption (high values mean less corruption) Constant

Water

Electricity

Table 14.3 Impact of Macroeconomic Variables on Availability of Infrastructure

0.82

68 0.675*** (4.92) –0.630*** (–11.50) 0.505*** (7.82) –0.014** (–2.21) –0.008** (–2.65) 0.202* (1.82) –1.994 (–1.45)

Rail Density

0.76

82 0.736*** (4.83) –0.720*** (–10.38) 0.507*** (6.84) –0.010 (–1.45) –0.004 (–1.03) –0.017 (–0.14) –5.024*** (–3.20)

Road Density

Transportation

378

George R. G. Clarke

Table 14.4 Correlation of Access Indicators After Controlling for Macroeconomic Regressors Access to Electricity Access to electricity Access to improved water

1.00 0.50*** (0.00) Mobile phone subscriptions 0.37*** (0.00) Fixed-line phone subscriptions 0.25*** (0.03) Rail density 0.01 (0.97) Road density –0.17 (0.38)

Access to Improved Water

Mobile Fixed-Line Phone Phone Subscriptions Subscriptions

Rail Density

Road Density

1.00 0.20** (0.02) 0.05 (0.53) 0.08 (0.53) 0.03 (0.79)

1.00 0.06 (0.43) –0.05 (0.70) –0.02 (0.83)

1.00 0.14 (0.27) 0.31*** (0.00)

1.00 0.47*** (0.00)

1.00

***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

(table 14.4).15 Although some correlations remain statistically significant (e.g., between rail and road density and between access to water and electricity), the point estimates of the correlations are significantly smaller—although they remain mostly positive—and most correlations are statistically insignificant, and the average value is 0.14. This suggests that the high correlations among the variables representing access to different types of infrastructure are largely due to access being higher in richer and more densely populated countries. PRICE OF INFRASTRUCTURE SERVICES

Six measures of price are used: (1) the price per kWh for electricity for household users; (2) the cost of an electricity connection for a business; (3) the price of a three-minute peak-time local fixed-line call; (4) the price of a three-minute peak-time cellular call; (5) the cost of a fixed-line connection; and (6) the cost of importing a 20-foot container. No comparable cross-country data were available on the cost of water. Higher values mean more costly service for all variables.

15. These comparisons are made by calculating the residuals from each of the regressions in table 14.2 and calculating the correlations between them.

how and why does the quality of infrastructure service delivery vary?

379

Table 14.5 Correlation of Price Indicators Before Controlling for Macroeconomic Regressors Price per kWh for Electricity (Household) Price per kWh for electricity (household) Cost of electricity connection (business) Price of 3-min. peak call (fixed line) Price of 3-min. peak call (cellular) Price of fixed-line connection (business) Cost of importing 20-foot container

Cost of Electricity Connection (Business)

Price of Price of 3-Min. Peak 3-Min. Peak Call Call (Fixed Line) (Cellular)

Price of Fixed-Line Connection (Business)

Cost of Importing 20-Foot Container

1.00

0.38*** (0.02)

1.00

0.10 (0.53)

0.16*** (0.05)

1.00

0.44*** (0.00)

0.16*** (0.06)

0.22*** (0.01)

1.00

0.39** (0.02)

0.30*** (0.00)

0.32*** (0.00)

0.03 (0.77)

1.00

–0.28 (0.07)

0.27*** (0.00)

0.03 (0.69)

0.21*** (0.01)

0.02 (0.84)

1.00

***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

As with the measures of access to infrastructure, the price variables are mostly positively correlated with one another, even across different services (table 14.5). For example, the price per kWh for electricity is positively correlated with the cost of an electricity connection for business, the price of a three-minute fixed-line phone call, and the cost of a three-minute cellular phone call. In contrast to the results from access, however, the correlations are generally smaller, and the average correlation is 0.18. Most correlations are between about 0.1 and 0.3—compared to between 0.4 and 0.7 for the indicators related to access. Given that the correlations are relatively modest, we would probably not expect the macroeconomic variables to be strongly and consistently correlated with the price variables. That is, if income or other variables explained most of the crosscountry variation in prices, the price variables would be more highly correlated. Table 14.6 shows the results from regressing measures related to the price of infrastructure services on macroeconomic and institutional variables. These are

380

0.479

42 0.021 (1.10) –0.021** (–2.68) 0.010 (1.18) –0.000 (–0.43) –0.001 (–1.63) 0.022 (1.56) 0.067 (0.38) 0.046

148 0.027 (0.20) 0.009 (0.14) –0.060 (–0.82) 0.002 (0.27) –0.007** (–2.01) –0.102 (–0.83) 10.774*** (8.40)

Cost of Connection (Business)

Note: T-statistics in parentheses. ***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Source: See appendix 14.1 for data sources.

R-squared

Observations Per capita gross national income (log) Area (log, sq. km) Population, total (log) Urban population (% of population) Exports of goods and services (% of gross domestic product) Control of corruption (high values mean less corruption) Constant

Price per kWh (Household)

Electricity

Table 14.6 Impact of Macroeconomic Variables on Price of Infrastructure Services

0.097

143 –0.031** (–2.00) 0.015** (2.19) –0.009 (–1.10) –0.000 (–0.12) 0.000 (1.19) 0.041*** (2.95) 0.353** (2.45)

Price of 3-Min. Peak Call (Fixed Line)

0.151

143 0.063 (0.74) 0.006 (0.18) 0.020 (0.45) 0.003 (0.79) –0.006*** (–2.96) 0.139* (1.85) –0.222 (–0.29)

Price of 3-Min. Peak Call (Cellular)

Telecommunications

0.116

111 0.359** (2.62) –0.054 (–0.94) 0.042 (0.59) –0.003 (–0.58) –0.006 (–1.62) –0.121 (–1.02) 1.208 (0.97)

Price of Fixed-Line Connection (Business)

0.305

155 –0.117** (–2.06) 0.102*** (3.68) –0.129*** (–4.08) –0.001 (–0.41) –0.001 (–0.99) –0.067 (–1.29) 9.223*** (16.89)

Cost of Importing 20-Foot Container

Transportation

how and why does the quality of infrastructure service delivery vary?

381

meant to capture the cost of infrastructure service. In contrast to the previous results for access to infrastructure, most of the coefficients on the macroeconomic and institutional variables are statistically insignificant. Moreover, they do not, generally, show a consistent relationship with price. For example, the coefficient on per capita income is statistically significant in only three of the six regressions. Moreover, even when significant, the sign on the coefficient on income varies. The coefficient on income is positive in the regression for a fixed-line phone call with the point estimate indicating that the cost of a fixed-line phone call would be about 0.05 higher if income were increased by 1 percent. However, the reverse is true for the price of getting a fixed-line connection and the cost of importing a 20-foot container; in those cases, higher income appears to be correlated with lower prices. Although it seems that access might be more limited in countries where population density is lower because it might be more expensive to serve spread-out rural customers, this does not seem to be the case. Neither urban population share nor population density (i.e., area and population) are consistently negatively correlated with the price indicators. Although the price of fixed-line phone calls is higher in geographically larger countries, the price of electricity is negatively correlated with area. The cost of importing a 20-foot container is higher in larger countries, possibly reflecting the greater inland transportation costs. One plausible explanation for the insignificant relationship between population density and infrastructure service prices is that the price of infrastructure services is less than the cost of provision. In many countries, prices are set in consultation with government-appointed regulators. In this respect, the lower levels of access observed in large countries might reflect rationing due to regulated prices rather than high prices per se. After controlling for the macroeconomic and institutional variables, the correlation between the different measures of the price of infrastructure services falls further (table 14.7). Although most remain positive, the correlations are mostly statistically insignificant and are smaller in absolute value; the average value is 0.16. The relatively modest drop is probably not surprising given that the macroeconomic variables are not consistently correlated with the price measures. QUALITY OF INFRASTRUCTURE SERVICES

For all of the variables, high values represent poor service quality. For example, high values of transmission and distribution losses, power and water outages, and losses during transportation all suggest poor-quality services. The simple correlations between the quality variables are mostly positive (table 14.8). While this suggests that the same countries generally have quality problems across infrastructure services, the point estimates are mostly small—less than 0.25 in most cases—and statistically insignificant; the average correlation is 0.15. Table 14.9 shows the results from regressing various measures related to the quality of infrastructure services on macroeconomic and institutional variables. The dependent variables are as follows: (1) the number of outages that firms

382

George R. G. Clarke

Table 14.7 Correlation of Price Indicators After Controlling for Macroeconomic Regressors Price per kWh Cost of Price of 3-Min. Price of 3-Min. Price of Cost of for Electricity Electricity Peak Call Peak Call Fixed-Line Importing (Household) Connection (Fixed Line) (Cellular) Connection 20-Foot (Business) (Business) Container Price per kWh for electricity (household) Cost of electricity connection (business) Price of 3-min. peak call (fixed line) Price of 3-min. peak call (cellular) Price of fixed-line connection (business) Cost of importing 20-foot container

1.00 0.20 (0.22) –0.13 (0.43) 0.29*** (0.07) 0.20 (0.28) 0.05 (0.76)

1.00 0.13 (0.12) 0.20*** (0.02) 0.25*** (0.01) 0.24*** (0.00)

1.00 0.25*** (0.00) 0.12 (0.20) 0.14 (0.10)

1.00 0.16 (0.10) 0.22** (0.01)

1.00 0.16 (0.10)

1.00

***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

face in a month; (2) electricity transmission and distribution losses; (3) faults per 100 fixed-line telephones; (4) losses due to breakage and spoilage during shipping; (5) percentage of roads that are unpaved; and (6) number of water service interruptions and shortages in a month. For the most part, service quality is higher in countries with higher per capita income. In particular, there are fewer power outages, lower transmission and distribution losses, fewer water service interruptions, and a lower percentage of unpaved roads in wealthier countries. This suggests that the higher coverage in high-income countries does not come at the expense of worse service quality. Population density has a mixed effect on service quality. Although power outages and water service interruptions are more common in densely populated countries (i.e., the coefficient on area is negative and the coefficient on population is positive), densely populated countries have lower distribution and transmission losses and fewer unpaved roads. After controlling for macroeconomic and institutional variables, the correlation between different quality indicators falls further (table 14.10). Although most remain positive, they generally become smaller, and almost all correlations are statistically insignificant; the average correlation is 0.10. Once again, given

how and why does the quality of infrastructure service delivery vary?

383

Table 14.8 Correlation of Quality Indicators Before Controlling for Macroeconomic Regressors No. of Power Electric Power Faults Losses to Percentage of No. of Water Outages Transmission per 100 Breakage or Roads That Shortages per Month and Distribution Fixed-Line Spoilage During Are Unpaved per Month Losses Telephones Shipping No. of power outages per month Electric power transmission and distribution losses Faults per 100 fixed-line telephones Losses to breakage or spoilage during shipping Percentage of roads that are unpaved No. of water shortages per month

1.00 0.18

1.00

(0.11) 0.04 (0.77)

0.04 (0.79)

1.00

0.19 (0.13)

0.03 (0.83)

0.20 (0.39)

1.00

0.20 (0.24)

0.21 (0.14)

0.12 (0.54)

0.27 (0.23)

1.00

0.72*** (0.00)

0.18 (0.13)

–0.24 (0.17)

0.01 (0.94)

0.04 (0.84)

1.00

***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

the lack of strong and consistent correlations in the regressions, it is not surprising that the correlations do not fall greatly after controlling for the macroeconomic variables.

Effect of Infrastructure Services on Firm Performance The results of the analysis suggest that although infrastructure access is mostly better in richer and more densely populated countries, the price and quality of infrastructure services are not consistently correlated with the macroeconomic control variables. In addition, the price and quality of service are not highly correlated across different infrastructure subsectors. For example, countries with high-quality telecommunications service do not necessarily have high-quality power or water service. This section discusses how the availability, price, and quality of infrastructure services affect firm behavior and performance.

384

0.25

113 –3.618* (–1.78) –2.869*** (–2.65) 3.088** (2.52) 0.005 (0.05) –0.045 (–0.59) –4.065* (–1.81) 23.789 (1.25) 0.22

122 –3.364** (–2.00) 0.853 (1.15) –2.418*** (–2.84) 0.049 (0.64) 0.005 (0.13) –2.048 (–1.47) 70.056*** (4.09)

Electric Power Transmission and Distribution Losses

Note: T-statistics in parentheses. ***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Source: See appendix 14.1 for data sources.

R-squared

Observations Per capita gross national income (log) Area (log, sq. km) Population, total (log) Urban population (% of population) Exports of goods and services (% of gross domestic product) Control of corruption (high values mean less corruption) Constant

No. of Power Outages per Month

Electricity

Table 14.9 Impact of Macroeconomic Variables on Quality of Infrastructure Services

0.18

61 2.108 (0.65) 2.102 (1.28) –2.973 (–1.43) –0.277* (–1.82) 0.054 (0.65) –4.030 (–1.25) 31.037 (0.87)

Faults per 100 Fixed-Line Telephones

Telecommunications

0.19

60 0.187 (0.26) –0.538 (–1.57) –0.192 (–0.42) 0.039 (1.33) –0.036 (–1.28) –1.686** (–2.12) 8.396 (1.08)

Losses to Breakage or Spoilage During Shipping

0.51

54 –24.141*** (–3.91) 9.450*** (–3.77) –10.912*** (–3.56) 0.191 (–0.65) 0.164 (1.20) 8.982 (1.67) 300.771*** (2.90)

Percentage of Roads That Are Unpaved

Transportation

0.31

94 –2.494** (–2.53) –1.266** (–2.57) 1.611*** (2.79) 0.019 (0.43) –0.039 (–1.00) –1.168 (–1.08) 16.185* (1.72)

No. of Water Shortages per Month

Water

385

0.17 (0.12) –0.05 (0.76) 0.15 (0.26) 0.22 (0.21) 0.63 (0.00)

1.00

Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

No. of water shortages per month

Losses to breakage or spoilage during shipping Percentage of roads that are unpaved

No. of power outages per month Electric power transmission and distribution losses Faults per 100 fixed-line telephones

No. of Power Outages per Month

–0.18 (0.20) 0.04 (0.79) –0.16 (0.26) 0.27 (0.02)

1.00

Electric Power Transmission and Distribution Losses

Table 14.10 Correlation of Quality Indicators After Controlling for Macroeconomic Regressors

0.20 (0.40) –0.03 (0.90) 0.01 (0.94)

1.00

Faults per 100 Fixed-Line Telephones

0.37 (0.12) –0.11 (0.42)

1.00

Losses to Breakage or Spoilage During Shipping

–0.02 (0.92)

1.00

Percentage of Roads That Are Unpaved

1.00

No. of Water Shortages per Month

386

George R. G. Clarke

MANAGERS’ PERCEPTIONS ABOUT INFRASTRUCTURE SERVICES

How large an impact does the quality and availability of infrastructure have on firm performance? One common way of assessing how seriously different aspects of the investment climate constrain firm growth is to ask managers what they see as the biggest obstacles that they face. For example, the World Bank’s Enterprise Survey asks managers to rank a series of investment climate constraints on a fivepoint scale ranging from “no obstacle” to “very severe obstacle” and also to say which of these are the biggest constraints. Two of the obstacles that the Enterprise Survey asks about relate to infrastructure: electricity and transportation.16 Figure 14.1 shows the seven constraints that the greatest number of firm managers identified as their biggest problem. By far the most common concerns are electricity (top constraint in 25 countries with available data) and access to finance (top constraint in 23 countries). Tax rates and competition with informal firms also ranked among the top concerns in more than 15 countries. In contrast, transportation did not rank as the top constraint in any of the countries with available data.17 The analysis in the previous section suggests that access to infrastructure is a greater problem in low-income countries than in middle- and high-income countries, although quality and price are much less strongly related to income. Firm managers are generally more concerned about electricity in low-income countries. In 16 of 38 mostly low-income countries in sub-Saharan Africa, electricity was ranked as the top constraint (Clarke and Dinh 2012). In comparison, it was ranked as the top constraint in only one of 16 mostly middle-income countries in Latin America. Consistent with this, Gelb and colleagues (2006) find that firms in the poorest countries in Africa tend to be most concerned about basic services and stability: macroeconomic stability, electricity, and access to finance typically rank among the top concerns.18 As income increases, firms tend to become more concerned about the quality of governance and the capability of the state; corruption, tax rates, tax administration, and regulation become increasingly binding.19

16. Questions about telecommunications are only asked to information-technology firms and firms in retail trade and so are ignored here. 17. It ranked as the second-greatest constraint in three countries: Gabon, Guinea, and Malawi. 18. Carlin, Schaffer, and Seabright (2010) also show that concern about physical infrastructure, including power and transportation, tends to be greater in low-income countries. 19. For example, the two countries where crime ranked as the top constraint—South Africa and Namibia—are both middle-income countries. During the middle of the 2007–2008 Enterprise Survey in South Africa, a serious power crisis hit the country. Since South African firms were used to cheap and reliable power, this was a shock to managers. Before the crisis hit, firm managers were most likely to say that crime was a serious problem. After the crisis, they were most likely to say that electricity was a problem (Clarke 2011a).

how and why does the quality of infrastructure service delivery vary?

387

Figure 14.1 Constraints Identified by Firm Managers as Most Difficult 25

Number of countries

20 15 10 5 0

Electricity

Access to finance

Tax rates

Informal firms

Political instability

Crime

Inadequate education

Note: The biggest constraint is based on the percentage of managers who identified that constraint as their biggest problem among 15 different constraints. The other options were access to land, corruption, courts, customs and trade regulation, labor regulation, tax administration, and transportation. Data source: Dinh, Mauvridis, and Nguyen (forthcoming).

