Skip Navigation Planning & Markets
Subscribe Submission Requirements Editorial Board Archive Links Search Home

The Costs and Benefits of Fragmented Metropolitan Governance and the New Regionalist Policies (1)

Alex Anas
Professor of Economics
University at Buffalo, SUNY


The proponents of the new regionalism observe that central city blight is spreading to the inner suburbs and that population and employment continue to disperse and decentralize. Many ills ranging from traffic congestion to cultural deterioration are blamed on this process. The proponents believe that various forms of metropolitan governance are needed to control this process from unfolding further. I argue that fragmented governance, as we know it today, entails substantial benefits. Local governments are more competitive when there is a multitude of them, and a wider variety of local public services is offered. Also, with a multitude of local governments there is more localized power to include/exclude undesirable land uses from one's own community. I then explain the high economic costs induced by Portland's growth boundary as an older but still much applauded example of the new regionalist policies. The proposed SGECA legislation for New York State, which I also explain and analyze, should have milder effects by comparison, but it could entail substantial costs to the taxpayer.



In recent years there has been an outpouring of literature, what I call the new regionalism, supporting the notion that we need new forms of effective regional governance to combat a variety of problems plaguing U.S. metropolitan areas.

The new regionalists lament the consequences of urban sprawl and metropolitan decentralization. (2) See, for example, Downs (1994), Moe and Wilkie (1997), American Planning Assoc. (1997), Ewing (1997), Preston (1998). Some observers have claimed that sprawl causes rising vehicle miles traveled, increasing traffic congestion (Dunphy, 1997), excessive spending on roads and on traffic management (Kunstler, 1996), deteriorating air quality, depletion of resource-lands, declining inner cities, segregated minorities, rising public service duplication and costs (Orfield, 1997; Rusk, 1993), and cultural deterioration (Judd and Swanstrom, 1994). Others have disputed these stylized beliefs and have argued that policies aimed to control sprawl -- e.g. urban growth boundaries, density zoning, adequate facilities ordinances and impact fees -- can create problems of housing affordability, increased traffic congestion, economic stagnation and diminished consumer choice (see National Association of Home Builders, 1997; Gordon and Richardson, 1997). Still others (e.g. Garreau, 1991) have pointed to the economic vitality of employment sub-centers and edge cities located within sprawling metropolitan areas.

The new regionalism's answer to these problems (preceived or real) is a new form of metropolitan governance designed at the state or federal levels and supported by appropriate regulations and subsidies. What the new regionalists mean by such regional governance is a form of intergovernmental compact among local governments in the same metropolitan region and possibly also encompassing its adjoining rural areas. Such a compact would be supported as necessary by incentives and policies at the state or federal levels and it is considered desirable that it would either emulate or approximate regional government. The compacts would also make it easier to employ revenue sharing among jurisdictions at the regional and metropolitan levels.

Most of the problems the new regionalists are concerned about are not new. They have been with us for decades. For example, central cities have been declining in economic vitality since the 1920s and the blight now observed in the central cities has been accumulating gradually. Concurrently, population and employment has been decentralizing within and beyond metropolitan areas increasing the dispersion of economic activity. However, only recently have observers become concerned (see, for example, Orfield, 1997; Rusk, 1993) that central city blight may be spreading to the inner suburbs which border central cities. This "contamination" of the inner suburbs has apparently heightened the concern and lent added rigour to the cause of the new regionalists.

Any attempt to counter sprawl and decentralization should start with an understanding of why decentralization and dispersion happen. The problems of decentralization and dispersion are not uniquely American. They are observed all over the world. There are four long-standing trends which have operated to support decentralization and dispersion internationally. These are:

1) Increases in real incomes which cause households to demand larger houses and to move out where the unit price of land is lower;

2) Advances in transport technology and the expansion of roads and freeways into the suburbs and the rural hinterland of a metropolitan area, which reduces the money and time cost of travel per mile and encourages peripheral location for businesses and households;

3) Changes in production technology which favor the utilization of one-story plants and one or two-story office buildings with large parking lots, which are cheaper to locate in the suburbs;

4) Advances in information technology have reduced the needs of businesses to be concentrated in the downtowns and central cities where they sought to minimize the costs of communicating and exchanging ideas with each other.

