«April, 2002 ABSTRACT Should suburbs help finance the core public services of their central city? Previous arguments for such assistance have stressed ...»
SHOULD SUBURBS HELP THEIR CENTRAL CITY?
Andrew F. Haughwout* and Robert P. Inman**
Should suburbs help finance the core public services of their central city? Previous arguments for
such assistance have stressed spillovers from city services to suburban residents or the fact that
suburban residents (should?) care about their city’s poor. We explore the validity of a third possible argument for such assistance, one now stressed by many large city mayors. Suburban residents – even those who never use city services nor care about the city’s poor – may wish to support the city’s budget if that budget contributes to the productive efficiency of the city’s private economy and if suburban residents consume directly or indirectly the output of city firms. The analysis here presents first-difference regressions of city and suburban home values, city and suburban population, and city and suburban incomes for 217 MSA’s for the decade 1980-1990. We find that weak city fiscal institutions and increases in the rate of city poverty depress both the city’s and the suburb’s private economies. The econometric results are replicated in a general equilibrium model of an open MSA economy, calibrated to the Philadelphia MSA. Our results suggest each suburban family in an average MSA will find it in their economic self-interest to pay from $100 to $250 per year to their central cities to facilitate the reform of weak central city fiscal institutions.
* Senior Economist, Federal Reserve Bank of New York, 33 Liberty Street, New York, NY.
** Miller-Sherrerd Professor of Finance and Economics, University of Pennsylvania, Philadelphia, PA, and Research Associate, NBER, Cambridge, MA.
Numerous have colleagues contributed to this research. Most importantly, Richard Voith generously shared his city and suburban data with us without which this project would not have been possible.
We received detailed and very helpful suggestions for revisions from Bill Gale, Janet Pack, Todd Sinai, and Jake Vigdor as well as seminar participants at the American Economic Association, Brookings, Colorado, Duke, Federal Reserve Bank of Philadelphia, the New School, and Wharton.
We received very able research assistance from Silvia Ellis, A. J. Glusman, and Minsun Park. The results and conclusions summarized in this paper are solely those of the authors and do not reflect official opinions of the Federal Reserve Bank of New York, the Federal Reserve System, or the NBER.
Should The Suburbs Help Their Central City?
by Andrew F. Haughwout and Robert P. Inman Federal Reserve Bank of New York and University of Pennsylvania Should the suburbs surrounding a central city help finance their city’s core public services?
This is a long-standing policy question in U.S. urban public finance and an issue of no less importance in other developed and developing federal public economies. Toronto has recently merged its financing and governance with its surrounding suburbs, and in configuring its new fiscal system, South Africa has opted for a form of metropolitan financing which “twins” wealthy suburbs with less wealthy central cities.1 What are the arguments for such suburb-to-city fiscal assistance, or stronger still, suburb plus city fiscal mergers?
Perhaps the most familiar argument for suburban-to-city assistance is to correct for the under provision of a city produced public good which benefits suburban residents. An arguably complete list of such city services would include zoos, museums, research libraries, airports, sports stadiums, city subways and buses, and city streets; the spillovers benefits to suburban residents from city education, city police and fire protection, recreation, and trash removal services are likely to be small.2 When spillovers are significant and suburban residents benefit from a city provided public 1 For an important early analysis of this question for the United States, see Neenan (1970).
For a valuable overview of the current arguments for regional financing of city services, see Pack (2002), particularly Chapters 1 and 6.
2 Acemoglu and Angrist (1999) find that the returns from education accrue almost exclusively to the benefit of the individual student. Even if there were significant social returns in the form of higher wages for other workers, given the high mobility of U.S. workers, suburban residents today would be unlikely to benefit significantly from helping to educate children in their particular city. Similarly, there is no convincing evidence that increased city police protection offers
financing favors user fees and average cost pricing, however, not intergovernmental transfers.3 The possible exception to this rule might be financing of city streets where for administrative reasons intergovernmental transfers to fund infrastructure might be appropriate; see Small (1992). But even here, parking fees might serve be a useful “third-best” policy; see Arnott, de Palma, and Lindsey (1991). While the spillover argument is often made for general suburban-to-city fiscal transfers, city spillovers are not pervasive. When they do occur, average cost pricing or targeted capital grants rather than general transfers are the appropriate policy responses.
The second familiar argument is that city poverty is a metropolitan wide concern and suburban residents ought to contribute towards meeting the needs of their city’s poor residents.
Mark Pauly (1973) has made such an argument, noting that individual’s redistributive preferences may give added weight to lower income households in close geographic proximity. National significant benefits to suburban residents. Capitalization studies which finds a significant effect of increased police services on home values – the best market measure of a city’s public service benefits – find a positive effect only at the neighborhood level; see Thaler (1978) and Hellman and Naroff (1979). Another often mentioned “spillover” from the city to the suburbs is the movement of people and resulting suburban “sprawl.” Increased suburbanization may create inefficient congestion of suburban roads and infrastructure and may lead to excessive development of open space, but suburban aid to cities is not an appropriate policy for correcting these problems.
Brueckner (2001) provides an excellent discussion of the causes and cures for suburban sprawl.