Figure 14.1 Lincoln_Ingram_Infrastructure

RELIABILITY OF PERCEPTION-BASED INDEXES

To assess what influences managers’ concerns about infrastructure, we regress aggregate measures of the percentage of firms that say infrastructure is a serious problem on the macroeconomic and institutional variables from the previous section and a vector of variables representing availability, price, and quality of infrastructure. The dependent variables come from the World Bank’s Enterprise Surveys (2012). Economists are often concerned about perception-based data (see, for example, Bertrand and Mullainathan 2001). Some researchers question whether managers have a good idea about binding constraints. One particular concern is that since only firms that exist can be interviewed—and by definition, these are firms that have managed to overcome any binding constraints—surveys of existing firms may underestimate the barriers caused by particularly binding constraints. Hausmann and Velasco (2005) illustrate this point with an analogy to camels and hippos. They note that the few animals found in the Sahara will be camels, which have adapted to life in the desert, rather than hippos, which depend heavily on water. Asking the camels about problems associated with life in the desert might not adequately represent the views of the missing hippos. Although underestimating the binding constraints is true, it seems that managers can better assess the constraints to running their businesses than can

388

George R. G. Clarke

outsiders like academics, politicians, and policy advisers. This would seem to be particularly true for broad constraints (e.g., whether electricity is a problem) rather than specific policy questions (e.g., whether the electricity company should be privatized or should invest in hydroelectric power). Moreover, it is important to remember that objective data also have problems—particularly for sensitive and difficult questions.20 In contrast, managers can easily answer questions about what they see as the biggest problems they face. SIMPLE CORRELATIONS

As a first exercise, we look at the simple correlations between measures of availability, price, and quality of electricity and transportation services and the two measures of perceptions: the percentage of firms identifying electricity as a serious problem and the percentage of firms identifying transportation as a serious problem. As discussed earlier, although many firms rated electricity as a serious problem, few rated transportation as a serious problem. The data come from the World Bank’s Enterprise Surveys, which, as noted, only include formal firms in manufacturing, retail trade, and services with at least five employees.21 The views of these firms might not reflect the views of informal microenterprises. Electricity All three measures of access are negatively correlated with firms’ perceptions about electricity (table 14.11). The negative correlation suggests that firms are more likely to say that electricity is a major or very severe problem in countries where access is lower. This is consistent with results in HallwardDriemeier and Aterido (2009), which show that the percentage of firms that complain about electricity is correlated with per capita electricity consumption. The correlation between access and perceptions could reflect that in countries with the lowest access rates, the mostly small and medium-size enterprises in the Enterprise Survey samples find it more difficult or more expensive to get facilities with electricity connections. That is, although most small and mediumsize formal urban firms in the Enterprise Surveys have utility service in most countries, this could reflect that they tend to purchase or rent properties that already have service.22 Another possible explanation for the correlation is that it reflects omitted variable bias.

20. For example, some work has shown that managers appear to find it difficult to answer questions that involve calculating percentages. Clarke (2011b) shows that when managers in sub-Saharan Africa report bribes as a percentage of sales, they report bribe payments that are between four and fifteen times higher than when they report them in monetary terms. 21. See footnote 5. 22. This does not mean that most enterprises in low-income countries have access to infrastructure services. For example, based on a nationally representative survey of microenterprises and small enterprises in Zambia, Clarke and others (2010) found that only 6 percent of microenterprises in rural areas and 24 percent in urban areas were connected to the public

how and why does the quality of infrastructure service delivery vary?

389

Table 14.11 Correlations with Electricity Obstacles Percentage of Firms Saying Electricity Is Serious Problem Access to electricity Per capita electricity consumption Per capita electricity production Cost of electricity connection Price per kWh for electricity (household) Price per kWh for electricity (industrial) No. of required procedures to get electricity connection No. of days to get electricity connection Losses due to power outages No. of power outages per month Percentage of firms with generators Electric power transmission and distribution losses

–0.24** (0.04) –0.38** (0.00) –0.32** (0.00) 0.16* (0.06) –0.38** (0.05) 0.01 (0.97) –0.03 (0.72) 0.26*** (0.00) 0.53*** (0.00) 0.56*** (0.00) 0.43*** (0.00) 0.30*** (0.00)

***, **, * = statistically significant at 1%, 5%, and 10% significance levels.

Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

grid. In contrast, all of the medium-size and large formal enterprises in the sample had both electricity connections and public water supply. The medium-size and large enterprises were more similar to the types of firms in the Enterprise Surveys. Indeed, firms in the Enterprise Surveys are not explicitly asked whether they have utility connections; it is implicitly assumed that they do when they are asked questions about infrastructure services.

390

George R. G. Clarke

In contrast to the access and availability indicators, the price indicators are much less strongly correlated with firms’ perceptions about electricity. When measured in dollar terms, there is a weakly significant positive correlation between the price of a business connection and firms’ perceptions about electricity.23 That is, firms appear to be more concerned about power in countries where the price of a business connection is higher. In contrast, the price per kWh for business users is not correlated with perceptions about power, and the price per kWh for households has a counterintuitive negative sign (i.e., firms are less concerned about power in countries with high electricity prices for households). Although this could reflect that service is of poorer quality in countries that do not charge households a sustainable price, it is important to note that the correlation between price and service quality is weak and statistically insignificant in most cases. Finally, there is a strong correlation between most measures of service quality and perceptions about electricity. Managers say that electricity is a greater problem in countries where it takes longer to get a new connection, where outages are more common and cause greater losses, where firms are more likely to have generators (a sign that reliability is a problem), and where transmission and distribution losses are higher.24 Transportation In contrast to the electricity-related variables, few of the transportation-related variables are significantly correlated with managers’ perceptions about transportation (table 14.12). Of the two measures of access to infrastructure—rail density and road density—only rail density is significantly correlated with perceptions about transportation. Managers in countries with greater rail density were less likely to say that transportation was a major or very severe obstacle. Similarly, most of the proxies for transportation costs are not significantly correlated with perceptions about transportation. In particular, the price of gasoline and diesel are uncorrelated with concerns about transportation. Moreover, the perception-based measures of costs from the Logistics Performance Index (i.e., the percentage of firms that said that rail, road, and port costs were high or very high) are also uncorrelated with the percentage of firms that said that transportation is a major or very severe obstacle. The only measure of cost that is correlated with perceptions about transportation is the cost of importing a 20foot container. Managers were significantly more likely to say transportation was

23. In contrast, when measured as a percentage of gross national income, the correlation is strong and more highly statistically significant. This is the way the data are presented in the Doing Business indicators. This could reflect that the second measure, but not the first, is highly negatively correlated with income. 24. This is consistent with the similar results reported by Gelb and others (2006), who note that firms complain more about power in countries where outages are more common.

Table 14.12 Correlations with Transportation Obstacles Percentage of Firms Saying Transportation Is Serious Problem Rail density Road density Cost of importing 20-foot container Price of diesel Price of gasoline Rail transport rate Road transport rates Port charges Losses to breakage or spoilage during shipping No. of days to complete import procedures Quality of port services Quality of rail services Quality of road services Percentage of roads that are unpaved Liner shipping connectivity index (maximum value in 2004 = 100)

–0.42** (0.00) –0.15 (0.33) 0.29** (0.00) –0.09 (0.33) –0.15 (0.11) –0.04 (0.73) 0.09 (0.40) –0.16 (0.14) 0.24** (0.05) 0.29*** (0.00) 0.14 (0.18) –0.12 (0.27) 0.11 (0.31) –0.04 (0.81) –0.32*** (0.00)

***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Note: See appendix 14.1 for full variable descriptions. P-values in parentheses. Source: See appendix 14.1 for data sources.

391

392

George R. G. Clarke

a serious problem in countries where the cost of importing a 20-foot container is high. Finally, the measures of quality are also mostly uncorrelated with perceptions about transportation. The coefficients on the measures of perceptions about the quality of service (i.e., the percentage of firms that said that the quality of port services, rail services, and road services was low or very low) are statistically insignificant in all cases. Similarly, the coefficient on the percentage of unpaved roads is also statistically insignificant. The only two statistically significant coefficients are related to the ease of engaging in international trade. Managers were less likely to say that transportation was a major or very severe obstacle in countries with better shipping connectivity and in countries in which it takes less time to import goods. Although these results might suggest that the main concern with respect to transportation is related to imported materials, it is important to note that these measures are highly correlated with similar measures related to exports (i.e., the cost of exporting a 20-foot container, the number of days to export a container).25 Further, shipping connectivity will also affect ease of exporting as well as ease of importing. In this respect, although the results suggest that transportation issues related to international trade are important, it is not clear that importing dominates exporting in terms of the effect on managers’ perceptions. EMPIRICAL MODEL

For the main regressions, we regress the two measures of perceptions about infrastructure (perceptions about electricity and transportation) on a set of macroeconomic control variables and a set of variables related to the relevant infrastructure services. Perceptions ! " # $ macroeconomic controls # % institutional quality # & infrastructure # ' The macroeconomic and institutional dependent variables are described in the previous section. In addition, measures of access, price, and quality of electricity and transportation services are added. In practice, including the infrastructure variables significantly reduces sample size: the different measures are not all available for all countries. Therefore, the focus is on those variables that are statistically significantly correlated with perceptions. When multiple measures are significantly correlated with perceptions within the same class of variables (e.g., price, quality of service, and access), the variables with the greatest coverage are generally selected. For example, per capita electricity consumption rather than access is used because the first vari-

25. The simple correlation between the two cost variables (import and export) is 0.94, and the simple correlation between the two time measures is 0.95.

how and why does the quality of infrastructure service delivery vary?

393

able is available for significantly more countries. If access were included rather than per capita electricity consumption, the regression would only have about 61 rather than 82 observations. ECONOMETRIC RESULTS

Table 14.13 presents results from the econometric analysis.

Table 14.13 Impact of Macroeconomic and Sector Variables on Perceptions About Infrastructure Observations

Percentage of Firms Saying Electricity Is Problem 120

Per capita electricity consumption No. of days to get electricity connection Cost of electricity connection (business) No. of power outages per month Cost of importing 20-foot container Liner shipping connectivity index (high values mean better connected) Rail density Urban population Area Population, total Per capita gross national income Exports of goods and services

82

Percentage of Firms Saying Transportation Is Problem 120

90

44

0.006*** (2.74) –0.166* (–1.81)

0.004 (1.50) –0.136* (–1.97)

0.172** (2.04) –1.087 (–0.87) 0.907 (0.61) –1.780 (–0.91) 0.069 (0.93)

0.796 (1.38) 0.256** (2.46) 2.563* (1.77) –1.444 (–0.90) –1.933 (–0.93) 0.128* (1.98)

–0.001 (–0.63) 0.077*** (3.32) 0.749 (0.36) 0.788*** (5.84)

0.337** (2.40) –3.553** (–2.06) 1.494 (0.80) –10.366*** (–3.25) –0.093 (–0.79)

0.504*** (3.40) –0.320 (–0.18) –0.256 (–0.13) –5.400 (–1.34) –0.086 (–0.74)

0.156** (2.03) 0.639 (0.68) –1.918* (–1.89) –3.181* (–1.82) 0.002 (0.03)

(continued)

394

George R. G. Clarke

Table 14.13 (continued) Observations

Control of corruption (high values mean less corruption) Constant R-squared

Percentage of Firms Saying Electricity Is Problem 120

82

–4.634 (–1.39)

0.541 (0.14)

128.832*** (4.55)

45.562 (1.08)

0.216

0.488

Percentage of Firms Saying Transportation Is Problem 120 –3.376* (–1.84) 61.969*** (3.99) 0.146

90

44

–2.222 (–1.03)

–5.360** (–2.45)

16.569 (0.78)

3.016 (0.12)

0.250

0.668

Note: T-statistics in parentheses. ***, **, * = statistically significant at 1%, 5%, and 10% significance levels. Source: See appendix 14.1 for data sources.

Macroeconomic Variables Before controlling for the quality and availability of infrastructure services, the coefficient on per capita income is negative and statistically significant in the regressions for both electricity and transportation. This suggests that managers are more likely to say that electricity and transportation are serious problems in low-income countries. After adding the infrastructure variables, however, the coefficients become smaller in absolute value and become statistically insignificant. In addition, the R-squared of the regressions increases, suggesting that managers’ perceptions of sectoral problems are informed by objective measures of difficulties with access and quality; that is, managers do not believe service is poor just because they believe that low income always means poor service. The coefficient on the percentage of the population living in urban areas is positive and statistically significant in both regressions. This suggests that managers are more concerned about electricity and power in countries with larger urban populations. For the most part, the quality and price of infrastructure service were not significantly correlated with the urban population share, as discussed earlier. Availability was higher in countries with larger urban populations, but since the Enterprise Surveys only cover urban areas in most countries, it is not clear that this should affect perceptions. For transportation, it is possible that this reflects congestion: countries with larger urban areas might be more likely to be congested. However, this would not explain the positive correlation between urban population share and the percentage of managers that identify electricity as a serious problem. After controlling for income and other macroeconomic variables, firms appear to be more concerned about transportation in countries where corruption is a greater problem. The coefficient on the “control of corruption” variables—with higher values meaning less corruption—is negative and statistically significant.

how and why does the quality of infrastructure service delivery vary?

395

This might suggest that the quality of transportation infrastructure is worse in countries with more corruption, perhaps because the quality of roads and other transportation infrastructure is worse or because the cost is higher in corrupt countries. If firms pay bribes to win government contracts, money will end up being diverted from the national treasury to the pockets of corrupt bureaucrats (Bardhan 1997).26 Similarly, corruption can affect the quality of construction when firms are able to bribe inspectors and regulators to avoid meeting contract provisions or quality standards. Quality will also suffer if firms pay bribes to avoid meeting technical requirements specified in the bidding documents.27 Infrastructure Services In addition to the macroeconomic and institutional variables, several objective measures are included that relate to the availability, price, and quality of infrastructure services. As noted above, the inclusion of these variables tends to restrict sample size, so only a limited number of variables were selected, based on sample availability and whether the simple correlation was statistically significant. In addition, at least one variable representing quality, access, and price was selected in each case. For electricity, per capita electricity consumption (availability), the cost of an electricity connection (price), the number of days to get an electricity connection (quality), and the number of power outages per month (quality) are included. While the R-squared more than doubles, the only statistically significant coefficients are on the variables representing quality: the number of days to get a connection and the number of power outages. These results suggest that the most important aspect of electricity service is quality and reliability. For transportation, rail density (availability), liner shipping connectivity (quality), and cost of importing a 20-foot container (price) are included. Including rail density reduces sample size considerably, so results are presented with and without this variable (see table 14.13). The coefficients on the cost of importing a 20-foot container and the index of shipping connectivity are both statistically significant, while the coefficient on rail density is statistically insignificant.28 Consistent with the previous results, this suggests that the most important aspect of transportation is the cost and ease of imports and exports.

26. Also consistent with this, Kenny (2007) shows that construction contractors are more likely to pay bribes and spend more on them than other firms. 27. Consistent with this, Kahn (2005) shows that natural disasters lead to more deaths in countries with weak institutions. He suggests that this could be because corruption leads to poorly enforced building codes and low-quality infrastructure. Anecdotal evidence is consistent with this. After the 2010 earthquake in Haiti, Billam (2010) argued that “buildings had been doomed during their construction.” The poor construction standards were attributed to corruption in procurement and building standards enforcement (Padget 2010; ScienceDaily 2010). Destruction during natural disasters in other countries has also been blamed on corruption (Kenny 2007). 28. When we include losses during transportation, all coefficients become statistically significant because including this variable significantly reduces sample size.

396

George R. G. Clarke

Conclusions This chapter looks at two questions: (1) how do different aspects of infrastructure service—availability, quality, and price—vary across countries; and (2) what aspects of infrastructure service have the greatest impact on firms? It examines the correlation between various measures of infrastructure services and the correlation between objective (and some subjective) measures of infrastructure services and managers’ perceptions about obstacles to firm performance imposed by poor-quality services. It also looks at the correlation between infrastructure services and macroeconomic variables. The analysis shows that access is highly correlated within countries for different infrastructure services (water, electricity, mobile phones, fixed-line phones, roads, and rail). In contrast, prices and service quality are not highly correlated within countries. That is, countries with poor service in one sector do not necessarily have poor service in other sectors. Similarly, although prices are often positively correlated across services (e.g., when per-minute charges for mobile phones are high, power prices are also high), the correlations are lower than for access. And price and quality and price and access are not strongly correlated within sectors (see, for example, table 14.15 for electricity).29 These results are broadly consistent with results in the 1994 World Development Report (World Bank 1994) for the 1990s. Macroeconomic variables—per capita income and population density in particular—explain much of the cross-country variation related to access to roads, rail, electricity, water, and telecommunications services. Given the weak withincountry correlations for price and quality, it is not surprising that macroeconomic variables like income and population density explain less of the cross-country variation for these variables.30 The strong correlation between income, population density, and access to infrastructure might suggest that income is destiny with respect to access: lowincome and sparsely populated countries are destined to have low levels of access. This does not mean, however, that governments can do nothing to improve access to infrastructure other than promote economic growth. As noted earlier, there is some variation in access, and even more for service quality and price, even after controlling for income and other macroeconomic factors. The recent experience with increased private sector participation in infrastructure supports this conclusion. Both cross-country econometric studies and individual country case studies show that governments can improve access and other aspects of service by introducing private sector participation, setting up in-

29. This, again, is broadly consistent with evidence from the 1994 World Development Report (World Bank 1994). 30. This is also consistent with results using data from the 1994 World Development Report (World Bank 1994).

how and why does the quality of infrastructure service delivery vary?

397

dependent regulators, and, where possible, allowing competition.31 For example, in 2006, the government of Kenya partially privatized the Kenya Electricity Generating Company (KenGen) and introduced a two-year management contract for the distribution company, Kenya Power and Lighting Company (KPLC). This, along with other reforms, resulted in connections increasing from 67,000 to 150,000 in two years (Public Private Infrastructure Advisory Facility 2010b). Similarly, starting in 2004, private and state-owned companies were allowed to bid for route-by-route contracts to provide bus service in Hanoi. This increased access to transportation and improved the quality of bus service in the city even though public subsidies were reduced (Public Private Infrastructure Advisory Facility 2010a). Although population density is correlated with access to service, the relationship between population density and the price of infrastructure services is not consistent. Given that access might be low in countries with large, spread-out rural populations because of the high cost of serving rural areas, it seems plausible that the cost of service should be higher in these countries. The insignificant relationship between population density and price of infrastructure services might be because the price of infrastructure services does not reflect the cost of providing service. That is, if prices are set by state-owned enterprises or in consultation with government-appointed regulators, they might not reflect the cost of providing service. The second part of the chapter shows that poor-quality infrastructure imposes substantial costs on firms. Enterprise managers in developing countries— and especially in low-income countries—were more likely to say that electricity is a serious obstacle than to say the same about any other area of the investment climate (e.g., access to finance, corruption, tax rates and administration, or regulation). Managers were less likely to say that transportation was a serious problem. Firm managers are most concerned about the quality of service for electricity. In contrast, most price measures are only weakly correlated with managers’ perceptions. Although managers tend to have worse perceptions about electricity in countries where access is poor, these correlations become statistically insignificant after controlling for the quality of service. Managers’ perceptions about transportation do not appear to be strongly associated with most measures of price, availability, or service quality in the transportation sector. Most of the objective indicators are not significantly correlated with managers’ perceptions about transportation. The most robust correlations appear to be related to the cost and time associated with importing materials—and potentially exporting, given the high correlation between the time and cost of exporting and importing. Firms were more likely to complain about transportation

31. Kessides (2005) and Megginson (2005) provide two recent summaries of the experience with private sector participation in infrastructure.