The expansion of U.S. metropolitan areas is also shaped by additional economic and governance factors which are unique to the U.S. These are:

1) Low agricultural land values in the United States (see, for example, Mayo, 1997) means that - other things being equal - a U.S. metropolitan area of a given population will be more dispersed than its European or Japanese counterparts;

2) Tax policies permitting the deductibility of nominal mortgage costs (and policies of the Federal Housing Administration) encourage(d) suburban home buying by the middle and higher income classes. The mortgage policies especially favor richer households and more so in times of rapid inflation (such as the decade of the 1970s);

3) The under-pricing or subsidization of infrastructure systems such as roads and freeways causes households to demand low densities and municipalities to build too much infrastructure (to the extent that state and federal agencies subsidize part of the bill);

4) Suburban low density zoning exists either for economic or for fiscal reasons or because of prejudicial preferences which exclude low income residents (and, hence, also high density residential developments) from the suburbs. As a result of such zoning, more land area is needed to accommodate a given number of residents.

5) Racial prejudice has caused wealthier white residents - and, more recently, wealthier minority residents - to flee to suburban and ex-urban areas from central cities where minority poor are concentrated (see Mills and Lubuele, 1997). This, together with the other reasons for decentralization, causes blighted conditions to emerge in inner cities and their tax bases to shrink, economic conditions to decline and crime rates to rise, causing increases in the tax rate and more flight from blight and crime by wealthier households, especially whites. This problem, together with exclusionary land use controls in the suburbs and residential discrimination against minorities (see Yinger,1995), has given rise to the mismatch problem (Kain,1968, 1994) whereby inner city blacks and other minorities have poor access to suburban jobs.

6) The fragmented nature of metropolitan governance causes suburban municipalities to compete with each other in the provision of infrastructure and the differentiation of their public services in order to attract residents and businesses as revenue sources. Such competition - known in economics as strategic competition - may create duplication of some services and too much underutilized local infrastructure capacity, which can cause more suburbanization.

The last factor in the above list, the fragmented nature of metropolitan governance, is what we wish to address next, since that is the new regionalism's favored approach to managing the consequences of sprawl and decentralization.


Is it desirable to reduce the degree of fragmentation in metropolitan governance by implementing compacts among municipalities, and how should the efficiency of such governance mechanisms be judged?

Mills and Hamilton (1994) have observed that most large U.S. metropolitan areas are composed of a multitude of local governments, which include municipalities and special purpose bodies such as school districts, water districts and regional transportation authorities:

    " integrated economic area, such as an MSA [Metropolitan Statistical Area] may contain an extraordinarily large number of local governments. The Chicago MSA has more than 1,100 local governments, and the New York City, Philadelphia and Pittsburgh MSAs have more than 500 local governments each." (Mills and Hamilton, p.316).

Is it desirable to implement a new set of State level regulations and laws which would induce such a multitude of jurisdictions in metropolitan areas to act as a compact? I interpret a compact to imply that jurisdictions belonging to it would act with much less autonomy and more coordination in many areas, than they do currently. It also means that members of the compact would accept constraints and regulations imposed at the state or federal levels and would engage in various forms of revenue sharing whether such revenue sharing is voluntary or imposed by state or county governments.

Here are some examples of how a metropolitan compact could affect the workings of local governments in the compact:

1) Services of a particular type could be consolidated across local governments. These could be public education, public health and welfare delivery, public housing, police and fire protection, public transportation, water supply, sanitation and wastewater collection and treatment.

