Urban areas face daunting economic challenges that have increased in scope in recent years. At the same time, cities provide exciting opportunities for growth and revitalization. The interplay of these challenges and opportunities create important tasks for policymakers and researchers.
Each year, the Brookings Institution and the Wharton School of Business cosponsor a conference to address these issues and provide cutting-edge, accessible research on issues unique to urban areas, as well as on broad economic and policy topics that have special applications in an urban setting.
The most recent conference, held at the Brookings Institution on October 25-26, 2001, sponsored two groups of papers being published this month in the Brookings-Wharton Papers on Urban Affairs (Brookings Institution Press, 2002). A symposium of three papers focuses on metropolitan tax and fiscal policy, examining the effects of political mergers between cities and suburbs, the links between the economic vitality of cities and suburbs, and firm-specific tax incentives for industry relocation. The remaining conference papers focus on changing demographics in urban areas, including the impact of alternative measures of gentrification on lower income city residents and the varying experiences of immigrant students in the New York City public school system.
Symposium on Metropolitan Tax and Fiscal Policy
Cities attract businesses and residents by providing high-quality amenities. But providing those amenities requires funding, and higher tax burdens increase incentives for city residents and firms to depart for lower tax locations. Balancing these considerations is an essential problem in urban public finance. The income disparity between most large central cities and their relatively wealthier suburbs makes these issues even more difficult and politically sensitive. In addition, the potential effectiveness of many fiscal options is unknown, and the connection between economic effectiveness and political feasibility is sometimes overlooked.
Large metropolitan areas in the United States are characterized by a very large number of local governments, with many urban areas containing more than one hundred separate municipalities. The fragmentation of local government has led to concerns regarding the distribution of government services and the efficiency with which these services are provided. Central city mayors and some analysts have advocated political and fiscal consolidation, but annexation of developed suburbs has rarely occurred.
Stephen Calabrese of the University of South Florida, Glenn Cassidy of Cassidy Policy Research, and Dennis Epple of Carnegie-Mellon University model voting behavior in multiple municipalities to evaluate the effects of mergers. Voters, who vary only in income, choose their preferred level of public services and redistribution, and the level and type of tax levied. They also choose their residential location based on these policies. In equilibrium, majority rule determines tax, public service, and redistribution policy; each municipality has a balanced budget; no one wants to move; and the housing market clears. The policy favored by the median-income voter will always be adopted. The model produces results consistent with observed patterns in cities: although both large and small municipalities provide public goods, redistribution occurs almost exclusively in large central cities. Small suburban municipalities depend primarily on property tax revenues to finance public services, but central cities use both income and property taxes. The policy choices result in income stratification across the metropolitan region. Low-income households with a preference for redistribution are more likely to locate in the central city, whereas wealthy households will choose suburbs with high levels of public service provision and less redistribution.
The stratification of municipalities by income implies that mergers are generally not politically viable. Residents of a poorer municipality, such as a central city, will support a merger with a wealthier suburb to obtain higher public good provision and redistribution, with lower overall tax rates. But residents of the wealthier suburb will oppose consolidation to avoid falling property values, reduced public good provision, and increased redistribution.
The aggregate welfare effects of mergers are more complicated. A merger between two jurisdictions will prompt the wealthiest individuals in the higher income jurisdiction to move from the consolidated city to a wealthier suburb in order to escape redistribution policies. These movers from the consolidated area will become the poorest residents in their new location and will purchase housing of less than the average value in that suburb. Housing prices in the new suburb will rise and public good provision will fall. This pattern will continue across suburbs in a domino effect. As a result, consolidation results in a negative impact on surrounding suburbs as well as for the wealthier residents of the merged municipalities.
Consolidation might still raise aggregate welfare, if mergers benefit poor voters more than they harm wealthier ones. Where this is the case, governments wishing to encourage annexations could compensate suburban residents for their losses and still improve social welfare.
Besides providing new insights into the dearth of consolidations, this paper advances researchers? ability to model simultaneous decisionmaking across multiple policy choices and offers a systematic explanation for income segregation that arises even when households have no explicit preference for the characteristics of their neighbors.
Suburban Fiscal Transfers to Central Cities
In the absence of political consolidation, financial transfers from suburbs to central cities are another, possibly more feasible, way to address metropolitan area public finance issues. But should the suburbs be interested in such an arrangement? Traditionally, proponents of such transfers have suggested that transfers are justified either because central cities fund public goods that benefit suburban residents, such as infrastructure, public education, and policing, or because central city poverty is a regional problem that should be addressed via transfers from the entire urban region.