3 When city services are “congested enough,” an appropriate fee designed to cover production and congestion costs will also provide the appropriate signal to set the efficient scale of the public service; see Mohring and Horwitz (1962) and Oakland (1972) more generally. Further, average cost pricing implemented through city user fees is often the efficient “tax” for funding city services. Average cost pricing will be preferred when the service in question is a consumption “complement” to the user’s leisure time and/or if the service is consumed disproportionately by upper income households; see Saez (2002). City zoos, museums, sports stadiums, and airports are public services which meet both criteria; again, suburban-to-city intergovernmental transfers will not needed.
governments can make supplemental contributions. In the presence of redistributive spillover benefits, however, city transfers will be too low relative to the efficient level of redistribution.
While theoretically compelling, the Pauly argument needs a credible mechanism to reveal suburban preferences for redistribution to the inner city poor. To the extent we rely upon the political process to provide a measure of these redistributive preferences, current state funding for city poverty may be efficient, at least from the perspective of the median, most likely, suburban resident. There may be a problem, but an appropriate fiscal institution – state government ! is already in place to correct it.4 Sensing perhaps the limited force of the usual spillover and redistribution arguments, U.S.
mayors have recently embraced a third argument for suburb-to-city fiscal assistance. The mayors argue that suburban residents need an economically vibrant central city if their own real incomes are to remain high and grow: “Economies don’t stop at the a city’s edge.”5 The mayors cite as evidence for this argument the now well-documented positive correlation between the average income of city and suburban residents in U.S. metropolitan areas. Their argument moves beyond correlation to causation, however. For the mayors, a weak city economy causes a weak suburban 4 An additional argument for suburban aid for city poverty is to neutralize the fiscal advantage to city poor families of relocating to a fiscally richer suburb. The suburbs might pay money to keep the poor from moving into their community. For most suburbs, however, fiscal zoning is a far more cost effective strategy for keeping the poor out of your town. Under the Mt.
Laurel decisions, New Jersey suburbs have been denied this option, so here wealthy suburbs have given transfers to their central cities to limit poor mobility; see Inman and Rubinfeld (1979).
5 Robert Lang of the Fannie Mae Foundation commenting on the recent United States Conference of Mayors meetings in Washington, D.C. See, “Cities and Their Suburbs Are Seen Growing as Units,” New York Times, July 10, 2001.
weak city has high tax rates and low public services which induces mobile firms and households to leave the city undermining the city’s private economy and thus, the mayors argue, the economy of the region as a whole.6 Suburbanites lose economically because of weak city finances. Suburbanto-city transfers which strengthen city finances will therefore strengthen the city’s economy which enhances, in turn, suburban residents’ private incomes. This paper seeks to evaluate the economic validity of this argument.
Section II presents evidence which documents the close interdependencies between central city and suburban economies for 252 U.S. metropolitan statistical areas (MSA’s), an interdependency which seems to have grown closer over the past three decades. We then seek to provide an economic foundation to the mayors’ intuition that weak city finances cause the correlations by suggesting one path through which weak city finances might depress the economic performance of both the central city and its surrounding suburbs. What links the central city economy to suburban resident welfare in our analysis is a production advantage enjoyed by center city firms from private sector agglomeration economies plus a location advantage for suburban residents from their proximity to the low cost central city. Weak city finances, which we specify here by inefficient and redistributive city fiscal institutions and regulations, leads to high city tax 6 The United States Conference of Mayors meeting in the summer, 2001, stressed first the documented interdependency between city and suburban economies and then pressed the idea that
correlation was causation running from city services to metropolitan economic performance:
Beginning in September, Mr. Morial aid he and other mayors would tour the cities to highlight the features, including a skilled work force, affordable housing, strong infrastructure and low crime rates, that they say make the metropolitan areas economically competitive (italics added). “Cities and Their Suburbs Are Seen Growing as Units,” New York Times, July 10, 2001.
thereby undoing some or all of the city’s agglomeration advantages. Production costs will rise and city firms will produce less output. Suburban residents must then pay for more for their private goods and services, either because of higher production costs (in the city or the suburbs) or because of the need to import more goods from other more distant production centers. Real incomes of city and suburban residents decline and city and suburban wealth measured by land values falls as well.
Weak city finances is the problem; for the mayors, suburb-to-city fiscal assistance is the solution.
Section III provides initial econometric evidence testing the plausibility of the mayor’s argument. Here we examine whether inefficient and redistributive central city fiscal institutions in conjunction with central city agglomeration economies might provide a causal link between the city and suburban economies. Cities are considered fiscally weak if their budgetary institutions – in particular, strong public employee unions, weak mayors, poverty obligations, redistributive tax structures – impose high tax rates and/or low public services on mobile firms and households. Not surprisingly, we find that such institutions do reduce the growth of city incomes and populations and depress city home value appreciation. More telling, we find those same city fiscal institutions also slow the growth of suburban incomes and population and depress suburban home value appreciation. The effects of weak city finances on the suburban economy are statistically significant and quantitatively important. Our estimates imply suburban households might be willing to pay from $100 to $250 per year to remove the adverse fiscal consequences of such fiscal institutions.
Section II argues the benefits of such transfers would come from protecting the agglomeration economies in center city production from the burdens of weak fiscal institutions. The