2) The compact could institute tax rates and zoning regulations in a coordinated manner in order to achieve some common objective deemed beneficial in the aggregate. Or, equivalently, individual jurisdictions may be allowed to set their own taxes while redistribution within compacts is achieved by various revenue sharing formulas.

3) Infrastructure planning and supply decisions could be deferred to higher level governments (e.g. county or state) or decided jointly by the jurisdictions within the compact.

The efficiency implications of such a compact are as follows. The compact should be implemented only if the economic benefits generated by it exceed the added economic costs that will occur when it is implemented.

Efficiency will depend on four factors: cost savings on the one hand and reduced benefits from the loss of competition, reduced variety in the provision of local services and loss of the localized power to include/exclude undesirable land uses from a community.

Cost savings: Cost savings are possible when there are economies of scale in the provision of public services by local governments. Economies of scale exist when the public service can be provided more cheaply on a per-capita basis when it is provided to a larger number of people than when it is provided to a smaller number of people.

Public transportation services are subject to relatively strong scale economies. It is not possible to provide public transportation -- especially fixed-rail public transit -- to a small number of people at a reasonable cost and quality. But such services are already regionalized in almost all metropolitan areas. The various local governments which share public transportation services have seen the wisdom of creating regional transportation authorities.

It is doubtful that other services such as schools, wastewater treatment, police and fire protection etc. are subject to a high degree of scale economies. That is why these services are not regionalized to the same degree as is public transportation. But to the extent that several municipalities can get together and save costs in supplying their citizens with public education or with mosquito abatement, they do that. There are school districts, sanitation districts etc. which act as special governmental entities. Nothing prevents adjoining municipalities from striking voluntary agreements to share services and save costs.

If there are any significant unrealized cost savings from the consolidation of public service provision, these are most probably in the area of administrative costs. Regionalizing a service could result in lower per-capita administrative costs. But even that is not certain. There is reason to believe that the larger the government agency, the less efficiently it will be run. If this is true, there would be added costs not cost savings from consolidation.

Competition: While the proponents of the new regionalism have emphasized cost savings from consolidation and the coordination of local government services, they have -- to my knowledge -- ignored the benefits from the increased competition among a large number of local governments in the same metropolitan area.

In 1956, the economist Charles Tiebout emphasized precisely that principle, and his original insights have since been made more precise and extended by many scholars. Tiebout noted that businesses and households choose local governments according to the quantity, quality and cost of the service-bundle which they provide. To choose a local government and to locate there means to get the public services provided by that local government and to agree to pay the cost (net of any subsidies) in the form of local taxes and fees. The public services provided by a local government contain the usual menu of education, parks, police and fire protection, garbage collection etc.

Local governments also provide a number of other benefits to the consumers who locate there. The most important of these is the character of the community insured by a set of zoning and other ordinances. Local governments also provide a particular style of financial management and a fiscal reputation. For example, some local governments are business-friendly and seek to attract businesses, while others are resident-friendly and seek to exclude businesses. Some local governments are pro-growth while others are growth-control oriented. Finally, local governments provide a profile of existing resident businesses and households.

Households and businesses choose local governments not only for the particular bundle of services which they provide, but also because of the bundle of zoning and other ordinances, their financial management and fiscal styles, their orientation toward growth or business and the profile of existing businesses and residents. In a metropolitan area with a multitude of local governments, these local governments compete with one another. For example, they compete in setting a structure of taxes and fees and in insuring a particular character by means of zoning ordinances and by pursuing growth or not.

Tiebout observed that if a particular government were run badly, failing to provide a bundle of services which appealed to its residents or because it charged too much for the bundle of services which it provided, then the residents and businesses could "vote with their feet" and move to another local government in the same metropolitan area where they could elect local politicians more responsive to their needs.

In this Tiebout framework of the "metropolitan marketplace" the presence of many other local governments means that each local government would be more responsive to the needs of its residents. The presence of many other local governments means that the disciplining force of competition insures that each local government does not get excessively out of line. Also, competition generates peer pressure: if one local government comes up with an innovation, others are motivated to emulate it.