In their paper, Andrew Haughwout of the Federal Reserve Bank of New York and Robert Inman of the University of Pennsylvania argue that neither of these arguments is compelling. They examine a new rationale for suburban transfers to central cities, based on two premises. The first is that cities create agglomeration economies. These economies occur because of the geographic concentration of firms within an industry and the resultant decline in transportation and labor costs, encouragement of innovation, and ease of spreading new ideas. These agglomeration economies reduce the cost of city-produced goods to both city and suburban residents. The second premise is that weak central city government—marked by a variety of financial practices and fiscal institutions—imposes costs on city residents and firms and induces them to relocate. The relocation, though, reduces the agglomeration economies available in the city and causes the price of city-produced goods to rise. If both premises hold, weak city governments hurt suburban residents, weak city finances cause poor suburban economic health, and suburban residents should be willing to pay to improve center cities' weak financial situation in order to preserve the benefits of agglomeration economies.
Haughwout and Inman demonstrate empirically that weak city finances are associated with negative city and suburban economic outcomes. In particular, they show that weak budgetary institutions, strong city unions, rising poverty rates and declining tax bases are associated with lower income, population growth, and rates of home value appreciation in both cities and their surrounding suburbs. They also develop a structural simulation model based on Philadelphia's economy that builds in a link between city finances and suburban economic outcomes. The effects of city finances on suburban health in Philadelphia are found to be similar to those found in the aggregate data. In this simulation a causal relation is assumed by construction, and thus the model implies that a suburban family should be willing to pay between $100 and $250 annually to improve city fiscal institutions in order to realize the benefits of agglomeration economies in the city.
The conclusion that weak city finances reduce agglomeration economies implies that transfers from suburbs to cities would only protect agglomeration economies if the funds were used to strengthen weak city finances. Such transfers would be counterproductive if the funds were used to raise pay for city workers or increase constituent services. To avoid these problems and ensure that the funds are used appropriately, Haughwout and Inman advocate the use of a number of specific mechanisms for transfers, including using suburban aid to fully fund state poverty mandates, reforming local property tax rules, and making aid dependent on the adoption of competitive bidding practices for city service contracts.
Tax Incentives and Business Location
If city-suburb consolidations are rare and suburban areas are reluctant to transfer resources to central cities, as the first two papers suggest, a third fiscal option for urban economic development is large, firm-specific tax breaks aimed at attracting or retaining particular businesses. Such actions have been highly publicized in the past, ranging from cities recruiting professional sports teams, to Alabama wooing Mercedes Benz in the early 1990s.
Despite the frequency of such actions, the research literature casts considerable doubt on the effectiveness of such incentives, on both theoretical and empirical grounds. One strand of the theoretical literature argues that under tax competition, all jurisdictions will select inefficiently low tax rates to prevent firms from exiting. This will result in a reduction of public service provision below efficient levels. Another strand of the literature argues that tax competition across communities results in an efficient allocation of resources, because people can choose where to live, and thus specific tax incentives introduce distortions. This approach assumes that if no tax incentives are offered, cities tax corporations' capital at rates equal to the marginal benefit of the public goods provided to the firms. Neither approach justifies large tax incentives for particular companies.
In their contribution to the conference, Teresa Garcia-Mila of the Universitat Pompeu Fabra in Spain and Therese McGuire of the Institute of Government and Public Affairs at the University of Illinois challenge the conventional wisdom. They develop a model in which cities compete for a mobile capital stock and benefit from productivity-enhancing agglomeration economies. Under these circumstances, the efficient tax rates on new firms equals the difference between the marginal benefit to firms of the public services they consume and the marginal agglomeration benefits to the city of the additional capital brought by the firm.
The authors then examine Chicago's decision in 2001 to offer Boeing $50 million in tax incentives to relocate its corporate headquarters from Seattle. Chicago's pursuit of Boeing is puzzling, given that it involved only the relocation of Boeing's headquarters rather than its manufacturing plants. Moreover, since most of the headquarters employees transferred from Seattle, few new jobs were created. In stark contrast, Chicago allowed a large local candy manufacturer employing nearly 1,000 people to leave the city without offering it tax incentives to stay. Garcia-Mila and McGuire suggest that Chicago planners believed that a management-oriented firm like Boeing would create greater potential for knowledge spillovers than would the retention of the manufacturing facility. The authors conclude that agglomeration of capital may have sufficient economic as well as political payoffs to justify firm-specific tax incentives.
Gentrification and Immigration
Two other papers presented at the conference provide new evidence on current issues in urban economics and urban policy. They demonstrate the breadth of topics that fall comfortably within the area of urban economics and the important insights that can be gained on urban issues from related fields of research, such as poverty and welfare or labor economics.