398

George R. G. Clarke

in countries where the cost of importing a 20-foot container is higher and where connections to international trade routes are worse. Previous studies have noted that managers appear more concerned about infrastructure in low-income countries than in middle-income countries (Carlin, Schaffer, and Seabright 2010; Gelb et al. 2006). Before controlling for quality of service, price, and access, we find similar results. The results from this study suggest that this mostly reflects that the quality of infrastructure is low in lowand middle-income countries. After controlling for this, the correlation between perceptions and per capita income becomes smaller and statistically insignificant. This might not be surprising given that, as noted above, many measures of access and service quality are positively correlated with per capita income.

references Arvis, J. F., M. A. Mustra, J. Panzer, L. Ojala, and T. Naula. 2012. Connecting to compete, 2010: Trade logistics in the global economy—The logistics performance index and its indicators. Washington, DC: World Bank. Bardhan, P. 1997. Corruption and development: A review of the issues. Journal of Economic Literature 35(3):1320–1346. Bertrand, M., and S. Mullainathan. 2001. Do people mean what they say? Implications for subjective survey data. American Economic Review: Papers and Proceedings 91(2):67–72. Billam, R. 2010. Lessons from the Haiti earthquake. Nature 463(18 February): 878–879. Carlin, W., M. Schaffer, and P. Seabright. 2010. A framework for cross-country comparisons of public infrastructure constraints on firm growth. London: University College London. Clarke, G. R. G. 2011a. Are managers’ perceptions about constraints reliable? Evidence from a natural experiment in South Africa. Journal of Globalization and Development 2(1):1–28. ———. 2011b. How petty is petty corruption? Evidence from firm surveys in Africa. World Development 39(7):1122–1132. Clarke, G. R. G., and H. T. Dinh. 2012. Introduction. In Quantitative analysis of manufacturing firm performance in Africa, ed. H. T. Dinh and G. R. G. Clarke, 1–26. Washington, DC: World Bank. Clarke, G. R. G., M. K. Shah, M. Sheppard, J. Munro, and R. Pearson. 2010. The profile and productivity of Zambian businesses. Lusaka, Zambia: World Bank. Clarke, G. R. G., and S. J. Wallsten. 2003. Universal service: Empirical evidence on the provision of infrastructure services to rural and poor urban consumers. In Infrastructure for poor people: Public policy for private provisions, ed. P. J. Brook and T. C. Irwin, 21–76. Washington, DC: World Bank. ———. 2006. Has the Internet increased trade? Developed and developing country evidence. Economic Inquiry 44(3):465–484. Dinh, H. T., D. A. Mauvridis, and H. B. Nguyen. Forthcoming. The binding constraint on firms’ growth in developing countries. In Performance of manufacturing firms

how and why does the quality of infrastructure service delivery vary?

399

in Africa: An empirical analysis, ed. H. T. Dinh and G. R. G. Clarke. Washington, DC: World Bank. Djankov, S., C. Freund, and C. Pham. 2010. Trading on time. Review of Economics and Statistics 92(1):166–173. Freund, C., and D. Weinhold. 2002. The Internet and international trade in services. American Economic Review 92(2):236–240. ———. 2004. The effect of the Internet on international trade. Journal of International Economics 62(1):171–189. Gelb, A., V. Ramachandran, M. K. Shah, and G. Turner. 2006. What matters to African firms? The relevance of perceptions data. Washington, DC: World Bank. Hallward-Driemeier, M., and R. Aterido. 2009. Comparing apples with . . . apples: How to make (more) sense of subjective rankings of constraints to business. Policy Research Working Paper 5054. Washington, DC: World Bank. Hausmann, R., and A. Velasco. 2005. Slow growth in Latin America: Common outcomes, common causes? Boston, MA: Kennedy School of Government. International Energy Agency. 2009. Energy prices and taxes: Quarterly statistics— Fourth quarter. Paris: International Energy Agency. International Telecommunication Union. 2012. World telecommunication indicators. Geneva, Switzerland: International Telecommunication Union. Kahn, M. E. 2005. The death toll from natural disasters: The role of income, geography, and institutions. Review of Economics and Statistics 87(2):271–284. Kaufmann, D., A. Kraay, and M. Mastruzzi. 2009. Governance matters VIII: Governance indicators for 1996–2008. Policy Research Working Paper 4978. Washington, DC: World Bank. ———. 2010. Response to “The worldwide governance indicators: Six, one, or none?” Washington, DC: World Bank. Kenny, C. 2007. Construction, corruption, and developing countries. Washington, DC: World Bank. Kessides, I. 2005. Reforming infrastructure: Privatization, regulation, and competition. Washington, DC: World Bank. Langbein, L., and S. Knack. 2010. The worldwide governance indicators: Six, one, or none? Journal of Development Studies 46(2):350–370. Megginson, W. L. 2005. The financial economics of privatization. New York: Oxford University Press. Megginson, W. L., and J. M. Netter. 2001. From state to market: A survey of empirical studies on privatization. Journal of Economic Literature 39(2):321–389. Padget, T. 2010. Chile and Haiti: A tale of two earthquakes. Time (1 March). www .time.com/time/world/article/0,8599,1968576,00.html. Public Private Infrastructure Advisory Facility. 2010a. PPIAF helps improve bus transport services in Vietnam’s capital. Washington, DC: World Bank. ———. 2010b. PPIAF supports private sector participation in Kenya’s energy sector. Washington, DC: World Bank. ScienceDaily. 2010. Industry corruption, shoddy construction likely contributed to Haiti quake devastation. ScienceDaily. www.sciencedaily.com/releases /2010/01/100114171539.htm. Shirley, M. M., and P. Walsh. 2000. Public versus private ownership: The current state of the debate. Policy Research Working Paper 2420. Washington, DC: World Bank.

400

George R. G. Clarke

Straub, S. 2008. Infrastructure and growth in developing countries: Recent advances and research challenges. Policy Research Working Paper 4460. Washington, DC: World Bank. Wallsten, S. J. 2001. An econometric analysis of telecom competition, privatization, and regulation in Africa and Latin America. Journal of Industrial Economics 49(1):1–19. World Bank. 1994. World development report: Infrastructure for development. Washington, DC: World Bank. ———. 2009. Enterprise survey and indicator surveys: Sampling methodology. Washington, DC: World Bank. ———. 2011a. Doing business 2012: Doing business in a more transparent world. Washington, DC: World Bank. ———. 2011b. World development indicators. Washington, DC: World Bank. ———. 2012. Enterprise surveys [standardized data 2006–2011]. Washington, DC: World Bank. www.enterprisesurveys.org.

appendix 14.1: sources of data Table 14.14 Electricity Variables Description

Year

Source

Access Access to electricity (% of population) Per capita electricity consumption (kWh per capita) Per capita electricity production (kWh per capita)

2009 2009 2009

World Bank (2011b) World Bank (2011b) World Bank (2011b)

2009

World Bank (2011a)

2008 2008

International Energy Agency (2009) International Energy Agency (2009)

Service Quality No. of power outages per month Losses due to power outages (% of sales) Percentage of firms with generators No. of required procedures to get electricity connection No. of days to get electricity connection Electric power transmission and distribution losses (% of output)

Various Various Various 2009 2009 2009

World Bank (2012) World Bank (2012) World Bank (2012) World Bank (2011a) World Bank (2011a) World Bank (2011b)

Other Percentage of firms saying power is major/very severe obstacle

Various

World Bank (2012)

Pricing Cost of getting electricity connection (% of per capita gross national income) Price per kWh for electricity (household, US$) Price per kWh for electricity (industrial, US$)

Note: Enterprise Survey data (World Bank 2012) is available for various years between 2006 and 2010 for each country.

401

Table 14.15 Transportation Variables Description

Year

Source

Access Rail density (km per sq. km) Road density (km per sq. km)

2009 2009

World Bank (2011b) World Bank (2011b)

Pricing Cost of importing 20-foot container (US$) Price of diesel (US$ per gallon) Price of gasoline (US$ per gallon) Rail transport rates (% of firms saying rates are high/very high) Road transport rates (% of firms saying rates are high/very high) Port charges (% of firms saying cost is high/very high)

2009 2009 2009 2010 2010 2010

World Bank (2011a)

Service Quality Losses to breakage or spoilage during shipping (% of output) No. of days to complete import procedures Quality of port services (% of firms saying quality is low/very low) Quality of rail services (% of firms saying quality is low/very low) Quality of road services (% of firms saying quality is low/very low) Percentage of roads that are unpaved Liner shipping connectivity index (maximum value in 2004 = 100)

Various 2009 2010 2010 2010 2009 2009

World Bank (2012) World Bank (2011a) Arvis et al. (2012) Arvis et al. (2012) Arvis et al. (2012) World Bank (2011b) World Bank (2011b)

Arvis et al. (2012) Arvis et al. (2012) Arvis et al. (2012)

Note: Enterprise Survey data (World Bank 2012) is available for various years between 2005 and 2009 for each country.

402

Table 14.16 Water and Telecommunications Variables Description

Year

Source

Access Access to improved water (% of population) Mobile phone subscriptions (% of population) Fixed phone subscriptions(% of population)

2009 2009 2009

World Bank (2011b) International Telecommunication Union (2012) International Telecommunication Union (2012)

Pricing Price of 3-min. peak call (fixed line, US$) Price of 3-min. peak call (cellular, US$) Price of fixed-line connection (business, US$)

2009 2009 2009

International Telecommunication Union (2012) International Telecommunication Union (2012) International Telecommunication Union (2012)

Quality No. of water shortages per month Faults per 100 fixed-line telephones

Variousa 2009

World Bank (2012) International Telecommunication Union (2012)

Note: Enterprise Survey data (World Bank 2012) is available for various years between 2005 and 2009 for each country. a Various years between 2006 and 2010.

Table 14.17 Macroeconomic Variables Description

Year

Source

Urban population (% of population) Area (sq. km) Population, total Per capita gross national income (purchasing power parity, adjusted constant US$) Exports of goods and services (% of gross domestic product) Control of corruption (high values mean less corruption)

2009 2009 2009 2009

World Bank (2011b) World Bank (2011b) World Bank (2011b) World Bank (2011b)

2009 2009

World Bank (2011b) Kaufmann, Kraay, and Mastruzzi (2009)a

See www.govindicators.org.

a

403

404

Price per kWh for electricity (industrial)

Price per kWh for electricity (household)

Cost of electricity connection (business, % of gross national income)

Cost of electricity connection (business, US$)

Per capita electricity production

Access to electricity Per capita electricity consumption

Table 14.18 Pairwise Correlations for Electricity Variables

1.00 0.50*** (0.00) 0.52*** (0.00) –0.24** (0.04) –0.55*** (0.00) –0.10 (0.69) –0.36 (0.16)

Access to Electricity

0.98*** (0.00) –0.22** (0.01) –0.23** (0.01) 0.24 (0.13) –0.20 (0.20)

1.00

Per Capita Electricity Consumption

–0.24*** (0.01) –0.24*** (0.01) 0.12 (0.44) –0.27* (0.08)

1.00

Per Capita Electricity Production

0.33*** (0.00) 0.38** (0.02) 0.22 (0.19)

1.00

Cost of Electricity Connection (Business, US$)

appendix 14.2: correlations of additional variables

–0.19 (0.22) 0.03 (0.85)

1.00

Cost of Electricity Connection (Business, % of gross national income)

405

Electric power transmission and distribution losses (% of output)

Percentage of firms with generators

No. of power outages per month

Losses due to power outages

No. of days to get electricity connection

No. of required procedures to get electricity connection

0.07 (0.56) –0.41*** (0.00) –0.40*** (0.00) –0.22* (0.07) –0.20* (0.10) –0.27*** (0.02)

–0.21*** (0.02) –0.19** (0.04) –0.43*** (0.00) –0.33*** (0.00) –0.41*** (0.00) –0.32*** (0.00)

–0.20** (0.02) –0.19** (0.04) –0.41*** (0.00) –0.31*** (0.00) –0.41*** (0.00) –0.33*** (0.00)

0.09 (0.22) 0.07 (0.39) 0.13 (0.15) 0.08 (0.35) 0.02 (0.84) 0.16 (0.09)* (continued)

–0.05 (0.53) 0.13* (0.09) 0.45*** (0.00) 0.24 (0.01)* 0.22 (0.01)** 0.09 (0.32)

406

1.00 0.68*** (0.00) –0.29* (0.06) 0.13 (0.42) –0.10 (0.63) 0.15 (0.49) –0.05 (0.83) –0.20 (0.19) –0.07 (0.66) 0.18 (0.29) 0.14 (0.48) 0.65*** (0.00) 0.29 (0.16) 0.25 (0.12)

1.00

0.26*** (0.00) 0.08 (0.36) –0.01 (0.94) –0.04 (0.67) 0.06 (0.49)

1.00

0.07 (0.45) 0.05 (0.60) 0.13 (0.15) 0.10 (0.29)

1.00

0.60*** (0.00) 0.43*** (0.00) 0.36*** (0.00)

1.00

0.48*** (0.00) 0.18*** (0.11)

1.00

0.25** (0.02)

1.00 1.00

Price per kWh Price per kWh No. of Required No. of Days to Losses Due to No. of Power Percentage Electric Power for Electricity for Electricity Procedures to Get Electricity Power Outages Outages per of Firms with Transmission and (Household) (Industrial) Get Electricity Connection (% of Sales) Month Generators Distribution Losses Connection (% of Output)

Note: P-values in parentheses. ***, **, * = statistically significant at 1%, 5%, and 10% significance levels.

Electric power transmission and distribution losses (% of output)

Percentage of firms with generators

No. of power outages per month

Losses due to power outages

No. of required procedures to get electricity connection No. of days to get electricity connection

Price per kWh for electricity (household) Price per kWh for electricity (industrial)

Table 14.18 (continued)

commentary Ahmed Abdel Aziz It is a challenge to model the variation of infrastructure services, particularly when they represent different sectors (electricity, water, telecommunication, transportation) and when they are measured across several countries. The work requires the study of various service indicators (e.g., for access/availability, price, and quality), the various independent and macroeconomic variables that affect the services (e.g., population, per capita income, exports), and the relationship between the indicators and the dependent variables. Both robust statistical analysis and good interpretation of results are required to make sound conclusions. In his chapter, George R. G. Clarke provides insightful discussions and conclusions explaining the variation among infrastructure services and the impact of a set of macroeconomic variables on them. Clarke uses a well-established source of data—the World Bank—to collect infrastructure services data for a large sample of countries. Through his research, he finds that access to infrastructure services is highly correlated within countries and that the price and quality of services are not highly correlated. He also documents that per capita income and population density play more important roles in explaining the variation in access to infrastructure services than they do in explaining the price and quality of the services. Clarke goes a step further by explaining the perception of company managers regarding variation in infrastructure services, finding that electricity is a more serious concern for managers than other services, such as transportation, or investment climate indicators like access to finance and corruption. Clarke presents great insights in the research discussion and draws significant conclusions from the results. Although he uses correlation and regression analyses in his work, further quantitative analysis to thoroughly investigate the various infrastructure sectors would buttress his conclusions. Building confidence in the results of regression analysis requires checking on various measures such as R-squared, T-statistics, and sample size, which reveal some weaknesses in the statistical analysis. For example, the coefficient of determination (R-squared) was relatively reasonable in identifying the variation of the availability of infrastructure services (table 14.3); however, it was weak, close to zero, for the price and quality of the services (tables 14.6 and 14.9). The statistical significance of the coefficients of the macroeconomic variables (as measured by the T-statistic) was reasonable for the coefficients of the availability measures (table 14.3), but not for the prices of the services (table 14.6). These findings underpin his policy conclusions. Good results require large sample sizes, which were available for some of the infrastructure measures (e.g., access to improved water, mobile phone subscriptions, cost of importing a 20-foot container). However, the small sample size for other measures (e.g., price of electricity) raises some doubts about the results for the particular indicators. Confidence in the results also requires looking at the 407

408

Ahmed Abdel Aziz

adequacy of the independent variables and checking if their behavior is consistent and well explained among the indicators and across sectors. While this was achieved in some variables, a number of macroeconomic variables had inconsistent behavior. For example, urban population and total population (table 14.6) had different impacts on the price indicators for the same sector (telecommunications), where the impacts changed the sign among the indicators. Although urban population and total population seem to be related, they had different impacts on the cost of connecting electricity (table 14.6) and on electric power transmission and distribution losses and shipping losses (table 14.9). Exports had negative impacts on nearly all the availability indicators (table 14.3). These surprising behaviors raise doubts about the adequacy of the variables used in the analysis, the type of regression, and the existence of correlation among the independent/ macroeconomic variables. In summary, in order to increase confidence in the results, we should explore ways of obtaining better and more meaningful analysis and results. We should look for multicolinearity among the independent variables; the existence of correlation among them would affect the stability of the coefficients. Along with the multiple regression analysis, it would be worthwhile to investigate the use of ridge regression and partial least squares regression to deal with the correlation among the independent variables. The adequacy of the macroeconomic variables and the infrastructure service indicators should also be examined. Perhaps other more representative variables and indicators could be used, such as unemployment rate, inflation, political status, and education. Furthermore, the models that Clarke presents covered the whole world, but perhaps regional/classified models could produce better results; cluster analysis could be used for that. However, because the data will be constrained (the world has a limited number of countries), we may need to consider how to get the best out of small sample data sizes or investigate other research methods. Finally, the data used in the work were mainly for 2009, and as it is highly likely that infrastructure services will change over time, more complex time-series regression analysis or panel data sets might be required.