The benefits of competition among local governments is lost when the entire metropolitan area has only one government or if it consists of many local governments tied to each other through a regional compact which reduces competition among local governments.

Consider the following example of how having a single government agency can hurt an entire region. Suppose that XYZ Corporation must set up a new plant and it is considering leaving the community where its current plant is located because of high state taxes. The officials of the local Industrial Development Authority (IDA) fail to provide it with the necessary incentives to offset the high state taxes. XYZ Corp. negotiates with another IDA in the same metropolitan area which provides the needed incentives and it sets up its new factory in the jurisdiction of the second IDA. XYZ Corp. is lost to the community where it was initially located but it is not lost to the region. Competition between the two IDAs increases the overall competitiveness of the region. XYZ's employees continue to hold on to their jobs but now commute to XYZ's new location within the same metropolitan area. For a parochially motivated region, seeking to keep as much employment within its boundaries, the presence of several IDAs competing with one another is an advantage that should not be easily given up. Others may observe that this may lead to excessive competition, causing a "race to the bottom" with the region paying too much to keep the employer. But there is little evidence that competition, in this context, will be excessive or wasteful.

What would happen if there was only one IDA in the metropolitan area? That metropolitan IDA would face competition from the IDAs of other metropolitan areas but it would not face any competition from within the metropolitan area. But that is less, not more competition for the IDA and fewer not more choices for XYZ Corp. Furthermore, a single IDA facing no local competition could easily become a lethargic bureaucracy unable to respond effectively. If that happened, the likes of XYZ Corp. would leave in droves. That is in part what has caused many of our central cities, which are large jurisdictions, to remain behind the curve. They have, for the most part, failed to effectively change their culture or to institute innovative business-friendly policies. They have had less pressure to compete, partly because they have received a higher level of state and federal subsidies and partly because they have had relative monopoly power because of their size.

What would residents do if they observed that their regional compact is instituting region-wide policies with which they disagreed or for which they did not wish to pay? They would have two choices. One would be to move out of the boundary of the regional-compact and into the rural towns where they can again exercise locally meaningful voting power. Such a move, being within commuting distances, would entail a change of residence without a change of job. A second and more costly choice would be to move to another metropolitan area or another state where regional compacts did not exist or where regional compacts operated more efficiently. Such a move is more costly as, in most cases, it would entail a change of job as well as a change of residence. To the extent that moving costs prevented residents from responding to failures in regional governance, their welfare would be reduced.

To be sure, the competition which occurs among local governments is not like the perfect competition or monopolistic competition which exists in many forms of industry. One well-known failure of local government competition is the fact that some services provided by one local government spill over to the residents and businesses of adjoining local governments. This was certainly true with the example of the region's efforts to retain XYZ Corp. If one IDA succeeds in doing that, the entire region benefits. Another example is that if a central city provides better public facilities for its residents, suburban residents can also use those facilities without having paid for their provision. If a suburb provides a new shopping mall with free and adequate parking, that benefits city residents as well. Countless other examples can be given. They all amount to one conclusion: that many residents and businesses benefit from the public services of other local governments for which they do not pay.

Spillover effects mean that competitive local governments can free-ride on one another. If the adjacent municipality built a new park that my community's residents can also use, then that makes it less necessary for my community to also build a park. But the presence of free-riders causes under-provision of public services. If my community cannot charge the residents of adjacent communities for the services provided in my community, then my community is not as willing to provide those services because it does not take into account the spillover of benefits. Conversely, if my community provides services or facilities which are costly to neighboring communities (for example, a new shopping mall which creates spill-over traffic congestion), then my community will over-provide such facilities and services because it does not take into account the costs it would impose on the neighboring communities.