Does Gentrification Harm the Poor?
Although white flight to the suburbs during the second half of the twentieth century is often viewed as a causal factor in the demise of central cities, the return of affluent households to city neighborhoods sometimes elicits similarly intense criticism. Gentrification, or the influx of upper-middle class or wealthy households into previously poor neighborhoods, is popularly seen as harmful to poor and minority residents. The closing of the Cabrini-Green Housing Project in Chicago and its partial replacement with townhomes, for example, produced an uproar among long-time public housing residents, who felt they were being driven from their homes despite city officials' reassurances that mixed-income housing developments would be beneficial to low-income as well as new middle-class neighborhood residents.
A paper by Jacob Vigdor of Duke University suggests that much less is known about the impact of gentrification on poor families than is commonly supposed. Rather than assuming there is a consensus definition, Vigdor begins by defining gentrification and makes the distinction between preference-driven and income-driven gentrification. In preference-driven gentrification, high-income households raise their valuation of the amenities available in poor neighborhoods. A common example is the two-earner family that decides it prefers a shorter commute and increases its willingness to pay to live in the more central neighborhood. Income-driven gentrification occurs when a change in the productivity of high-income households raises the demand for, and hence the price of, housing in upper-income neighborhoods. This forces some of the people in these neighborhoods to move to lower priced areas.
Under both types of gentrification, housing prices rise in the formerly poor neighborhoods, so that renters there either have to move or absorb the higher rental costs (and possibly higher amenities). One difference between the two lies in housing prices in the upper-income neighborhood, which fall under preference-driven gentrification but rise under income-driven gentrification. Vigdor proposes policy options that both directly and indirectly reduce potential harm caused by gentrification. Rent subsidies or relocation assistance directly address rising costs of housing for low-income households. Job training or education subsidies could make poor residents more able to compete in the housing market.
Gentrification may also have effects on the poor apart from through the housing market. Many of these effects are likely to be positive. Rising housing prices can raise property tax revenues, increase redistribution, and improve public services. An influx of higher income households might create job opportunities for low-income residents or relocate jobs closer to the neighborhood. Poor residents might benefit from improvements in neighborhood quality, such as lower crime rates.
Vigdor argues that most work on gentrification focuses too narrowly on spatial displacement and does not in fact demonstrate that displacement is caused by gentrification or that it causes harm. Using Boston as a case study, he finds that households with low educational attainment (who are more likely to be among the long-term poor) living in gentrifying areas are no more likely to move than other households in the area or than low-education households in other areas. Gentrification has not increased the segregation of Boston neighborhoods by socioeconomic class; in fact, gentrification seems to lead to more mixed-income neighborhoods.
Immigrants and School Segregation
The proportion of the U.S. population that is foreign-born—currently about 10.4 percent—is at its highest level since 1930. Research suggests that segregation among racial groups is significant and negatively affects children's educational attainment, but it is unknown whether the same holds true for immigrants. Peer effects—proximity to low-income, less-educated classmates—appear to negatively affect racial minorities, particularly for in the areas of education and employment. Peer effects may have a different impact on the children of immigrants because although the parents generally have little education, immigrant groups often demonstrate a strong preference for educating their children. Similarly, although schools with large racial minorities have been found to receive less funding than average, no previous studies have investigated the impact of immigrant segregation on school funding. Indeed, segregation of immigrants may improve access to resources, as concentration of a group may make it more efficient for the government to provide particular services, such as classes in English as a second language.
Using evidence from the New York public schools, in which 16 percent of students were foreign-born in 1998-99, Ingrid Gould Ellen, Katherine O'Regan, Amy Ellen Schwartz, and Leanna Stiefel of New York University evaluate the degree of immigrant segregation and its relationship to resource allocation and student performance. Having assembled a data set that contains information on the academic and socioeconomic characteristics of all children in New York City public schools in 1998-99, aggregated to school level, the authors find a relatively low level of segregation for immigrants overall.
Some groups of immigrant students, particularly students from the former Soviet Union and the Caribbean, are considerably more segregated than foreign-born students overall, but still less segregated than nonwhite students. The authors show that different immigrant groups, particularly Soviet and Caribbean immigrants, have substantially different peer influences and access to educational resources. Like native students, immigrant students are highly segregated by race. Racial segregation is accompanied by peer characteristics, teacher quality, and classroom and aggregate school spending patterns that have negative effects that overwhelm differences in educational attainment due to nativity. Soviet students who attend schools with high percentages of white students have higher quality teachers and higher achieving peers, while Dominican students in predominantly black schools are educated with students characterized by extremely high poverty rates and low test scores.