CONTRIBUTORS Editors gregory k. ingram President and CEO Cochair of Department of International Studies Lincoln Institute of Land Policy

karin l. brandt Research Analyst and Project Administrator Interdepartmental Programs Lincoln Institute of Land Policy

Authors alex anas

david j. crapo Managing Partner Crapo | Smith PLLC

mirjam de bruijn Professor and Director African Studies Centre University of Leiden

louise nelson dyble Assistant Professor of History Social Sciences Department Michigan Technological University

nancey green leigh

Professor Department of Economics State University of New York at Buffalo

Professor School of City and Regional Planning Georgia Institute of Technology

janice a. beecher

victor a. matheson

Director Institute of Public Utilities Michigan State University

Associate Professor Department of Economics College of the Holy Cross

george r. g. clarke

robert picciotto

Killam Distinguished Associate Professor of Economics Division of International Banking and Finance Studies A. R. Sanchez Jr. School of Business Texas A&M International University

gary c. cornia Dean and Professor Marriott School of Management Brigham Young University

Visiting Professor Department of War Studies Kings College London

anu ramaswami Professor Department of Civil Engineering University of Colorado Denver

felix rioja Associate Professor Department of Economics Georgia State University 409

contributors

410

katherine sierra

dolores koenig

Senior Fellow Brookings Institution

Professor Department of Anthropology American University

yan song Associate Professor Department of City and Regional Planning University of North Carolina at Chapel Hill

david m. levinson

lawrence c. walters

Senior Fellow Price School of Public Policy University of Southern California

Stewart Grow Professor Romney Institute of Public Management Brigham Young University

Commentators ahmed abdel aziz Associate Professor Department of Construction Management University of Washington

Professor Department of Civil Engineering University of Minnesota

richard g. little

david e. luberoff Senior Research Associate Rappaport Institute for Greater Boston Harvard University

w. ross morrow

Independent consultant

Assistant Professor Departments of Economics and Mechanical Engineering Iowa State University

timothy j. brennan

don pickrell

Professor of Public Policy and Economics Department of Public Policy University of Maryland, Baltimore County

Chief Economist Volpe National Transportation Systems Center

alain bertaud

j. fred giertz Professor of Economics Institute of Government and Public Affairs University of Illinois at Urbana-Champaign

anthony m. townsend Research Director Institute for the Future

waheed uddin Professor Department of Civil Engineering University of Mississippi

INDEX Abu-Lughod, J. L., 199, 202 accountability, regulation and, 105 Actor Network Theory, 62 Adams, A., 139 Adams Express Co. v. Ohio State Auditor, 134, 137 Adaptation Fund, 280 “adders,” 109 Adolphson, D. L., 141, 142 advanced manufacturing, 320n1, 321 Advanced Manufacturing Partnership, 320–321 Advisory Commission on Intergovernmental Relations (ACIR), 134 Africa: displaced persons in, 256, 256n18; Facebook in, 76–77; mobile phone companies, 67–68; mobile phone coverage, 63–64; mobile phone jobs, 73; mobile phones, 61–80, 83–84; mobile phone service delivery, 67–68; population growth, 269, 270t; refugees in, 256n18; rural areas, 63, 66; social media in, 75–77; text messaging in, 78; urban areas, 63 African Development Bank, 255 African socialism (Ujamaa), 238 African Studies Centre, University of Leiden, 61n1, 80 Agénor, P.-R., 353 Agostini, S. J., 234 Ahmedabad, India, 11 airline travel allocation, 298 air pollution, 367–368 airtime sales, in Africa, 69–71, 77, 78 Airtouch Communications, Inc. v. Department of Revenue, State of Wyoming, 146 Akosombo Hydroelectric Project, 15, 249 Alexander, I., 112 Allmers, S., 220t, 230 Alonso, W., 19 Altshuler, A. A., 203, 233, 234 American Planning Association, 323n3 American Public Power Association, 97t American Recovery and Reinvestment Act (ARRA), 90, 350 American Society of Civil Engineers (ASCE), 17, 90, 329, 330, 345, 348

American Water Works Association, 91 American Water Works Company, 110n15 Anas, A., 13, 155, 156, 157, 157n1, 158f, 159, 160f, 164f, 165, 166, 168, 172, 179, 182–185 Anas-Xu model, 156 Angel, S., 19, 257 Angola, 72 Angwafo, T., 65 antitrust investigations, 98, 124 Apgar, W. C., 155 appeals, property valuation, 130, 146–148 Applegate v. Ernst, 133 Appolinaire (mobile phone repairman), 73–74 apportionment, 145–146 Arab Financing Facility for Infrastructure (AFFI), 16 Archambault, J., 78 Ardita-Gomez, A., 284n24 Argonne National Lab, GREET model, 307 Arnott, R. J., 165, 166, 168 Arrow, K. J., 350 Arvis, J. F., 372 Aschauer, D. A., 351 Asia: population growth, 269, 270t; urban industry, 341 Asian Cities Climate Change Resilience Network, 288 Asian Development Bank, 255, 287n29 assessment ratios, 145 Assessor of Robert Mills County v. Unit Drilling, 147 Astrium GEO-Information Services, 227f Aterido, R., 388 Athens Olympics, 222 Atherton, T. J., 166 Atkinson, R. D., 320, 322 Atlanta, Georgia, Olympic Games, 14, 229; sports infrastructure, 224–225, 226 AT&T, 98, 123–124 Austin, Texas: Climate Protection Plan, 288–291; sustainable energy infrastructure, 283–284, 288–291, 292 Austin Energy, 289–291, 292 Austin Icons Revisit MCC, 289 Australia: cap-and-trade policies, 273; eminent domain powers, 245; National

411

412 Australia (cont.) Water Initiative, 274; sustainable development, 287 automobiles: energy use per mile, 309; fuel efficiency, 172; type choice, 161 Averch-Johnson (“AJ”) effect, 107–108 Avery, K. B., 269 aviation infrastructure investment, 37–38 Ayaba (student), 69–70 Aziz, A. A., 407 Baade, R., 220t, 222, 223t, 224, 225f, 226f, 230 “back to the city” movement, 327 Bailey, E. E., 106, 107, 118 banks: climate funds, 282; loans, in China, 31–32, 49 Banks, J. P., 276t, 290, 291 Banobras (Banco Nacional de Obras y Servicios Públicos), 280 Barabas, A., 248 Barcelona Summer Olympics, 229 Bardhan, P., 395 Barrett, P., 194 Barro, R., 350–351, 353 Bartolome, M., 248 Basheda, G. N., 113 Batchelor, S., 71, 72 Baumann, R., 220t Beaver County, Utah, 131, 148, 149 Beaver County v. Property Tax Division of the Utah State Tax Commission, 138, 147 Beaver County v. WilTel, Inc., 135 Beecher, J. A., 11, 13, 87, 96, 101, 123 behavioral feedback, to promote energy conservation, 289, 305, 306 Beijing, 14, 50 Bell Laboratory, 106 Benmaamar, M., 359n1 Bennet, M., 205 Bennett, L., 198, 199, 202, 203 Bennett, M. I. J., 202 Benson, B. I., 246 Berg, S., 116 Berlin, K., 280, 282 Berra, Y., 221 Bertaud, A., 341 Bertrand, M., 387 Berube, A., 322 Betancur, J. J., 205 Betten, N., 191

Index beyond-net systems, 336 bicycles, 302–303, 309–310 Biewald, B., 113 Biggar, D., 123n1 Bilal, H., 77, 78 bilateral development assistance, 11 Biles, R., 202 Billam, R., 395n27 Blackley, D. M., 168 BNDES (Banco Nacional de Desenvolvimento Econômico e Social), 280 Bockstael, N. E., 41, 45 Bogotá, Colombia, 11 Bolivia, 335t, 354 Bonbright, J. C., 104 bond financing, 10 Bonniwell, D., 199–200 Born, P. R. D., 351 Borrero, O., 11 Boston Gas Company v. Board of Assessors, 138 Boston Redevelopment Authority, 324t Boulder, Colo., 305 Bradford, P., 113 Brandt, K. L., 1, 7, 11, 255 Brazil: FIFA World Cup, 219, 222, 226; infrastructure financing, 11, 352; involuntary resettlement in, 251–252; Summer Olympics, 226 Brazilian Football Confederation, 222, 227 bread and circuses, 230 Brennan, T. J., 109, 115, 123 Brenner, N., 207 bribery, 388n20 Briner, G., 282, 283 Brinkman, I., 61n2, 62, 77, 78 Briscoe, J., 248n13 British Standards Institute, 299 broadband infrastructure needs, 91 Brookings Hoover Institution, 290–291 Brown, J., 278n14, 282 brownfield development, 327; industrial redevelopment, 331–334, 345–346; infrastructure concessions, 213; remediation costs, 332, 333 Brox, J. A., 356, 356n6 Brueckner, J. K., 35 Brundtland Commission, 271, 271n3 Bruton, M. J., 39 Bruton, S. G., 39 Bruzelius, N., 203 BSI (British Standards Institute), 299

Index BTS (Bureau of Transportation Statistics), 309 Buder, S., 191 budget approach to infrastructure maintenance, 358, 359 Buhl, S., 198 building energy consumption: efficiency, 276t, 303; factors affecting, 302; greenhouse gas emissions, 301, 306; mitigation measures, 303–305, 306 build-operate-transfer (BOT) schemes, 14 Bujagali Dam, Uganda, 248n13 Bureau of Economic Analysis, NAICS, 95t Burke, E., 205 Burkina Faso, 65n4, 335t Burnham, D., 194 business rental market, 166 bus rapid transit (BRT): in China, 309; in Mexico, 284–286, 291 bus travel, 309 Butterfield, W., 12f, 259 C40 cities: master plans, 274–275; mitigation measures, 275f; sustainable development tools, 276–277 C40 Climate Leadership Group, 272, 272n5 cable companies: property taxation of, 135, 149 Calderon, C. A., 8, 9, 21, 236 California State Board of Equalization, 132 Callahan, L., 327, 328, 334 Calumet: Chicago Skyway and, 193, 195, 199–200; economic decline of, 199– 200; economic development of, 206; economy, 197; Indiana East-West Toll Road and, 191, 193; industrial growth of, 191; isolation of, 200; Loop highway connection, 193; transportation issues, 191–194 Calumet Skyway, 196f. See also Chicago Skyway Cambodia, 251 Cameroon: mobile money in, 72; mobile phone jobs in, 73–75; mobile phones in, 65–66, 69–70 Canada, 356 Canadian Survey of Capital and Repair Expenditures, 356 cap-and-trade policies, 273, 282 capital intensity, of public utilities, 93, 95f capital leases, 136 Caravani, A., 283

413 carbon credits, 15–16 Carbon Disclosure Project, 272, 274, 274n11, 276t carbon emissions. See greenhouse gas (GHG) emissions carbon footprint, 272–273 carbon markets, 279, 282–283 Carbon Partnership Facility (CPF), 283 Carey, J. M., 112 Carlin, W., 386n18, 398 Carlock, G., 273 Carrion-Flores, C., 39 Castells, M., 63, 65n4, 207 CBC, 221 CDC Associates, 324t Celtel, 67–68, 70–71 Center for American Progress and Global Climate Network, 282 Central African Republic, 63 Central Artery (Big Dig) project, Boston, 13, 14 central business districts (CBDs): Chicago Skyway and, 190; housing location choice, 172; transportation infrastructure and, 50; urban highways and, 199 central government infrastructure financing, 30–31 centrally assessed properties: appraisal cycles, 148; electric assets, 130–131; GDP and, 128; property taxes for, 127; property valuation, 126–131; as revenue source for local governments, 128 Centre on Housing Rights and Eviction, 233 Cernea, M., 240, 248, 249, 263, 264 Certified Emission Reduction (CER), 283 Chad: Facebook and, 76–77; mobile money, 72; mobile phones, 63, 70–71, 77, 78 Chapman, C. M., 132, 136, 138 Charles River Associates, 166 Chatterton, I., 5 Chavez, A., 296, 297, 298, 299, 300, 301 Chen, C., 12f, 259 Chen, Z., 269 Chernick, P., 113 Chicago: Crosstown Expressway, 201; “downtown versus the neighborhoods” orientation, 201–202; economic decline, 199; economic development, 201, 202–203, 205; as entrepreneurial city, 202; as global city, 202–203, 207; infrastructure development, 203; infrastructure leases, 204–206; local traffic

414 Chicago (cont.) congestion, 194; neighborhood human infrastructure fund, 204; population shifts away from, 199; racial issues, 201; regional highway system, 194; service-based economy, 202; transportation issues, 191–194; urban highway funding, 193, 201; urban planning, 194; urban renewal, 201; Wrigleyville neighborhood, 224, 225f Chicago and North Western R.R. Co. v. State, 137 Chicago Area Transportation Study (CATS), 200 Chicago Central Area Committee (CCAC), 200–201 Chicago City Council, 205 Chicago Department of Aviation, 206 Chicago Department of Planning and Development, 206 Chicago Department of Streets and Sanitation, 200 Chicago Loop: Calumet connection to, 193; economic development of, 200–201; support for, 198–199 Chicago MSA: calibration of RELU-TRAN CGE model to, 166–170; central business district model, 158; RELU-TRAN2 model of, 156–159, 182–185; RELUTRAN zones, 158–159; road-pricing policies, 156–179, 182–185; traffic congestion, 156 Chicago Plan Commission, 191, 194 Chicago Skyway, 14, 189–207, 195f, 212– 214; avoidance of, 197; central business district and, 190; concession period, 213; construction, 193; decline of, 199–202; design of, 194–195; economic decline of Calumet and, 199– 200; failure of, 189–190, 199, 212; financing, 193, 195, 197–199; history, 189–194, 207; limited-access design, 194, 195; map, 192f, 196f; as megaproject, 190; original intention of, 189–194; political backlash, 204; private operation of, 203–206; purpose of, 194–195; reconstruction of, 213; redefinition of, 202–206; as revenuegenerating facility, 203–206; traffic congestion and, 194, 197–199, 212; traffic project inaccuracies, 197–198; urban development effects, 189–190

Index Chile, 274 China: aviation investment, 37–38; bus rapid transit, 309; cap-and-trade policy, 273; displaced persons in, 256; economic growth, 21–22; financing mechanisms, 30–33, 49; Five-Year Plans, 22, 28, 274; green stimulus packages, 278; high-speed rail, 28–29; infrastructure, urban scale, and land prices, 33–38; infrastructure capacity per capita, 48– 49; infrastructure financing, 11, 49; infrastructure inadequacies, 22, 24; infrastructure investment, 30–31, 55; infrastructure returns, 8–9; intercity migrants, 37; intra-urban railway infrastructure, 29; involuntary resettlement in, 249; Landsat images, 39–41; Magic Bullet, 54–55; monocentric model of urban growth, 33–38; National Stadium (“Bird’s Nest”), 225, 227f; public infrastructure expenditure, 259n20; railroad infrastructure investment, 27–29, 37, 54–55; regional imbalances in infrastructure investment, 22–27, 49; rural-to-urban migrants, 37; Summer Olympics, 217, 221, 233; sustainable spatial development in, 50– 51; Three Gorges Dam displacement, 15; transportation infrastructure investment, 27–29; urban growth, 33–38, 56; urban industry, 341; urban infrastructure investment, 21–51, 54–56; urban land conversions in, 38–48; Water Cube, 226, 227f China City Statistical Yearbook, 2011, 35 China Statistical Yearbooks for Land and Resources, 2010, 35 China Statistical Yearbooks for Urban Construction, 2011, 35 Chixoy Dam, 247–248 Chu, C., 168 Chupka, M. W., 90 CIF Pilot Program for Climate Resilience, 280 Cincinnati, Lafayette and Chicago Railroad: property valuation, 130–134 Cintra, 204 Cities and Climate Change: An Urgent Agenda (World Bank), 287n29 Cities and Climate Change Policy (World Bank), 287n29 City of Los Angeles, 335 City of Philadelphia, 324t

Index City of Portland, 324t City of San Jose, 324t, 326 “City Protests,” 191, 193 civic pride, 230 Civil Aeronautics Board, 117 Clapp, C., 282, 283 Clark, D., 245, 248 Clarke, G. R. G., 17, 370, 373n9, 374n10, 386, 386n19, 388n20, 388n22, 407– 408 Clean Air, Clean Jobs Bill, 305 Clean Development Mechanism (CDM), 282–283 clean energy: Austin Energy programs, 289–291, 292; Pecan Street Project, 289–290 Clean Energy Finance and Investment Authority, 280–282 Clean Technology Fund (CTF), 279, 279n17, 284, 284n24, 285 clean technology industries, 272n4 Clemenz, G., 112 Cleveland, Cincinnati, Chicago & St. Louis Railway Co. v. Backus, 133–134 Clifton, J., 119 climate change, 6; industrial infrastructure and, 318; mitigation costs, 277; mitigation financing, 277–281; population growth and, 269–270; urban industry and, 342; utility infrastructure and, 90. See also global temperature rise climate funds, 282 Climate Investment Funds (CIFs), 279 Climate Protection Agreement, 314, 317 Climate Protection Center, U.S. Conference of Mayors, 314, 314n2 Climate Protection Plan, Austin, 288–291 Climate Risks and Adaptation in Asian Coastal Megacities (World Bank), 287n29 Climate Smart Loan Program, 305 CO2 emissions, 368–369. See also greenhouse gas (GHG) emissions coal plants, 305 Coates, D., 224 Cohen, J., 236 Cohen, N. R., 198 Colchester, M., 248 Collender, G., 61n2 Collier’s magazine, 83 Colorado Interstate Gas Co. v. Property Tax Administrator Mary Huddleston, 146

415 Colorado Renewable Portfolio Standard, 305 Colten, C., 191 Comcast Corp. v. Department of Revenue, 135 ComEd (Exelon), 115n17 Commonwealth Games, 219 commuter ride-sharing programs, 309, 310 commuting time, 162 compensation: for involuntary resettlement, 238, 242, 246, 246n11, 251, 253–254, 264–265; land-for-land, 244 compensatory pricing, for utility monopolies, 103f competitive markets, 98–99 competitiveness, green growth and, 272 Comprehensive Superhighway Program 1939 (Burnham), 194 computable general equilibrium (CGE) models. See RELU-TRAN2 CGE model concession agreements, 213 Condit, C. W., 200, 201 conflict-related refugees, 256, 256n18 congestion taxes, 13 congestion tolls, 170–171 Congress for the New Urbanism, 323n3 Congressional Budget Office, 347, 354, 355t, 357, 357f Connecticut, 280–282 Consolidated Edison Co. of New York, 13 construction contractors, 374n11 construction elasticity, 169 construction standards, 395n27 consumers: car type choice, 161; fuel taxes and, 174–176; housing type choice, 161; quasi-Pigouvian tolling and, 174– 176; in RELU-TRAN2 CGE model, 159–165, 167t; residence location choice, 161; road-pricing policies and, 174–176 container shipping: cost of, 378–381; quality of service, 381–383, 392, 395; service obstacles, 391t contract-based regulation, 116 Convention on Human Rights, 245–246 converged utilities, 96 Cook County, 193 cooling degree days, 302 Cooper, A. B., 315 cordon tolling, 179 Core Area designation, 326 Corfee-Morlot, J., 282, 283 Cornia, G. C., 13, 126, 141, 142, 149, 152

416 correlation approach, in unit approach to property taxation, 144–145 corruption, 374, 388n20, 394–395 cost approach, in unit approach to property taxation, 140–142 Costello, K., 111, 115 cost functions, 7 cost indexing, 114 cost-of-service ratemaking, 123 cost-plus ratemaking, 102, 105–106 Cotula, L., 256 country income: infrastructure maintenance and, 348–349; infrastructure value and, 5–6 Coutard, O., 335, 336t Coverdale & Colpitts, 197, 198 Crapo, D. J., 13, 126, 152 crime, 386n18 Cronon, W., 193 cross country studies, 7 Crosstown Expressway, Chicago, 201 Cui, S., 309 Cutler, L., 191 Daley, R. J., 199, 200–201, 202 Daley, R. M., 202, 203–206, 212–213 dam construction, 247–250 Damodaran, A., 142 Daniels, M., 212–213 Danielson, A. L., 104 Danielson, M. N., 233 Das, V., 79 Dash, P. K., 31, 47 Da Silva, L., 236 date-certain regulations, 305 Davis, M., 14 Davis, S. C., 164 Davoodi, H., 359 Deakin, E., 157 de Bruijn, M., 9, 61, 61n2, 62, 65, 67, 68, 70n, 71n, 74f, 77, 78, 79, 80, 83–84 debt repayment, in China, 32 Declaration on the Right to Development, UN, 244 De Leuw, Cather and Company, 194 Delli Priscoli, J., 248n13 Delta Airlines, Inc. v. Department of Revenue, 136 Delta Alliance, 288 demand: elasticity of travel demand, 168, 303; income-elasticity of, 95–96; for industrial land, 327–329, 345;