A major benefit of a regional compact of local governments, if accompanied by appropriate regional planning, would be that these spill-over effects could in principle be calculated by a single regional planning agency. Then, the over-provision and under-provision effects could be reduced by compensating transfers which would be possible within a regional revenue sharing arrangement. But although that sounds attractive in principle, no one knows how to reliably calculate the spill-over effects and the kind of regional planning which is required would be impossible to do in practice. Therefore, even with the spill-over effects present, it would be wrong to conclude that competition among a multitude of local governments does not improve efficiency in the provision of public services.

Variety: Another aspect of Tiebout's observation was on the variety of services and characteristics provided through the multitude of local governments. He observed that having a multitude of local governments in a metropolitan area, with each government providing a different variety of services, consumers tastes for local services are better matched.

Consider an elderly couple with the children out of high school or a single childless unmarried individual. What these two households have in common is that they do not wish to pay (or pay too much) for public education. In a metropolitan area with a variety of educational services provided by local governments, these households will reduce their share of property taxes which support education, by choosing a community where the education portion of the property tax rate is low. In a regionally managed governance structure where the same tax rate would apply to each municipality, there would be no opportunity for a similar choice to be made. Countless other examples can be given.

The benefits of a rich variety in public service provision are surely as important as the benefits of competition. It would be a mistake to embrace alternative forms of metropolitan governance without giving due consideration to how variety in public service provision would be sacrificed. One of the most important characteristics of U.S. metropolitan areas, as compared to European metropolitan areas, is that a richer variety of local service bundles are offered in the U.S. As a result of competition among them, local governments strive to differentiate their services and because of this differentiation, consumer choice is improved.

Exclusion/Inclusion: One of the most controversial benefits of local governance is the ability of residents to exclude or include certain land uses from their community. This observation is also a part of the Tiebout perspective. There are various forms of exclusion/inclusion.

Fiscal inclusion occurs when the local government zones land use in favor of uses such as commercial or multifamily residential which may be paying higher taxes on a per acre basis. To the extent that land is so zoned, additional single family development is excluded. This has the benefit that the majority of residents (e.g homeowners in single family homes) lower their tax bill in exchange for accepting a higher proportion of an otherwise undesirable land use. They may experience more traffic congestion locally and a loss of residential character but they are compensated by the lower taxes.

Economic exclusion occurs as an offset to the free-rider problem. The free-rider problem refers to the fact that a lower income resident who lives in a higher density and cheaper housing unit pays less for the same bundle of local public services than does a higher income resident who lives in a lower density and more expensive housing unit. Consider the following, now classical, example. In a suburb all of the housing is large lot residential single family, costing about $ 100,000. Public services such as schools and police protection are paid out of property taxes on single family housing of about 2% of housing value, making the annual bill $ 2,000. Now suppose that the same community also allows the construction of some multiple family apartments (or condominiums) each worth about $ 50,000. The residents of these units are poorer but also pay 2% of property value in annual taxes or $ 1,000 per year. The result is that both groups of residents receive the same bundle of public services but one pays twice as much as the other. The lower income residents are said to be free-riders because they are underpaying for the public services.

As a result of free-riding by lower income residents, the higher income residents are motivated to exclude the lower income residents (by zoning) and to cause the same land to become developed in the single family use. Notice that doing this could result in a reduction in the tax rate from 2% as long as the value per acre is higher in the single family use than it is in the multifamily use. Also, if residents prefer living in an all-single -family environment, the exclusion would cause their housing values to appreciate and an even lower tax rate would be required.

Of course, there are other responses to the free-rider problem. Instead of excluding the use, some communities would charge the free-riding use a higher tax rate. That, in turn, serves to induce partial exclusion by means of taxation, instead of total exclusion by means of zoning. The multiplicity of local governments in a metropolitan area allows several solutions to the free-rider problem to be tried out in different communities. Residents can then sort themselves out among the communities according to their preferred solution.

Prejudicial exclusion, although morally repugnant, is also prevalent. In this case, residents with similar prejudicial tastes live in communities which exclude lower income residents or minorities. A good example of such exclusion is a low-density suburban community which fights in the courts the introduction of multiple family housing even if it can be shown that it is financially beneficial. The residents believe in preserving the racial or income profile of their school district, or they believe that the introduction of such housing will reduce their property values since others are prejudiced and will flee the community. Residents of such a community are often willing to pay higher income or property taxes to maintain a high degree of racial or income homogeneity in their own communities as long as those taxes are used to subsidize the cities and other localities where the "undesirable" groups are largely confined. In other words, they believe in paying to insure "not in my community" or "not in my school district." It cannot be denied that the multiplicity of local governments benefits those who are inclined to isolate themselves by forming a community which excludes groups they consider undesirable.

To conclude, the presence of economies of scale in the provision of public services means that regional compacts can save cost in selected areas. At the same time the loss of competition and of variety in the provision of public services together with the loss of the power to selectively include/exclude means that the loss of benefits from instituting regional compacts could be sizable. I am, therefore, unconvinced that the new forms of regional governance which are advocated are desirable. I am prepared to concede, however, that there may be specific areas of public service provision where regionalization may be desirable. I leave it to others to provide the examples.


Would regional governance reduce central city blight and would it limit the process of employment and population decentralization without creating other undesirable side effects?

The answer clearly depends on the form of regional governance which is advocated. The new regionalism espouses a variety of possible policy approaches. It is, therefore, impossible to define it precisely. To the extent that regional governance means the fostering of voluntary agreements among local governments so that regional or inter-jurisdictional problems can be tackled jointly, I cannot see anything wrong with that. Truly voluntary agreements are always mutually beneficial and anything which is mutually beneficial to the parties involved can only improve economic efficiency. Thus, for example, if inner-ring suburbs voluntarily decide to cost-share in the mitigation of the blight of those central city neighborhoods which border them and if the voters of those inner-city suburbs supported the idea, it must be assumed that it would benefit at least the majority of the residents.

But the forms of grand regional governance which are advocated go much beyond voluntary agreements among local governments. I will deal with two examples: one existing, the other proposed. The existing example is the law in the State of Oregon which prescribes that all urban areas institute growth boundaries. The second is the currently proposed Smart Growth Economic Competitiveness Act of 1999 (Hoyt and Rath, 1999) which may be enacted in New York State and is an example of similar concepts and existing legislation implemented in other states. The Act calls for the direction of "...future state infrastructure investments toward existing population centers and/or existing public infrastructure" by means of a regional compact of local governments.

The language in the proposed legislature states explicitly that redirecting funds in the manner proposed, can address the problems of "...rising property taxes; decentralization of employment centers; air and water pollution; congestion of public ways; isolation of older cities and suburbs, and the consequent abandonment of substantial public infrastructure investments; diminishment of agricultural lands; and the reduction of public services."

How does New York State's SGECA, differ from what is being practiced in Portland, Oregon? Both policies designate the central parts of metropolitan areas as locations to which future development should be attracted, while indicating that peripheral areas, which were the growth areas of the recent past, should not attract additional growth. Beyond that, the two approaches differ as to the means employed: Portland attempts to direct development to the center by means of direct land use controls at the metropolitan level, while SGECA proposes to do it by targeting subsidies to the central area and denying them or reducing them to the periphery.

In the case of Portland, there is a precise growth boundary beyond which certain types of development are not allowed to occur and other types of development are allowed if they provide their own infrastructure. For example, families that do not wish to live within the growth boundary can set up single family homes in the rural areas outside the boundary but they must provide their own septic tanks and some other services. If members of such families continue to hold jobs within the boundary, they are forced to endure longer commutes for the luxury of choosing their preferred residential density outside the boundary.