Index price-inelasticity of, 95–96, 168, 183; rent elasticity of, 95–96 Demissie, F., 202 Dempwolf, C. S., 323–324 Démurger, S., 21, 33, 47, 49 Deng, T., 284n23 Deng, X., 33 Dentsu Institute for Human Studies, 220t Denver, Colorado: fleet upgrades, 308; transboundary infrastructure footprint (TBIF) accounting, 299, 300f Denver Regional Council of Governments Ride-Arrangers program, 309 Department of Taxation and Finance, 134, 149 depreciation rate, 103 deregulation: trucking, 328; unit approach to property valuation and, 149; in the United States, 13; of utilities, 97–100, 117 De Sousa, C., 332 developers: in RELU-TRAN2 CGE model, 159, 165–166, 167f; road-pricing policies and, 175, 182–183 developing countries: infrastructure financing in, 11, 352–353; infrastructure maintenance in, 350–352, 356, 358– 359, 360n9, 361–362, 366; infrastructure service delivery in, 370–398, 401–408; involuntary resettlement in, 236–260, 263–265; mega-events in, 215–231, 233–234, 256; mobile phone subscriptions, 63f; return on infrastructure investment in, 351; short-run benefits of hosting megaevents, 219–221; sporting events in, 219; sustainability in, 15–16; urban infrastructure development in, 54 development assistance agencies: infrastructure investment by, 238; involuntary resettlement by, 238; public distrust of, 238 Development Assistance Committee, DECO, 259n21 Development Plan for the Central Area of Chicago 1958, 201 Dhom Dam, 251 Diegel, S. W., 164 “digital exclusion” costs, 91 Digital Impact Group, 91 Dill, J., 309 Ding, C., 32

Index Dinh, H. T., 386, 387f DiPasquale, D., 168 direction capitalization method, 143, 146 Direct-Plus-Supply Chain, 311 discounted cash flow analysis, 143 displaced persons: equitable treatment of, 263–265; infrastructure investment costs and, 239–240; numbers of, 247– 249; resettlement of, 256–257; unified policy for, 259–260. See also involuntary resettlement distribution centers, 319, 326–328, 332 Dixit, A., 155 Djankov, S., 374n10 Doing Business Indicators, World Bank, 372, 390n23 Donner, J., 78 Dornfest, A. S., 148 Downie, A., 223t drinking water, 91, 330 driving, road-pricing policies and, 176–177 Dudley, S., 110 Durán, E., 11 Durand-Lasserve, A., 251 Dwarakanath, N., 298, 301 Dyble, L. N., 14, 189, 193, 194, 199, 204, 212–214 Earle, R., 90 Easterly, W., 236, 355 economic development: green growth, 271– 273; infrastructure-driven, 237–238; mega-events and, 224–228, 233–234 economic growth: in China, 21–22; infrastructure and, 4, 5–9, 21, 237–238; infrastructure value and, 5; telecommunications infrastructure and, 61–62 economic regulation of public utilities, 100–105, 110 Economic Research Institute for ASEAN and East Asia, 22, 24, 27, 48, 49 economic transactions, mobile phones for, 71–72 economic welfare, 360 economies of scale, 55 Economist, The, 223t, 327 Edison Electric Institute, 105n9 efficiency, 17–18 Eggers, D., 90 Egypt, 63 Eisinger, P., 202 Ekyne, S., 61n2

417 electricity: access to, 374–378, 407; clean energy programs, 288–291, 292; converged utilities and, 93; cost of, 378– 381; data sources, 401t; firm managers’ perceptions of, 386–390, 393–394t, 394–398; firm performance and, 384t, 385t, 386, 387f; generators in rural Africa, 65–66; nonavailability of, in Africa, 66n6; pairwise correlation, 404– 406t; per capita consumption, 395; quality of service, 381–383; reliability of, 371; service obstacles, 389t, 397; transmission line costs, 89 electricity infrastructure: condition of, 348– 350, 349t; inadequacy of, 89–90; infrastructure needs, 90; power lines, 350; power outages, 89–90; regulation of, 125; restructuring, 99, 100t; sustainability and, 15; urban industry and, 330–331 electric utility companies: green energy programs, 288–291, 292; market restructuring, 109; ownership of, 97f; property taxation, 127–128, 134; unit approach to property valuation, 140, 149 Elk Hills Power, LLC v. Board of Equalization, 138 embodied energy, 298 eminent domain: free market alternatives to, 246–247; involuntary resettlement and, 245–246; legal challenges to, 246–247 employer-based commuter programs, 309, 310 employment: centralization or decentralization of, 177–178, 184; changing jobs, as tax-avoidance behavior, 172; infrastructure-related resettlement and, 240; job mobility and, 182; in manufacturing, 342, 343; road-pricing policies and, 174t, 176, 177–178, 182; traffic congestion policies and, 157; urban industry and, 324, 337, 338, 341–343 energy conservation: behavioral feedback, 289, 305, 306; regulatory approaches, 304–305; time-of-sale ordinances, 304–305; voluntary actions, 304 energy consumption: in buildings, 302; embodied energy, 298; land use patterns and, 302–303; transboundary infrastructure footprint, 297–301; in transportation, 302–303

418 energy efficiency: Austin Energy programs, 289–291, 292; date-certain regulations, 305; green growth and, 271; infrastructure projects and, 15–16; loans, 305 Energy Information Administration, 106, 368f Energy Policy Act of 2005, 109 Energy Prices and Taxes database, International Energy Agency, 372 energy technology, 272–273. See also green growth; sustainable development “engineering-economics,” 93 Englehardt, B., 220t Enterprise Surveys, World Bank, 372, 386, 388, 389n22 entrepreneurial city government, 202 Environmental Audit Committee, 280 environmental issues: infrastructure problems and, 15–16; safeguards, 242; utilities and, 125 environmental services for sustainable development, 276t Equator Principles, 255, 258 equity/ethics issues: intergenerational, 93; involuntary resettlement and, 239– 240, 263–265; road-pricing policies and, 176 Erzo, S., 61n2 Escobal, J., 361 Estache, A., 259, 350 Europe: cap-and-trade policy, 273; economic issues, 279; eminent domain, 245–246; public infrastructure financing, 11 European Bank for Reconstruction and Development, 255 European Investment Bank, 280 European Union, 245–246, 335 Ewing-Thiel, E., 296 exempt property values, 145 exports: electric infrastructure and, 330– 331, 341; government support for, 320, 321–322, 333, 346; infrastructure deficits and, 323, 329, 345; infrastructure service delivery and, 374; urban industry and, 319, 338 Ezell, S. J., 320 Facebook: in Africa, 76–77; home energy use analysis, 289

Index “Failure to Act” (AMCE), 330 “Failure to Invest” (AMCE), 330 false resettlement benefit claims, 253 Fay, M., 350, 360 Feddersen, A., 220t Federal-Aid Highway Act of 1956, 201 Federal Communications Commission (FCC), 88n1, 91, 97, 99 Federal Energy Regulatory Commission (FERC), 97, 99, 109, 124 Federal Fund for Infrastructure (FONADIN), 285, 286 Federal Highway Administration, 302–303 federal highway trust fund, 193 Fédération Internationale de Football Association (FIFA), 217, 221, 222. See also FIFA World Cup Federation of Zambian Road Hauliers, Ltd., 348, 348t, 353 feed-into-grid systems, 335 Feehan, J. P., 351n2 fees for services, 32 Feldstein, M., 351n1 Feng, X., 224 FERC, 127t Ferguson, J., 79 Ferman, B., 202 Fernández-Ardèvol, M., 63, 65n4 FIFA World Cup: appeal of hosting, 234; benefits and costs, 219–230, 223t; economic impacts, 219–228; host countries, 218–219; infrastructure for, 215, 222; intangible benefits, 228–231; short-run benefits, 219–221; short-run costs of hosting, 222–223; stadium reuse, 224 “fifth world,” 65n4 financing methods, 10–13; in China, 30–33, 49 Finer, J., 220t Firman, T., 287, 288 firm managers: electricity reliability and, 371; infrastructure quality and, 371; infrastructure service delivery perceptions, 383–398; perception-based indexes of, 387–388; transportation infrastructure and, 371 firms: in RELU-TRAN2 CGE model, 159, 165 first-generation road funds, 358–359 first-mover cities, 272

Index Fitzgerald, J., 326, 328 fixed utilities, 93 fleet upgrades, 308 flood management infrastructure, 286–288 Florio, M., 111 Flyvbjerg, B., 198, 203 Fogelson, R. M., 194 food supply: energy consumption for, 298, 301; food insecurity, 241 forced displacement. See involuntary resettlement Ford Rouge Truck manufacturing plant, 334 foreign direct investment (FDI), 31 formula rate plans (FRPs), 115–116 Foster, V., 12f, 259 4R Act (Railroad Revitalization and Regulatory Reform Act of 1976), 139 Fox-Penner, P. S., 90, 119 Freemark, Y., 29f free-market, eminent domain vs., 246–247 Fresno, L. O., 39 Freund, C., 374n10 Fuchs, E. R., 202 Fuchs, R., 270 fuel consumption, 297–301 fuel intensity, 164f fuel taxes, 170, 171–172; centralization or decentralization and, 178; driving and, 176–177; equity of, 176; for highway financing, 213; impacts of, 173–179; rents and, 178–179; revenues raised by, 174–176; for road funds, 358, 359; tax-avoidance behavior, 172, 173; traffic congestion and, 156, 157; wages and, 178; welfare and, 174–176 Fujishima, S., 156 Fulani culture, 76 full-cost accounting, 102 fund of funds, 281t Fuss, M., 69 G-7 nations, 219 Gabon, 63, 386n17 Gambia, 335t Garner, R., 202 Garrison, W. L., 54, 54f, 55f Gelb, A., 386, 390n24, 398 Geller, S., 113 Gellert, P. K., 190 gender relations, mobile phones and, 78

419 general equilibrium models: for monocentric cities, 155; for polycentric cities, 156 generators, for charging mobile phones, 65–66 Geoghegan, J., 39 geographic information systems (GIS), 38–48 George, H., 245n10 Georgiopoulos, P., 315 geothermal capacity, 131 Germany: FIFA World Cup, 219, 224; International Climate Initiative, 280 Ghana: Akosombo Hydroelectric Project, 15, 249; infrastructure maintenance, 360; mobile phones in, 84 Gills, D. C., 205 Ginn, J. R., 155 Glaeser, E. I., 234 Global Environment Facility (GEF), 279 Global GT LP v. Golden Telecom, 147 “global shadows,” 79 Global System for Mobile Communications (GSM) networks, 71–72 global temperature rise: greenhouse gas emissions and, 269–270; UN goals to limit, 269–270, 277. See also climate change Glomm, G., 351 Goedhart, M., 142 Goel, R. K., 112 Gold, J. R., 231 Gold, M. M., 231 goods markets: in RELU-TRAN CGE model, 166; road-pricing policies and, 176 Gottlieb, P. D., 322 “Gov. Craig,” 193 government: infrastructure maintenance funding, 357–358; resettlement role, 257–258; urban development and, 193–194, 319 Graham, S., 190, 193, 207 Gramlich, E. M., 21 Granovetter, M., 75 Grant Thornton South Africa, 220t, 221 Great Recession: employment and, 337; manufacturing and, 320–321 Green, R. K., 168, 169 green banks, 280–282, 281t green bonds, 281t, 282 Green Climate Fund (GCF), 280

420 green energy purchases, 306 greenfield development, 332–333 greenfield toll roads, 212 green growth, 271–273; bus rapid transit, 284–286, 291; energy systems, 288–291; flood management, 283, 286–288, 291; future of, 291–292; inclusion of the poor and, 272; industrial redevelopment, 334–336; innovation and, 272; productivity and competitiveness and, 272; quality of life and, 271; resiliency and, 272; sustainable development tools, 276–277; urban infrastructure, 271–273 Green Growth Program, Jordan, 283 greenhouse gas (GHG) emissions, 16; accounting protocol, 299; building sector mitigation strategies, 306; cap-andtrade agreements, 282; global temperature rise and, 269–270; growth of, 269– 270; industrial, 343; infrastructure supply chain, 297–299; low-emission bus technologies, 284–286, 291; measuring, 296–297, 313–315; mitigation measures, 303–311, 307f, 308f, 313, 315–317; population growth and, 296; regulation of, 15–16, 308f; transboundary, 297–301, 313; transportation, 301, 302–303, 307–310, 367–368; urban, 269–270, 274–276, 294–311, 313–317; vehicle miles traveled and, 302–303; voluntary mitigation methods, 308f Greenstein, J., 361 green stimulus packages, 278 grey infrastructure: urban industry: industrial sprawl and, 334; urban development and, 318–319. See also industrial development Grobman, J. H., 112 gross domestic product (GDP): centrally assessed properties and, 128; infrastructure investment and, 7, 22–27, 23f, 258–259; infrastructure maintenance and, 350, 353, 356; private utilities and, 93–94; value added by railroads and public utilities, 95f, 126–127 gross operating revenue, 128 gross rent, 128 Grotius, H., 245 Group of 20 (G-20), 273 growth models, 350–354

Index Grübler, A., 5f Grudgings, S., 222 GSM forum, 74 Guatemala, 247–248 Guinea, 386n17 Gutfreund, D., 199 Gwilliam, K., 358, 359, 359n1 Haas, R., 366, 367f Habsatu (African woman), 75–77, 79 Hackworth, J., 207 Hahn, H. P., 78 Haiti earthquake, 395n27 Hall, T., 202 Hallward-Driemeier, M., 388 Handy, S., 309 Hanoi, 397 Hanser, P., 113 Harral, C., 350, 352 Harvey, D., 202 Hausmann, R., 387 HCLD (historical cost less depreciation) model, 140–141 He, M., 33, 35, 38 heating degree days, 302 Heggie, I. G., 348, 348t, 358 Helper, S., 321, 322, 324 Hemphill, R. C., 106 Hernández, J., 11 high-speed rail, 28–29 high-tech manufacturing, 332 highways. See roads and highways Highway Trust Fund, 213 Hillman, T., 297, 298, 299–300, 301, 302, 316 Hiramatsu, T., 155n, 156, 157, 157n1, 164f, 169, 172, 179 Hirsch, A. R., 194 historical cost less depreciation (HCLD) model, 140–141 Historic American Engineering Record, 195f Hledik, R., 90 Hoelzel, N., 323, 325t, 331 Hoffer, E., 236 Hogan, S., 205 Holm, M. S., 198 Holtz-Eakin, D., 357 Hong, Y.-H., 11 Hong Kong, 8, 11 Horst, H. A., 75 Hou, D., 109

Index housing: choice of type, 161; construction elasticity, 169; energy consumption, 289, 302, 306; energy efficiency, 306; floor space supply elasticity, 168; location choice, 172; price elasticity, 168, 183; quantity, 161; RELU-TRAN CGE model variables, 168–170; supply elasticity, 169; tracking energy use, 289; voluntary reduction of greenhouse gases, 306 Hua, C., 108 Hubbard, P., 202 Hudson, W. R., 366, 367t Huilongguan, 50 Hultman, N., 273 human rights, involuntary resettlement and, 244–245 Humphreys, B., 224 Humphreys, J., 220t Hurley, A., 191, 199 ICLEI-USA, 299, 310–311 ICOLD, 248, 248n14 Idaho State Tax Commission, 138 Idroes, I., 287, 288 Illinois Toll Highway Commission, 193 in-boundary infrastructures, 297–301 incentive-based regulation, 105–114, 110t, 118 income approach, 142–143 income-generating potential, 133 income inequality, infrastructure investment and, 9, 19, 237–238 income restoration, involuntary resettlement and, 244, 253 Independent Evaluation Group (IEG), 248, 255, 359–360 India: Commonwealth Games hosted by, 219; displaced persons in, 256; involuntary resettlement in, 248, 249, 251 Indiana East-West Toll Road, 191, 205, 212, 213; Chicago Skyway and, 194, 197 Indiana Toll Road Commission, 191 indirect cost recovery, 10 individualized travel marketing (TM), 309, 310 Indonesia, 238, 252 industrial density, 318 industrial development: in Calumet, 191; demand for land, 327–329, 345; infrastructure, 318–338, 341–346; location,

421 16; low-impact, 334–336; sprawl, 157, 318, 334. See also urban industry industrial parks, 332 infrastructure: capacity per capita, 48–49; country income and, 5–6, 348–349, 349t; depreciation rate, 351; economic growth and, 4, 5–6, 16; efficiency and, 17–18; global conditions, 3–19; locating, 16; performance determinants, 373–374; sustainability and, 15; technology, 4–6; urban development and, 21–56; value of facilities, 5 infrastructure asset management system (AMS), 366–367 infrastructure designer-operators, 304 infrastructure-driven development, 237–238 infrastructure financing, 10–13; bus rapid transit, 285–286, 291; in China, 49; debt repayment, 32; in developing countries, 352–353; infrastructure maintenance, 354–363; innovative, 277–281; social risk and, 255; sustainable development, 283–291; tax rate and, 352–353 infrastructure financing mechanisms. See financing methods infrastructure investment: to attract private investment, 39, 48–51; benefits of, 6–9; in China, 21–51; in developed countries, 22, 23f; economic growth and, 6–9, 21, 237–238; GDP and, 22–27, 258–259; human welfare and, 9; income inequality and, 9, 19; infrastructure maintenance and, 350–354; involuntary resettlement for, 236–260, 263–265; poverty reduction and, 237– 238, 350; by private sector, 6, 7; productivity and, 7–9; by public sector, 7; rate of return on, 8; ratio of maintenance to, 356; return on, 351; social benefits of, 9; sources of, in China, 30–31; for sustainable spatial development, 50–51; urban growth and, 18, 21; urban land conversions and, in China, 38–48. See also infrastructure maintenance infrastructure leases, 204–206 infrastructure maintenance, 17, 347–362, 366–369; budget approach to, 358, 359; country income and, 348–349; depreciation rate and, 351; in developing countries, 350–352, 356, 358–359,

422 infrastructure maintenance (cont.) 360n9, 361–362, 366; economic rate of return for, 354–358; economic welfare of residents and, 360; environmental sustainability and, 367–368; federal funding of, 356–358; financing, 352–362; fuel taxes for, 358, 359; GDP and, 350, 353; infrastructure asset management system, 366–367; infrastructure investment and, 350–354; life cycle analysis, 367; neglect of, 347– 350, 353–354, 354–355, 360–361; operating costs and, 347–350, 366, 368; poverty and, 350, 360; premature reconstruction and, 350; productivity and, 347; ratio to new investment, 356; road funds, 358–360; service disruptions and, 349; state and local funding of, 356–358; sustainable issues, 366–369; theoretical model, 350–354; user charges, 359; user costs, 348 infrastructure service delivery: access, 374– 378, 407; firm managers’ perceptions of, 386–395; firm performance and, 383–395; macroeconomic variables, 372–375, 377–387, 378t, 380t, 382t, 383t, 384t, 385t, 392, 393t, 394–395, 396, 403t; per capita income and, 371; performance measures, 372–383; population density and, 371; price measures, 378–381; quality of, 370– 398, 401–408; service availability, 374–378; service quality, 381–383 INGAA Foundation, 91 Ingram, G. K., 1, 7, 11, 155, 255 Inspection Panel, World Bank, 242 intangible property: excluding from property taxes, 137–138; unit assessment and, 136, 145 Inter-American Development Bank, 255, 360; Planet Banking, 282 intercity migrants, 37–38 intergenerational equity issues, 93 International City/Country Management Association, 323n3 international climate finance, 279–280 International Climate Fund, United Kingdom, 280 International Committee on Large Dams (ICOLD), 248, 248n14 International Economic Development Council, 323n3

Index International Energy Agency (IEA), 269, 269n1, 273n6, 277, 277n12; Energy Prices and Taxes database, 372 International Financial Corporation (IFC), 255 International Monetary Fund (IMF), 358 International Olympic Committee (IOC), 217, 222 International Rivers Network, 257n19 International Telecommunications Union (ITU), 6f, 63, 371–372; World Telecommunication Indicators, 371–372 international trade, 229 Interstate Commerce Act, 124n2 Interstate Commerce Commission (ICC), 13, 117 interstate properties, 134 InterVISTAS Consulting, 220t intra-urban railway infrastructure, 29 involuntary resettlement, 236–260, 263– 265; alternatives to, 252; compensation for, 242, 246, 246n11, 251, 253– 254, 258, 264–265; consolidated policy framework for, 258; country-based approach to, 258; dam construction and, 247–250; development policy, 236–237; of displaced persons, 256– 257; eminent domain and, 245–246; equity issues, 263–265; ethical issues, 239–240; failure of, 252–254; government role, 257–258; history of, 238– 240; impacts of, 240–241, 249–251, 263–265; indirect effects, 244; institutional competence and, 252; land-forland strategies, 253; lessons learned, 251–254; market-based solutions, 246–247, 251; market valuation and, 245; mega-events and, 256; numbers of people affected, 247–249; participatory process and, 239–240, 252–253; policy issues, 239–240, 242–243, 258– 260; political will and, 252; poverty reduction and, 237; property rights and, 247; pseudo-settlers claiming benefits, 253; psychological impacts, 241; recommendations, 257–260; rightsbased approaches, 239, 244–245; as social engineering, 238; voluntary resettlement vs., 239–240, 257; World Bank and, 239–245, 254–255. See also resettlement; resettlement policies

Index Irwin, E. G., 39, 355 Irwin, T., 112 Istrate, E., 321, 322 Itaparica Dam, 251–252 ITT World Communications, Inc. v. San Francisco, 134–135 ITU World Communications/ICT Indicators database, 63f, 64f Jack, W., 72 Jackson, K. T., 199 Jacobs, M., 278n14, 282 Jagannathan, V., 113 Jakarta: climate change goals, 276; flood management infrastructure, 283, 286–288, 291 Jamaica, 335t, 354 Jamison, M. A., 111, 114 Janata, J. F., 138, 139, 140n2 Janson, B., 298 Japan, 245, 279 Javanese farmers, 238 Jefferis, C., 205 Jenkins, J., 274, 274n8 Jessiman, W. A., 166 jobs. See employment Johannesburg, 10 Johnson, C. L., 205 Johnson, I., 243n7 Johnson, S., 231 Jones, D. W., 199 Jones v. Southern Natural Gas Co., 138 Jordan, 283 Jozi bonds, 10 Kahn, A. E., 105n10, 106n11 Kahn, M. E., 395n27 Kain, J. F., 155 Kalaitzidakis, P., 351n3, 353, 356 Kalyvitis, 353, 356, 357, 357f Kamal-Chaoui, L., 270n2 Kamerschen, D. R., 104 Kanher Dam, 251 Kaplan, I., 205 Kashora, M., 71, 72 Katz, B., 321 Kaufmann, D., 372, 374n12 Ke, S., 33, 35, 38 Kedung Ombo Dam, 252 Keeley, J., 256 Keidel, A., 256 Kelly, T., 205

423 Kennelly, M., 194, 199, 200 Kenny, C., 374n11, 395n26, 395n27 Kenworthy, J., 302 Kenya: electricity, 397; involuntary resettlement, 238; mobile money, 72; mobile phones, 84 Kenya Electricity Generating Company (KenGen), 397 Kenya Power and Lighting Company (KPLC), 397 Kerin, P., 118n18 Kessides, I., 397n31 “K” factor, 111 Kibora, L., 78 Kihm, S., 113 Kikuyu tribes, 238 Kim, H. M., 315 Kim, I., 156 Kim, K., 202 Kimmel, J., 168 King, D., 56 King, M. L., Jr., 201 King, S. P., 111 Kingdom, W., 113 Klein, N., 256 Knack, S., 374, 374n12 Kniesner, T. J., 168 Koenig, D., 263 Koller, T., 142 Koval, J. P., 202 Kraay, A., 372, 374n12 Krizek, K., 309 Krueger, T., 321, 322, 324 Kuala Lumpur, Malaysia, 14 Kumar, A., 359n1 Kumar, E., 298, 301 Kurbanov, S. I., 205 Kurz, M., 350 Kushler, M., 119 Kwoka, J. E., 94 Kyoto Protocol, 283 labor hours, 161 labor market, 166 Lake Calumet Airport, 206 Lambin, E. F., 39 land: landlessness, 240; prices, 38; as private property, 245n10; titling programs, 250; transfer fees, 32; value capture, 10–11 Land Acquisition Act, Australia, 245 land-for-land compensation, 244

424 Landsat images, 39–41, 39n1 land use: density, 303; energy consumption and, 302–303; in RELU-TRAN2 CGE model, 182–183; travel behavior and, 302–303; urban industry and, 324–326, 325t, 341–345 land use conversion model: China, 33–38, 43–47; road-pricing policies and, 182 Langbein, L., 374, 374n12 Lankao, P. R., 286 Lanthier, P., 119 Laos, 248n13 laptop computers, 84 Latin America, 11, 350 Latour, B., 62 Lavado, R., 108 Lawrence, M., 205 leased properties, 136 Lea Valley, 233 Leigh, N. G., 17, 318, 323, 324, 324t, 325t, 326, 327, 328, 331, 341–345 Leipziger, D., 238 leisure, in RELU-TRAN2 CGE model, 161 Leonard, R., 256 Lerman, S. R., 166 Leseur, A., 282, 283 Levinson, D. M., 54, 54f, 55f, 56 Li, S.-M., 48 Li, Y., 39 Library of Congress, Geography and Map Division, 56, 57, 58, 59 life cycle analysis (CLA), 367 lighting, 334 Ligthart, J. E., 351 Lin, J., 238 Lin, S., 47 Ling, R., 75 Liston, C., 112 Little, R. G., 212 Little River/Little Haiti industrial district, Miami, 325 Liu, J., 33, 157 Liu, Y., 39, 156, 158f, 159, 160f Liu, Z., 7, 11, 22, 255 local governments: centrally assessed properties as revenue source for, 128–131; greenhouse gas emissions goals, 274– 276; infrastructure financing, 31–32; infrastructure maintenance funding, 357–358; property assessments and, 128–131; sustainable development

Index tools, 276–277; unit approach to property taxation and, 147–148, 152–153 Logistics Performance Indicators, World Bank, 372, 390 London: mega-events and, 229; Summer Olympics, 14, 221, 222, 233; urban growth, 56; Wembley Stadium, 14, 222, 233 Long, H., 39 Long, J. G., 233 long-distance telecommunications services, 135 Longman, J., 223t long-run price elasticity of aggregate housing stock, 168 Loo, B. P. Y., 48 loop-closing systems, 335–336 Lopez, J. H., 9 Los Angeles: cap-and-trade policies, 282; Low Impact Development Ordinance, 334–335; Summer Olympics, 222– 223, 226 Loube, R., 111 low-density residential development, 318 low-emission technologies, 285 low-impact development, 334–336 Low Impact Development Ordinance, Los Angeles, 334–335 low-income people: displacement of, 14–15; green growth and, 272; involuntary resettlement and, 263–265; utility expenditures, 96. See also poverty Luberoff, D., 203, 233, 234 Luby, M. J., 205 Lynch, B. D., 190 Lyon, T. P., 105n10, 108, 109, 114 Maass, A., 234 Macquarie Atlas Roads, 205 Macquarie Infrastructure Group (MIG), 204, 205 macroeconomic variables, 372–375, 377– 387, 378t, 380t, 382t, 383t, 384t, 385t, 392, 393t, 394–395, 396, 403t; data sources, 403t Maennig, W., 220t, 229, 230 Magic Bullet, 54–55 maintenance-needs assessment, 367 Major League Baseball, 224, 225 major league sports stadiums, 224 Malawi, 386n17 Malaya, 238

Index Maldonado, J. K., 247n12 Mali, 66, 73 Malpezzi, S., 168, 169, 342, 343f Mandela, N., 230 Manning, M., 269 manufacturing: advanced manufacturing, 320n1, 321; brownfield development, 332; employment, 342, 343; exports, 322; federal policy, 320–321; importance of, 320; location issues, 321– 322; to order, 327; suburbanization of, 326–328, 344; trends, 326–328; types of, 322; U.S. position, 319, 320–321 MAP-21 transportation act, 213 Marchio, N., 321, 322 marginal-cost pricing, 95 marginalization, 241 market-based evictions, 251 market depreciation, 140–141 market valuation, 245 market value, 132–133 Marques, R. C., 116 Marquis, M., 269 Marshall, A., 318 Marvin, S., 190, 193, 207 mass transit: bus, 284–286, 291; fleet upgrades, 308; high-speed rail, 28–29; individualized travel marketing, 309, 310; infrastructure development in China, 29; population density and, 18; regional transportation patterns and, 302–303; switching to, as tax-avoidance behavior, 172, 177 master plans, 274–275 Mastruzzi, M., 372, 374n12 Matheson, V. A., 14, 215, 219, 220t, 221, 222, 223t, 225f, 226f, 228, 230, 233 Mathios, A. D., 112 Mathur, H. M., 263, 264 Mauvridis, D. A., 387f Maya Achi people, 247–248 Mayo, S. K., 168, 169 Mayor’s Committee for Economic and Cultural Development, 201 Mazatec people, 248 M-banking (mobile banking or mobile money), 71–72 McClendon, D., 192f McCourt, K., 198, 199 McCully, P., 257n19 McCurry, J. W., 328 McDermott, K. A., 106

425 McGrattan, E., 351n1, 351n3, 356 MCI, 98 McLean, B., 205 McMullen, J., 199–200 Meester, L., 77 mega-events: appeal of, 234; civic pride and, 230; displacement of people for, 14–15; economic impacts of, 219–228, 233–234; hosting costs, 223t; infrastructure investment, 6, 14, 215–231, 233–234; intangible benefits of, 228– 231; international trade and, 229; involuntary resettlement and, 256; longrun benefits, 224–230; rent-seeking behavior and, 233, 234; revenues from, 222–223; short-run benefits, 219–221; short-run costs, 222–223; tourism and, 221, 234; types of, 215–219 megaprojects: challenges, 6, 13–15; Chicago Skyway, 189–207, 212–214; costs, 13–14; defined, 190; displacement for, 14–15; financing, 14; pros and cons of, 206–207; rent-seeking behavior and, 233, 234 Megginson, W. L., 371n2, 397n31 Mertens, B., 39 Meschi, M., 69 Mexico: bus rapid transit, 283–286, 291; dam construction, 248 Miami, 325 Miami21, 325 microenterprises, 388, 388n22 Midway Airport, 205 Mier, R., 202 Miguel Aleman Dam, 248 Millennium Development Goals, 237–238, 250, 254, 259, 350 Miller, D., 75 Miller, D. L., 193, 200 Mills, E. S., 155 Ministry of Housing and Urban-Rural Development, China, 23f, 24f, 25f, 50–51 Ministry of Land and Resources of China, 32 Minneapolis: highway map, 57f, 58f, 59f, 60f; urban growth, 56 Missouri River, Fort Scot & Gulf Railroad v. Morris, 133 MIT Media Lab, 84 mixed-use industrial areas, 325–326 Mobile Africa Revisited, 61n1, 62, 62n3 mobile banking, 77–78 mobile money, 71–72

426 mobile phones: in Africa, 61–80, 83–84; airtime sales, 69–71, 77, 78; charging, 66; gender relationships and, 78; generators for, 65–66; “global shadows” and, 79; impacts of, 61–62; political activism and, 84; poverty and, 77–78; repair of, 73–75; SIM cards for, 69, 75; social dynamics and, 9, 10f, 62, 66, 75–79 mobile phone service: access to, 374–378, 407; companies, 67–72; cost of, 378– 381; coverage, 63–64; delivery of, 67– 68; economic development and, 61–62; economic infrastructure, 69–72, 83; economic transactions, 71–72; market unpredictability, 77; masts, 64, 66; prepaid systems, 69; quality of service, 381–383; subscriptions worldwide, 63f, 64f; success of, 4–5; technology, 73– 75, 83; wireless technology and, 64–66 Mobile Telephone Network (MTN), 65, 65n5, 68, 72, 77; “callmeback” service, 78 mode choice, in RELU-TRAN CGE model, 163 Moe, K. J., 202 Mohl, R. A., 191 Mohr, C., 309 Molony, T., 78 monocentric cities, 155 monopolies: compensatory pricing for, 103f; utilities as, 96–97, 100–101 Montaña, M., 11 Moral-Benito, E., 8 Morrison, M., 360 Morrow, W. R., 313 Morse Commission, 243n8 Moses, L., 19 Moss, Allan, 205 Motor Carrier Act of 1980, 139 Mozambique, 63 M-Pesa, 72 Muggah, R., 240 Mullainathan, S., 387 Multi-campus Research Program and Initiative (MRPI), 155n multifamily housing stock, 169–170 multi-jurisdiction companies, 134 multilateral development banks, 11, 280 Multilateral Investment Guarantee Agency (MIGA), 255 multi-stakeholder partnerships, 272–273 Munich Olympics, 229

Index municipal bonds, 10 municipal utility contracts, 116 Muro, M., 272n4, 274, 274n8 Muslim women, 78 Mustra, M. A., 372 Nagin, R., 230 Nakahigashi, M., 236 Nakhooda, S., 283 Namibia, 386n18 Nam Theun 2 hydropower plant, Laos, 248n13 Nangbeto Dam, 252 Narayanpur project, Karnataka, 251 Nataraj, G., 31, 47 National Aquatic Center (“Water Cube”), Beijing, 14 National Association of Local Government Environmental Professionals, 323n3 National Association of Tax Administrators, 132 National Audit Office, 31 National Best Workplace for Commuters Program, 309 National Bureau of Economic Research (NBER), 155 National Bureau of Statistics of China, 23f, 24f, 25f, 27f, 28f, 31f national development banks, 280 National Export Initiative, 321 National Research Council, 157, 303 National Stadium (“Bird’s Nest”), China, 225, 227f National Surface Transportation Policy and Revenue Study Commission, 350 National Water Initiative, Australia, 274 Native Americans, 238 natural disasters: construction standards and, 395n27; involuntary resettlement due to, 256 natural gas industry: converged utilities and, 93; infrastructure needs, 90–91; power plants, 305; restructuring, 99, 100t Naula, T., 372 Negroponte, N., 84 Nehru, J., 18 Nelson, J. D., 284n23 Nepal, 335t Netherlands, 287 net operating income: in discounted cash flow analysis, 143; in traditional perpetuity capitalization, 142

Index Netter, M., 371n2 networked infrastructure, 335 Newbery, D. M., 112 New Chicago, 202 New Jersey Turnpike, 198 Newman, P., 302 New Policies Scenario, 270n2 New York, 56 New York State Department of Taxation and Finance, 149 Nguyen, H. B., 387f Nikolova, M., 225f, 226f Nkwi, W., 69, 72 Nokia, 84 nonoperating properties, 136 Nordhaus, T. M., 274, 274n8 Norris, P., 5f North Africa, 66n6 North American Electric Reliability Corporation (NERC), 88n1, 90n2 North South Institute Conference on the Future of Multilateral Development Cooperation in a Changing World Order, 238n2 Norway Winter Olympics, 228 Nowak, J. E., 245 Nyamnjoh, F. B., 61n2, 62, 65 Nyden, P., 198, 199 Nyerere, J., 238 Nyerere, Julius, 238 Obadare, E., 78 Obama, B., 321 occupation permits, 251 Odendaal, N., 83 OECD, 12 off-grid systems, 335 Office of Sustainability, 288, 289n34 Office of the Gas and Electricity Markets, 112 Office of Water Services, 111 O’Hara, S. P., 191, 199 Ojala, L., 372 Oliver-Smith, A., 248, 254 Olympics, 6, 14; appeal of hosting, 234; bidding for, 229; economic impacts of hosting, 219–228; host cities, 215, 216–218t; infrastructure impacts, 215– 221, 222, 233–234; intangible benefits, 228–231; Olympic effect, 229; shortrun benefits, 219–221; short-run costs, 222–223. See also Summer Olympics; Winter Olympics

427 One Laptop per Child program, 84 operating leases, 136 operating properties, 136 Operations Evaluation Department, 244n9, 249 operations & maintenance (O&M) expenditures, 356–358 O-Power, 305 Oppenheim, J., 113 Orange (telephone company), 68, 72 Organisation for Economic Co-operation and Development (OECD), 217, 271, 272, 273n7, 275, 276t, 277, 278n13, 283; Development Assistance Committee, OECD, 259n21 Ozanne, L., 155 PacifCorp v. Idaho State Tax Comm’n, 138, 146 PacifCorp v. Property Tax Division of the Utah State Tax Commission, 147 PacifCorp v. State of Montana, 146 Pacini, E., 193 Padget, T., 395n27 Pak Mun Dam, 252 Panzer, J., 372 Papalambros, P. Y., 315 Paris Declaration of 2005, 254, 259, 259n21 parking garages, 205 participatory process, involuntary resettlement and, 239–240, 252–253 PAS, 299, 311 Patillo Industrial Real Estate, 327 Patkar, M., 249n16 Payne, B., 223t Pecan Street Project, 289–290 Peeters, T., 221 Peltzman, S., 117 Pennsylvania Turnpike, 213 per capita electricity consumption, 395 per capita income, infrastructure service delivery and, 294, 296, 371, 373, 375–376, 381, 382, 396, 398 per capita vehicle miles traveled (VMT), 302–303 perception-based indexes, 387–388 perception of infrastructure service delivery, 386–395 performance-based contracts, 360n9 performance-based regulation (PBR), 112–113