In the case of New York State, as I understand the proposed legislation, there would be no strict growth boundary. However, many peripheral areas would not be eligible to receive state subsidies for the development of certain kinds of public infrastructure. Presumably, they would be able to build the same or substitute infrastructure with private capital. Meanwhile, the close-in areas targeted for future development would presumably operate as a regional compact, sharing infrastructure subsidies according to a formula which would direct funds toward existing population centers or toward areas of existing infrastructure. What this means, for example, is that the City of Buffalo would find it easy to get money from the state to upgrade streets or to replace old wooden water pipes, while the rural suburb of Newstead would find it impossible (or, perhaps, very difficult) to get money to build more local streets or to install sewer lines. This would raise the private cost of development in Newstead, lowering it in Buffalo and would thus induce development destined for Newstead to go to Buffalo instead.

What happened in Portland is by now pretty well known (see, for example, Knaap 1985; Knaap and Nelson, 1988, 1992). The strict growth boundary limited the area in which development could occur. Because Portland was a rapidly growing area, the price of land within the boundary increased sharply, by 400% by some accounts (National Association of Home Builders, 1997). A 400% increase in the price of land, if true, translates into an 40%-80% increase in the price of a housing unit on average assuming that land costs were on average 10%-20% of housing costs before the growth boundary was instituted.

What did Portland area residents get for such a sharp increase in their housing costs? Those who owned land inside the boundary clearly made exceptional windfall gains, but these gains came at the expense of those who owned land outside the growth boundary: their land lost value because future development prospects were either wiped out or greatly diminished. Thus, the growth boundary acted as a device which distributed wealth away from peripheral land owners (including farmers) and in favor of centrally located land owners (including homeowners and city landlords). The property rights of the peripheral owners of land were taken without compensation.

What about tenants? If the above estimates are correct, businesses and households who rented buildings within the growth boundary are on average paying 40%-80% more than they would be paying. This is exceptionally harmful to those who have low incomes. Many such households were forced to move out of the boundary and into surrounding rural towns, where the cost of housing remained quite reasonable relative to the areas inside the boundary. Others moved out because they could not afford a large enough home at the high land prices inside the boundary. Such households located in the peripheral towns and rural areas where they could install their own septic tanks. To the extent that they continued to work inside the boundary, they incurred longer commutes than they would have in the absence of the growth boundary. A hidden cost, difficult to quantify but real, was also borne by those households and businesses which would have located in Portland but decided not to do so because of the high land prices there. By the same reasoning, some businesses must have moved out of Portland or out of state, while others are considering doing the same: Intel, for example (see National Association of Home Builders, 1997). But cost increases were not confined to just land. Businesses and households buy a variety of things (restaurant meals, theater visits and all kinds of other shopping goods) which are produced or sold locally. To the extent that land is an input in the production or sale of these goods and services, their prices must have increased as well. Wages must have increased in Portland because of the higher cost of living there induced by the growth boundary. But labor is a mobile factor, whereas land is immobile, and the increase in wages could come nowhere near offsetting the sharp increase in land prices caused by the boundary.

I have included a supply-demand diagram for the land market in the Portland metropolitan area. Figure 1 illustrates how the limitation in the use of land defines the effective land supply function SS which is completely inelastic in its binding portion. The demand curve DD represents the aggregate quantity of land all users would demand at any given price for the land. The supply function, SS, is that which would hold were there no growth boundary. Then, the equilibrium price of land would have been P1, whereas now the equilibrium price is P2, reportedly five times higher. If there were no growth boundary, the total benefits in dollars to the users of land would be measured by the triangular area adf and the benefits in dollars to the owners of land by the triangular area dfh. The growth boundary regulation causes the benefits to the users of land to shrink from adf to just abc, while the benefits to the owners of land shrink by egf and increase by bced. The loss efg is incurred by those who cannot develop their land while the gain bced is a transfer from the users of land to those owners of land who are within the boundary. Note that total benefits (to users and owners alike) have decreased by the area cgf. This measures the deadweight loss of the growth boundary regulation in Portland exclusive of the costs of thinking up, implementing and monitoring the growth boundary.