428 performance-based regulatory model (RIIO), 112 Peru, 360–361, 362 Peruvian Rural Roads Project, 360–361, 362 Peterson, C. R., 106 Peterson, G., 234 Pfeifenberger, J. P., 109, 113 Pham, C., 374n10 Philips, C. F., 141n3 phone cards, 70–71 Piatkowski, D., 309 Picciotto, R., 15, 236, 252, 263–264 Pickrell, D., 155n, 182 Pigouvian tolling, 171, 184. See also quasiPigouvian tolling pipeline companies, 134 Planet Banking, Inter-American Development Bank, 282 Plan of Chicago 1906 (Burnham), 194 pledge funds, 281t, 282 Plummer, M., 220t Points of Presence, 329 political activism: mobile phones and, 84; social media and, 76–77 political issues, in unit approach to property taxation, 147–148 Pollock, J., 106, 110 polycentric cities, 156 Ponce, C., 361 Poole, D., 79 population: climate change and, 269–270; global, 269–270, 269–270t, 296; infrastructure service delivery and, 373, 408; road-pricing policies and, 183; in urban areas, 269–270, 270t, 296 population density: infrastructure service delivery and, 371–373, 373n8, 376, 381–382, 396–397, 407; quality of infrastructure service and, 382; transportation infrastructure and, 18 Porter, M. E., 106 Portland, Oregon, 332–334 Port of Portland, 332, 333t post-networked urbanism, 335, 336 poverty: infrastructure investment and, 9, 237, 350; infrastructure maintenance and, 360; mobile phone sales and, 77–78; urban industry and, 324, 338, 342. See also low-income people power outages, 89–90 Pratt Center, 324t

Index President’s Council of Advisors on Science and Technology, 319, 320n1 Preuss, H., 223t price-cap regulation (PCR), 111–112 price elasticity of housing demand, 168, 183 price feedback, 305 price inelasticity, 95–96 Pritchett, L., 7 private infrastructure, 13 private investment: infrastructure development, 6, 7, 11, 12f, 39, 48–51; innovative methods, 286; leveraging, 280–282; in low-impact development, 334–335; in sustainable development, 279–282. See also public-private partnerships private participation in infrastructure (PPI) investment, 6; in developing countries, 11, 12f; environmental issues and, 16 private property: compensation for takings, 246; eminent domain, 245–247; land as, 245n10 private sector, involuntary resettlement and, 258–259 private toll roads, 204–207, 212–214 private utilities, 93–94 privatization of Chicago Skyway, 204–207, 212–214 production, distribution and repair (PDR) activities, 323–324 production functions, 7 productivity: efficiency and, 17–18; green growth and, 272; infrastructure and, 7–9, 337, 347; measuring infrastructure impacts, 7–9; operations & maintenance (O&M) expenditures, 356–358 profit sharing, 113–114 property rights, 247 property taxation: administration, 127– 128; of centrally assessed properties, 127; of electric utility operating companies, 127; geothermal capacity and, 130–131; intangible property and, 136, 137–138; of interstate properties, 134; of mixed-used industrial areas, 325; of private infrastructure, 13; of railroads and public utilities, 126–150, 152–154; residential property, 128; unit approach to, 126–150, 152–154. See also taxation

Index Property Tax Division, 131, 143 property valuation: appeals, 130, 146–148; apportionment, 145–146; appraisal cycles, 148; assessment ratios, 145–146; methods, 126–150, 152–154; national data, 128–130; of nonoperating properties, 136; of operating properties, 136; railroads, 131–134; by state agencies, 126–131; unit approach to, 132–133 PROTRAM, 285 prudence reviews, 106, 107–108 psychological trauma, resettlement and, 241 public interest, 91–92 Public-Private Infrastructure Advisory Facility, 12, 397 public-private partnerships: contract-based regulation for, 116; green growth, 280–282; for green infrastructure financing, 278; for infrastructure financing, 11, 12f; involuntary resettlement and, 258–259; Pecan Street Project, 289–290 public transit. See mass transit public utilities: capital intensity, 93, 94, 95f; characteristics, 87, 91–97; economic characteristics, 93–96; economic regulation, 100–105; externalities, 96; fixed utilities, 93; infrastructure replacement needs, 89; institutional characteristics, 96–97; intergenerational equity issues, 93; leased properties, 136; load characteristics, 93; as natural monopolies, 92; property taxation, 126–150, 152– 154; property valuation, 126–130; public interest and, 91–92; reliability, 92–93; restructuring, 97–100; role of, 88–91; sales comparison approach to property evaluation, 144–145; service reliability, 88; siting, 93; state valuation, 132; technical characteristics, 92– 93; unit approach to property valuation, 140; vertical integration, 93. See also utilities public works investment, involuntary resettlement for, 236–260, 263–265 Puerto, O. S., 5 Pushak, N., 12f, 259 Qatar, 219 Qin, D., 269 Qiu, J. L., 63, 65n4 quality of life, 271

429 quasi-Pigouvian tolling, 170–171; avoidance of, 173, 177; centralization/decentralization of jobs/residences, 177– 178, 179; driving and, 176–177; equity of, 176; impacts of, 173–179; rents and, 178–179; revenues, 174–176; traffic congestion and, 156; wages and, 178; welfare and, 174–176 Quest Corp. v. Colorado Div. of Property Taxation, 135 Quigley, J. M., 234 racial issues, 201 rail density, 374–378, 390, 395 Railroad Revitalization and Regulatory Reform Act of 1976 (4R Act), 139 railroads: Chicago, 193–194; China, 27–29, 37, 54–55; economies of scale, 55; industry restructuring, 139; interstate, 134; leased properties, 136; property taxation, 126–150, 152–154; property valuation, 131–134; quality of, 329–330, 345; state valuation of, 132; unit approach to valuation, 132–134, 150; urban industry and, 327–328, 329–330; in U.S., 54, 56; warehouses and, 327–328 rainwater capture, 335 Ramachandran, V., 386, 390n24, 398 Ramaswami, A., 16, 296, 297, 298, 299–300, 301, 302, 303, 304, 304f, 305, 307f, 308t, 310, 313–317 Ranjan, R., 298, 301 Ranney, D., 207 rapid transit. See mass transit Rast, J., 200, 201 rate base model, 141 rate base/rate-of-return (RB/ROR) regulation, 87, 102, 105–106, 108, 111, 112, 118 rate indexing, 114 ratemaking, 101–105; cost-of-service, 123; cost-plus, 102, 105–106; principles, 101; rate base/rate-of-return (RB/ROR) method, 87, 102, 105–106, 108; rate of return and, 104, 106, 109; social, 105; steps, 102 rate of return: formula rate plans, 115; on infrastructure investment, 8; ratemaking and, 102–106, 109 ratepayers, 101–102 rate shock, 105

430 rate stabilization plans, 115–116 rate year (test year), 102 Ravikumar, B., 351 Reagan, R., 99 real estate values, 189–190. See also property valuation real estate variables, in RELU-TRAN CGE model, 168–170 real-time energy meters, 305 Reeve, K., 296 refugees, 256 regional technology banks, 280 regulation: accountability and, 105; advantages and disadvantages of, 118; alternative models, 109–117, 110t, 118; boundaries of, 123–124; competitive markets vs., 98–99; energy conservation, 304–305; incentives and, 105– 109, 118; purpose of, 108; ratemaking, 101–105; regulatory lag, 106–107, 108f; regulatory risk, 104, 107; standards and, 105; unit approach to property taxation and, 138–139; in the U.S., 13; utility infrastructure, 87–119, 123–125 Reid, H., 286 Reinikka, R., 349 RELU-TRAN2 CGE model, 157–170; calibration, 166–170, 183; Chicago MSA application, 156–179, 182–185; consumers in RELU, 159–163; consumers in TRAN, 163–165; critique of, 182–183; developers in, 165–166; firms in, 165; fuel taxes and, 170, 171–172; interzonal road links, 159, 160f; markets, 166; purpose, 157–158; quasi-Pigouvian tolling and, 170–171; RELU (land use) component, 182–183; structure of, 159–166; TRAN (transportation component), 183; zones, 158–159 remediation costs, brownfield development, 332, 333 remote sensing data, 38–48 Rendell, E., 212–213 renewable energy, 276t; Austin Energy programs, 289–291, 292; industrial uses, 335–336; subsidies, 278 Renewable Energy and Energy Efficiency Export Initiative, 321 Renewable Portfolio Standard, Colorado, 305

Index rental market, in RELU-TRAN CGE model, 166 rent elasticity of housing demand, 168 rents, road-pricing policies and, 178–179 rent-seeking behavior, 233, 234 replacement cost new less depreciation model, 142–143 reproduction cost less depreciation model, 141–142 resettlement: funding, 253–254; history, 238–240; market-based solutions, 246–247, 251; megaprojects and, 15; participatory process and, 239–240; refusal of, 239–240; voluntary, 250– 251, 257, 265. See also involuntary resettlement resettlement policies: free market and, 246–247; inequities in, 240–241; property rights and, 247; rights-based approaches to, 239, 244–245 residence location choice, 161; road-pricing policies and, 174t, 177–178, 184–185; as tax-avoidance behavior, 172 residential property: low-density development, 318; property taxes, 128; traffic congestion policies and, 157; voluntary reduction of greenhouse gases, 306 resiliency, green growth and, 272 restructuring: electricity industry, 99, 100t; natural gas industry, 99, 100t; telecommunications industry, 99, 100t; utilities, 97–100 return on equity, 104 revenue-assurance regulatory methods, 110 revenue caps, 114 revenue decoupling, 114–115 revenue requirements, utilities, 103–105 Rhee, H.-J., 156 Rice, E., 252 Richardson, Bill, 89 Richter, F., 229 Ride-Arrangers program, Denver, 309 Rifkin, J., 335 right of use, 251 rights-based resettlement policies, 239, 244–245 Rincon de los Esteros, San Jose, 325–326 Rio de Janeiro, 217, 269 Rioja, F., 17, 347, 351, 351n4, 352 road funds, 358–360 road-pricing policies: centralization/decentralization of jobs/residences, 177–178;

Index consumer behavior and, 174–176, 182; design of, 184; developers and, 175, 182–183; driving and, 176–177; equity of, 176; impacts, 156–157, 173–179, 184–185; location effects, 170–179, 182–185; rents and, 178– 179; revenues, 174–176; wages and, 178, 182; welfare and, 174–176 roads and highways: conditions, 348–350, 348t, 349t; economic rate of return on maintenance, 354; financing, 201, 213; firm managers’ perceptions of, 386– 388, 390–392; firm performance and, 385t, 386; greenhouse gas emissions, 302–303; infrastructure maintenance vs. reconstruction, 350; maintenance financing, 358–362; quality of, 329– 330, 345; road density, 374–378, 390; service obstacles, 391t, 397; vehicle operating costs and, 348t Robert, A., 270n2 Robinson, H. M., 88 Rockefeller Foundation, 288 Rogers, R. P., 112 Roller, L.-H., 61 roll goods, 91 roofing, 334 Roques, F. A., 112 Rose, A., 229 Rothengatter, W., 203 Rothschild, M., 351n1 Rothwell, J., 272n4, 321 Rotunda, R. D., 245 route choice: in RELU-TRAN CGE model, 163–164; tax-avoidance and, 172 Rowe, J., 105n9 Rugby World Cup, 230 Rule, H., 289 rural areas: in Africa, 63, 64–66; generators for, 65–66; mobile phone masts in, 64–66 rural-to-urban migrants, 37 Russia, 217, 219 Rutherford, J., 335, 336t Rwanda, 251 Ryan, J. R., 197 Sabarini, P., 288n32 Safaicom, 72 safeguard policies, World Bank, 241–242 “safe-harbor” approach, 113 Saha, D., 272n4

431 Sahoo, P., 31, 47 Saleemul, P., 286 sales comparison approach, 143–144 Salt Lake City, 229 San Francisco, 324t Sangare, B., 66 sanitation infrastructure, 3, 4 San Jose, 325–326 Santiago, 282 Sappington, D. E. M., 112, 113 Sardar Sarovar (Narmada River) project, 243n8, 249 Sartor, O., 282, 283 Sassen, S., 202 Satterwaithe, D., 286 Savva, N., 112 scale economies, utilities, 93–95 Schaffer, M., 386n18, 398 Schalateck, L., 283 Schmalensee, R., 351n1 Schmitz, J., 351n1, 351n3, 356 Schroter, H., 119 Schumpeter, J. A., 98, 118 ScienceDaily, 395n27 Science to Achieve Results (STAR), 155n scope economies, utilities, 94 Scudder, T., 250, 265 Seabright, P., 386n18, 398 second-generation road funds, 359–360 sedum, 334 self-financing: in China, 31, 55; in U.S., 55 Seli, D., 76–77, 78 Seltzner, M., 200 Sen, A., 239 Senegal, 72 Seo, D., 112 Seoul, 275 Seoul Olympic Games, 14, 215, 216 Serven, L., 8, 9, 21, 236, 355 service area, 373 service disruptions, 349 sewer systems, 330 Sey, A., 63, 65n4 Shah, M. K., 386, 390n24, 398 Shalizi, Z., 358, 359 Shanghai, 56 Shanghai Metro, 33 shared rides, 309 Shellenberger, M., 274, 274n8 Shenzhen City, Guangdon Province, 39–48 Shikou Dam, 252 Shin, J., 112

432 shipping containers. See container shipping Shirley, M. M., 371n2 shopping trips, 161 Shum, Y.-M., 48 Sierra, K., 15, 269, 273, 281t, 286, 286n26 Silverstein, A., 88n1 Simarmata, H., 287, 288 Simatovic, R., 361 SIM cards, 69, 75 Simpson, D., 205 simulation road-pricing models, 155–179, 182–185 Singapore, 8 single-family housing stock, 169–170 Sloan, D., 223t slums, 250 Smart Grid 2.0, 290 Smart Growth Leadership Institute, 323n3 smart growth movement, 323, 325t, 327, 344 Smart Growth Network, 323n3 “Smart Growth’s Blindside” (Leigh and Hoelzel), 323 smartphones, 73–74, 84 Smart Regulations, Boulder, Colo., 305 Smith, A., 347 Smith, B. A., 169 Smith, C., 194 Smith, N., 207 Smolensky, E., 234 SoCal Edison, California, 305 Soccer City, Johannesburg, South Africa, 225, 228f soccer World Cup. See FIFA World Cup Sochi, Russia, 217 social disarticulation, 241 social dynamics: of infrastructure investment, 9; involuntary resettlement and, 238–239; mobile phones and, 62, 66, 75–79; text messaging and, 78; voice communication and, 75–76 social engineering, 238 social media, 75–77 social network theory, 75 social policy, 241–243 social ratemaking, 105 social regulation of utilities, 110 social risk, 255 social safeguards, 241–242, 249 social services, 241 Société Nationale d’Electricité du Cameroun (SONEL), 66

Index Soehodo, S., 275, 286 Solomon, L. D., 205 Solomon, S., 269 Song, Y., 8, 21, 33, 35, 38, 41, 54, 56 South Africa: crime, 386n18; FIFA World Cup, 219, 222, 224; mega-events, 221; mobile phones in, 63, 78, 84; Rugby World Cup, 230; Soccer City, Johannesburg, 225, 228f South African World Cup, 219 Southeast Chicago, 200 South Korea, 278 South Side Chicago, 199 south-south financing, 11, 12f sovereign bond financing, 10 Spatial Plan (World Bank), 286n28 Spatz, D., 194, 200, 201 Special Economic Zone (SEZ) (China), 39 Spiegel, M., 229 Spivey, J. M., 194 sporting events, 219 sports infrastructure: economic development and, 224, 233–234; for megaevents, 222–223, 233; reuse of, 225–227 Sprint, 98 squatters, 253 Squires, G. D., 198, 199 stadium economics, 224 Stapleton, G. P., 205 State Council Development Research Center Information Network, 26f, 30f state governments: boards of equalization, 146; green banks, 280–282; infrastructure maintenance funding, 357–358; property valuation by, 126–131; unit approach to property taxation and, 152–153 Stern, S., 106 Stiles, L. A., 132, 136, 138 Stilwell, F., 205 St. Lawrence Seaway, 191 stock-and-debt method, 143, 146 storm water management, 91, 334 Strategic Hamlets program, 238 Straub, S., 8t, 370n1 structured competition, 116–117 Struyk, R. J., 155 sub-Saharan Africa: bribery in, 388n20; electricity, 66n6, 386; infrastructure investment, 9, 10f, 11; south-south financing, 12f

Index suburban industry, 326–328, 344 Sudan, 77, 78 Suhrbier, J. H., 166 Sullivan, A. M., 155 Summer Olympics: benefits and costs of hosting, 219–230; history of, 215–218; host cities, 215, 216–218t; hosting costs, 223t; infrastructure impacts of, 215. See also Olympics; Winter Olympics Sun, H., 39 Surbakti, I., 287, 288 surface transportation, 329–330, 345 Suri, T., 72 sustainable development, 15–17, 269–292; affordability, 290–291; building energy efficiency, 276t; bus rapid transit, 284– 286, 291; defined, 271; energy infrastructure, 283–284, 288–291; environmental services, 276t; flood management infrastructure, 283, 286–288, 291; future of, 291–292; industrial redevelopment, 334–336; industry trends, 335– 336; infrastructure financing, 283–291; infrastructure investment for, 50–51; infrastructure maintenance and, 367– 368; leveraging private investment for, 280–282; national strategies, 273–274; private investment in, 279–280; renewable energy, 276t; resettlement issues and, 258; transboundary infrastructure footprint and, 301; transportation, 276t; urban design, 276t; urban infrastructure, 271–273; urban strategies, 274–276; urban transportation, 283–286 Sustainable Sydney plan, 275 Suttles, G. D., 201 Svensson, J., 349 Sydney, 222, 275 Sykes, R., 277 Sylla, F. S., 71, 72 Szymanski, S., 221 Tabital Pulaaku association, 76, 139 Tanzania, 78, 238 Tanzi, V., 359 Tawney, L., 274, 274n8 taxation: congestion taxes, 13; infrastructure financing and, 32, 352–353. See also fuel taxes; property taxation; unit approach to property taxation

433 tax-avoidance behavior: quasi-Pigouvian tolling and, 177; under traffic congestion policies, 172–173 Tax Equity and Fiscal Responsibility Act of 1982, 139 Taylor v. Secor, 133 technology: barriers, 278–279; high-tech manufacturing, 332; telecommunications infrastructure and, 64–66 Techwood Homes, 14 Teixeira, R., 227 telecommunications industry: antitrust investigations, 98; property taxation of, 134– 135; regulation of, 123–125; restructuring, 98, 99, 100t; telephone company property taxation, 134–135; unit approach to property valuation, 140 telecommunications infrastructure: condition, 349, 349t; economic development and, 61–62; historical development of, 4–5; mobile phones, 69–72, 83; needs, 91; role of, 3; technology and, 64–66 telecommunications services: access to, 374–378, 407; cost of, 378–381; firm performance and, 384t, 385t, 386; quality of service, 381–383 telegraph lines, 134 tenure security, 251 Tesla, N., 83 test year (rate year), 102 Texas Workforce Commission, 289n34 Texier, P., 287 text messaging, 78 Thailand, 252 Thatcher, M., 99 The Henry Ford, 334 thermoplastic olefin (TPO) tiles, 334 “Third Industrial Revolution, The” (The Economist), 327 Three Gorges Dam, China, 15 Tignor, M., 269 time allocation, in RELU-TRAN2 CGE model, 161–162 time-of-sale ordinances, 304–305 T-Mobile USA, Inc. v. Utah Tax Comm’n, 137 Togo, 252 Tokyo, 11, 282 toll roads: Chicago Skyway, 189–207, 212– 214; cordon tolling, 179; greenfield, 212; privatization of, 204–207, 212– 214. See also quasi-Pigouvian tolling

434 Toronto, 332 total factor productivity, 356–358 tourism, 221, 234 Townsend, A. M., 83–84 Trade Promotion Coordinating Committee, 321 traditional perpetuity capitalization model, 142 traffic congestion: Chicago MSA, 156; congestion tolls, 170–171; delays, 170; excess fuel consumption, 170; fuel taxes, 170, 171–172; policy impacts, 173–179; tax-avoidance behavior, 172–173; urban sprawl and, 157 traffic projections, 197–198 traffic-related CO2 emissions, 368–369 Train, K. E., 166 transboundary GHG emission footprint, 299–300, 310 transboundary infrastructure footprint (TBIF), 297–301; accounting method, 299; avoiding double counting in, 299; convergence of, 299–300, 315; for Denver, 299, 300f; measurement, 297, 313–315; policy, 315–317; policy relevance, 300–301; standardization of, 310–311; value of, 315–317 transformative strategies: enabling environment for, 273–277; local policy framework, 274–276; local policy toolkit, 276–277; national policy framework, 273–274 transit. See mass transit Transit/Employment Residential District Overlay, 326 transportation: bus rapid transit, 284–286, 291; data sources, 402t; energy consumption, 302–303; firm managers’ perceptions of, 386–388, 390–398, 394t; firm performance and, 385t, 386; greenhouse gas emissions, 301– 303, 367–368; greenhouse gas emissions mitigation measures, 307–311; industrial costs, 328; land use patterns and, 302–303; maintenance financing, 358–362; mitigation measures, 303–305; in RELU-TRAN2 CGE model, 183; service obstacles, 391t, 397; sustainable design, 276t; transboundary infrastructure footprint, 297–298. See also mass transit; roads and highways

Index transportation companies, 140 transportation infrastructure: historical development, 4; investment in China, 27–29; involuntary resettlement for, 247; leases, 204–206; maintenance, 17; population density and, 18; role of, 3; sustainability and, 15; urban growth patterns and, 9, 16, 50; urban planning and, 50–51. See also mass transit; railroads; roads and highways travel: elasticity of demand, 168, 303; voluntary changes to reduce greenhouse gases, 308f Trebing, H. M., 109 Trembath, A., 274, 274n8 Tri-State Expressway (Chicago area), 191, 194, 197 trucking deregulation, 328 Tsui, K., 49 tsunami, 256 Tu, C. C., 224 Tunisia, 335t Turner, G., 386, 390n24, 398 Turner, M. A., 155 Twain, M., 54 Uddin, W., 366, 367, 367f, 368f, 369f Uganda: carbon credits, 15–16; dam construction, 248n13; mobile phone jobs in, 73 Understanding the Economy: Promising Signs of Recovery in Manufacturing, 320 undeveloped land, 174t unemployment: infrastructure-related resettlement and, 240; in RELU-TRAN2 CGE model, 160 Union Pacific Railroad v. Utah State Tax Commission, 136, 138 unit approach to property taxation, 126– 150, 152–154; administrative issues, 145–146; benefits of, 132; context of, 126–131, 149; correlation, 144–145; cost approach, 140–142; defined, 133; defining the unit, 134–135; determining what should be valued, 139–140; direct capitalization method, 144, 146; excluding intangible property, 137–138; HCLD model, 140–141; history of, 133–134; income approach, 142–143; political issues, 147–148; rate base model, 141; regulations and,

Index 138–139; replacement cost new less depreciation model, 142–143; reproduction cost less depreciation model, 141–142; sales comparison approach, 143–144; stock-and-debt method, 143, 146; traditional perpetuity capitalization model, 142; types of, 132–133; unit contents, 135–137; unresolved issues with, 152–153; valuation methods, 139–145 unitary valuation methods, 146–148 United Kingdom: compensation for takings, 246n11; green banks, 280; International Climate Fund, 280; price-cap regulation in, 111, 112; utility regulation in, 87 United National High Commissioner for Refugees, 256n18 United Nations, 256, 270f, 271, 278n14, 286, 296; Conference on Environment and Development (Earth Summit), Rio de Janeiro, 269; Declaration on the Right to Development, 244; Environment Program, 270f, 270n2, 271; Framework Convention on Climate Change, 279; Habitat, 287n29; infrastructure-driven development and, 237–238; Millennium Development Goals, 237–238, 250, 254, 259, 350; temperature rise goals, 269–270, 277; World Commission on Environment and Development, 271n3 United States: deregulation in, 13; economic issues, 279, 320–321, 337; economic rate of return on road maintenance, 335t, 354; eminent domain, 245; FIFA World Cup, 219, 222; green stimulus packages, 278; infrastructure deficits, 323, 329, 337, 345; infrastructure replacement, 89; interstate highway system, 55; large infrastructure projects, 13–14; price-cap regulation in, 111; railroads, 54; rate base/rate-of-return (RB/ROR) method, 102; regulation, 13; self-financing in, 55; utility infrastructure conditions, 89–90; utility infrastructure regulation, 11, 13, 87 University of Chicago Library Map Collection, 196f urban agglomeration theory, 318 urban areas: in Africa, 63, 64; bus rapid transit, 284–286, 291; cap-and-trade

435 markets, 282; carbon footprint reduction efforts, 272–273; carbon markets, 282–283; China, 21–56; energy consumption characteristics, 301; entrepreneurial city government, 202; greenhouse gas emissions, 16, 269–270, 296–311, 313–317; greenhouse gas mitigation strategies, 303–311, 314– 317; growth of, 3, 9, 16, 18–19, 21; industrial policy, 321–326; infrastructure service delivery, 373–374, 376, 394, 408; land conversions, 38–48; land prices, 33–38; land use patterns, 302–303; master plans for sustainability, 274–275; mitigation measures, 275t; mobile phone masts in, 64; population growth, 269–270, 270t; simulation models, 155–179, 182–185; sustainable infrastructure, 271–273; transboundary infrastructure footprint, 297–301; transformative strategies, 273–274 urban design: industry and, 325; for sustainable development, 276t urban development: general equilibrium models for, 155; government role in, 193–194; industrial infrastructure and, 318–319; infrastructure investment and, in China, 33–38; involuntary resettlement for, 247, 250; megaprojects and, 189–190; monocentric model, 33–38; right of use during, 251; transportation infrastructure and, 50 urban highways: central city decline and, 199; financing, 193. See also roads and highways urban industry, 318–338, 341–346; brownfield development, 327, 331–334; in China, 341; climate change and, 343; conversion of, 323; electricity infrastructure and, 330–331; employment and, 324, 341–343; “geographic high road” policies, 322; government support of, 319, 337–338, 341, 342, 345– 346; greenfield development, 331–334; greenhouse gas emissions and, 343; infrastructure deficits, 323, 329, 337, 345; infrastructure needs, 329–331, 337, 345; land demand, 327–329; land economics, 341–343; land use and, 324–326, 325t, 341–345; “low road policies,” 322; manufacturing

436 urban industry (cont.) issues, 319–321, 326–328; metropolitan area policy and, 321–326; mixed use, 325–326; national policy and, 319–321; planning and zoning issues, 323–326, 324t, 344; poverty and, 324, 338, 343; railroads and, 329–330; renewable energy, 335–336; renovations for, 327; road quality and, 329–330; sewer systems and, 330; smart growth movement and, 323, 325, 327, 344; strengthening, 319–326, 336–337; suburban industry vs., 326–328, 344; sustainable development, 335–336; transportation costs, 328; transportation infrastructure and, 325–326; types of, 322; vacancy rates, 327, 343–344; value of, 319; wages, 322; warehouses and distribution centers, 319, 326–328, 332, 344; water systems and, 330; worker density, 344. See also industrial development urban infrastructure: in China, 22–27, 49, 54–56; development, 33–38; financing, 277–281; investment barriers, 278–279; networks, 273; productivity and, 337; service delivery, 373–374, 376, 394, 408; transportation, 27–29; for urban industry, 329–331, 337, 345. See also infrastructure urban land use models, 155–179, 182–185 urban planning: industrial land and, 323– 326; master plans, 274–275; transportation infrastructure and, 50–51 Urban Planning, Land and Resources Commission of Shenzhen Municipality, 41f, 43f, 45f urban renewal, 201, 251 urban sprawl, 157, 318, 334 U.S. Bureau of Economic Analysis, 128 U.S. Bureau of Labor Statistics, 342f U.S. Cellular Park, 224, 226f U.S. Census Bureau, 127, 129t U.S. Conference of Mayors: Climate Protection Agreement, 314, 317; Climate Protection Center, 314, 314n2 U.S. Congress Joint Economic Committee, 320 U.S. Department of Energy, 331 U.S. Department of Justice, 98, 124 U.S. Department of Transportation, Pipeline & Hazardous Materials Safety Administration, 90n3

Index U.S. Energy Information Administration, 88n1 U.S. Environmental Protection Agency, 88n1, 89, 91, 96, 97t, 323n3, 331 user costs, infrastructure maintenance and, 348–349, 348t, 368 user fees, 10, 359 U.S. Highway Trust Fund, 359 U.S. Manufacturing Enhancement Act, 320 U.S. Mayor’s Climate Protection Agreement, 296 U.S. Overseas Private Investment Corporation, 282 U.S. Supreme Court: antitrust cases, 124; eminent domain and, 245; unit approach to property valuation and, 133–135, 137–138; utilities and public interest, 91–92 Utah Ass’n of Counties v. Tax Comm’n, 146 Utah State Tax Commission, 131 utilities: alternative regulation models, 109–118; characteristics of, 87; commodity costs, 95; competitive markets, 98–99; consumers, 98; converged, 96; cost (rate) indexing, 114; customers, 95–96; demand, 95–96; deregulation, 97–100, 117; economic regulation, 100–105, 110; environmental issues and, 125; fixed, 93; formula rate plans, 115–116; incentive-based regulation, 105–114, 110t; load characteristics, 93; marginal-cost pricing, 95; as monopolies, 96–97, 100–101; network economics, 95; ownership of, 96–97, 97f; performance-based regulation, 112– 113; policy reforms, 98; price-cap regulation, 111–112; private ownership of, 87; profit sharing, 113–114; property taxation, 13; prudence reviews, 106, 107–108; public interest and, 91–92; public-private partnerships, 116; rate base/rate-of-return regulation, 87, 102, 108, 111, 112, 118; rate of return and, 104, 106; regulation of, 87–119, 123–125; regulatory lag and, 106–107, 108f; restructuring, 97–100; revenueassurance regulatory methods for, 110; revenue caps for, 114; revenue requirements, 103, 104t; role of, 88–91; scale economies, 93–95; scope economies, 94; social regulation of, 110; standards, 95; stock indexes, 93n5; struc-

Index tured competition for, 116–117; vertical integration of, 93. See also public utilities utility bills, mobile payment of, 72 utility infrastructure: capital replacement costs, 89; characteristics of, 88–89; costs of, 95; inadequacy of, 89–90; investment needs, 89–91; value added to GDP, 95f utility infrastructure regulation, 87–119, 123–125; evaluation of, 87–88; jurisdiction, 87; in the U.S., 11, 13 valuation appeals, 130, 146–148 valuation methods: local revenues and, 130; for railroads and public utilities, 126–150, 152–154; for unit approach to property valuation, 139–145 value capture, 10–11, 246 van Beek, W. E. A., 65 Van Berkel, R., 301 van Dijk, R., 79 Vanski, J., 155 van Wicklin, W., 252 vehicle emissions, 367–368 vehicle-fuel technologies, 307 vehicle miles traveled (VMT), 302–303, 309 vehicle operating cost (VOC), 368, 368t; road condition and, 348 Velasco, A., 387 Veldkamp, A., 39 Vella, E., 357, 357f Vermeulen, S., 256 vertical integration of public utilities, 93 Vertovec, S., 79 Vogelsang, I., 111 voice communication, 75–76 Voight, K., 220t, 222, 223t voluntary organizations, 247 voluntary resettlement, 250–251, 257, 265 Vuyani, H., 78 Wachs, M., 198 wage elasticity of labor supply, 168 wages: industrial, 322; road-pricing policies and, 178, 182 Wales, 246n11 Wallis, M., 274 Wallsten, S. J., 373, 373n9, 374n10 Walsh, P., 371n2 Walters, L. C., 13, 126, 141, 142, 149, 152 Wang, D., 49

437 Wang, L., 39 warehouses, 319, 326–328, 332, 344 Warren, M., 110 Washington, H., 201–202 Washington Consensus, 237 Washington Metro System, 13–14 waste disposal, 301 wastewater management, 91, 301 water bills, 72 water commissions, 101 Water Cube, Beijing, 226, 227f water markets, 274 water pipes, 89 water resources: access to, 374–378, 407; converged utilities and, 93–94; data sources, 403t; efficient use of, 271; firm performance and, 384t, 385t, 386; greenhouse gas emissions, 301; master planning, 275; national policies on, 274; quality of service, 381–383; regulation of, 99–100, 100t water resources infrastructure: historical development of, 4; involuntary resettlement for, 247–249, 250; needs, 91; rainwater capture, 335; replacement needs, 89; role of, 3; urban growth patterns and, 9; urban industry and, 330 water rights registers, 274 water utilities: customers, 95; as monopolies, 96; ownership of, 97f Waverman, L., 61, 69 Wealth of Nations, The (Smith), 347 weather: building energy consumption and, 302; persons displaced by, 256–257 Wein, H., 106, 106n12, 107 Weinhold, D., 374n10 Weisman, D. L., 112 Welch, R. B., 132, 136, 138 welfare: economic, 360; fuel taxes and, 174–176; infrastructure investment and, 9; road-pricing policies, 174–176 Wembley Soccer Stadium, 14, 222, 233 Wenzel, A., 283 Wessels, D., 142 Western Union Telegraph Co. v. Taggert, 134 West Nile Electrification Project (WNEP), 15–16 Weston, A. L., 132, 136, 138 Wheaton, W. C., 168 Wial, H., 321, 322, 324

438 “widow and orphan stocks,” 93n5 Wiewel, W., 199 Wille, L., 200 Williamson, H. F., 189 “willing buyer, willing seller” concept, 132 Wilson, B., 41 WilTel, 135 wind energy, 289 Winter Olympics: history of, 215–218; host cities, 215, 216–218t; hosting benefits and costs, 219–230, 223t; infrastructure impacts of, 215; intangible benefits of, 229. See also Olympics wireless technology, 64–66 Wisconsin Public Service Commission, 97n6 Wisconsin Steel, 199 Witness for Peace, 248 Wolak, F. A., 111, 112 Wold Bank, 370n Wolf, C., 99 Wolf, T., 113 World Bank, 6f, 7, 10f, 12f, 21, 22, 23, 38, 237n1, 239, 242, 243n7, 247, 249, 255, 257n19, 271, 277, 279, 284n24, 285t, 286, 286n28, 287n29, 287n30, 288n31, 348, 349–350, 349t, 354, 355, 355t, 358, 359, 360, 361, 371, 371n2, 371n3, 372, 372n5, 387, 396, 396n29, 396n30, 401t, 402t, 403t; criticism of, 254; dam projects, 247–250, 248, 248n13; displacement by projects, 247–249; Doing Business Indicators, 372, 390n23; Enterprise Surveys, 372, 386, 388, 389n22; human rights and, 244–245; infrastructure asset management system, 366; involuntary resettlement and, 239–245, 254–255; legal authority to call loans, 243n6; Logistics Performance Indicators, 372, 390; resettlement policy, 239–244, 253, 254–255; safeguard policies, 241–242; Sustain-

Index ability Infrastructure Action Plan, 280n18; sustainable development and, 280; sustainable infrastructure financing, 287; utility market structures and, 99; World Development Indicators, 371 World Bank Group, 255 World Commission on Dams, 249n15, 249n16, 257n19 World Cup (soccer). See FIFA World Cup World Cup sports events, 218n1 World Development Report (WDR), 349–350, 354, 371, 372, 396 World Economic Forum, 345 World Energy Outlook, 66n6 World Telecommunication Indicators, ITU, 371–372 WOTRO, The Netherlands, 62n2 Wrigley Field, 224, 225f Wrigleyville neighborhood, Chicago, 224, 225f Wu, W., 48 Xiao, J., 39 Xie, F., 56 Xu, R., 156 Yantan Dam, 252 York, D., 119 Yoshimo, N., 236 Zain, 67–68, 71 Zambia, 348, 359–360 Zeilig, N., 91 Zenou, Y., 33 Zhan, J., 33 Zhang, C., 32, 33 Zhang, J., 22 Zhengzhou, 50 Zietlow, G., 359n1, 360n9 Zimbalist, A. S., 233 zoning, 323–325, 324f, 344

ABOUT THE LINCOLN INSTITUTE OF LAND POLICY The Lincoln Institute of Land Policy is a private operating foundation whose mission is to improve the quality of public debate and decisions in the areas of land policy and land-related taxation in the United States and around the world. The Institute’s goals are to integrate theory and practice to better shape land policy and to provide a nonpartisan forum for discussion of the multidisciplinary forces that influence public policy. This focus on land derives from the Institute’s founding objective—to address the links between land policy and social and economic progress—which was identified and analyzed by political economist and author Henry George. The work of the Institute is organized in three departments: Valuation and Taxation, Planning and Urban Form, and International Studies, which includes programs on Latin America and China. The Institute seeks to inform decision making through education, research, policy evaluation, demonstration projects, and the dissemination of information through our publications, website, and other media. The Institute’s programs bring together scholars, practitioners, public officials, policy makers, journalists, and citizens in a collegial learning environment. The Institute does not take a particular point of view, but rather serves as a catalyst to facilitate analysis and discussion of land use and taxation issues—to make a difference today and to help policy makers plan for tomorrow. The Lincoln Institute of Land Policy is an equal opportunity institution.

113 Brattle Street Cambridge, MA 02138-3400 USA Phone: 1-617-661-3016 or 1-800-526-3873 Fax: 1-617-661-7235 or 1-800-526-3944 E-mail: [email protected] Web: www.lincolninst.edu