New York State's SGECA legislation would operate in a different manner but with some similar effects. The infrastructure subsidies to the central areas would reduce the cost of development in those areas relative to the outlying areas, to the extent that these improvement costs would be otherwise borne by developers. To the extent that the improvement costs would be otherwise borne by local governments, the subsidies would stimulate the demand for land by improving the infrastructure and by reducing the local tax rates. Developers would now demand a higher aggregate quantity of land, because it is cheaper to build on it on an after-subsidy basis and because the land has been made more desirable by the infrastructure investment or the lower tax rates. This is illustrated in Figure 2 by the outward shift in the demand curve for land from position DD, prior to the infrastructure subsidies, to position D'D' after the infrastructure subsidies. That causes land prices to increase and the developed quantity of land to expand. To the extent that subsidies would not be forthcoming, the opposite would occur outside the targeted areas, with the demand curve for land in those areas shifting inward. The effects of this would be similar along general lines to what happened in Portland, although the same effects would likely be much milder because there would not be a gross limitation on land development. The diagram shows that, without the subsidy, the benefits from land development are equal to the area abc (split up as abd to users of land and dbc to owners of land). After the subsidy, the benefits are now given by efc and are higher by efba. Correspondingly, the benefits would be lowered outside the targeted area. But this is not the whole story. The cost of the subsidy itself needs to be deducted from the increase in the benefits to land use and ownership in order to arrive at a final measure of total net benefit. If the infrastructure subsidies cost the taxpayer more than efba less the amount by which the benefits are lowered outside the targeted area, then the program generates a net loss. Also, to the extent that existing subsidies to peripheral areas are cut back, there will be economic losses in those areas as well.

Figure 1: Effects of Portland's Growth Boundary on the Land Market
Figure 1

Figure 2: Effects of SGECA on the Land Market
Figure 2

What is unclear about the proposed New York legislation is the likely effectiveness of the centrally targeted infrastructure subsidies. The poor quality of infrastructure is only one of the reasons that residents and businesses and, hence, developers have flown from the central cities. The other more important reasons are that neighborhoods are blighted, that crime rates are too high, that streets are too congested, that building densities are too high and that city governments are not business friendly. To correct all these ills sufficiently to attract development back into the central cities might be so expensive that state income taxes would have to go up and the state would regress in its economic competitiveness even as it tries to improve that of its central cities. Finally, it is not clear that the withholding of public infrastructure subsidies from peripheral areas would discourage development in those places. Americans have shown a strong desire to move away from the cores of urban areas. This desire seems to be getting stronger as new forms of communication such as telecommuting, teleshopping and other forms of Internet communication become easier to employ, reducing the need for physical accessibility. Much of the infrastructure in peripheral areas is privately provided and does not rely on subsidies. Developers will provide the infrastructure as long as they know that they can pass on the cost to willing buyers.

Finally, any program which relies on subsidies may be regarded with at least some skepticism. Such a program would have to be designed very carefully. Surely it would be wasteful if all it did was to cause central cities to chase after newly available subsidies. Lessons from the past are not very encouraging: when the federal government was subsidizing the construction of fixed rail transit systems, planners in Houston did studies which greatly overestimated land use densities in Houston. Houston got a transit system but it was not an efficient investment. Subsidies could cause central cities and other central areas to become complacent and less competitive. If so, the problem of decline could be exacerbated not mitigated by the new regionalist programs of revenue sharing.


1. Paper prepared for presentation at "Regionalism: Promise and Problems," a symposium held at the State University of New York at Buffalo School of Law, March 6, 1999 and sponsored by the State University of New York at Buffalo School of Law.

2. The phenomenon of decentralization is widely documented. See, for example, Cervero and Wu (1997), Giuliano and Small (1991), Gordon and Richardson (1997), McDonald and Prather (1994), McMillen and McDonald (1998), Mieszkowski and Mills (1993).


American Planning Association (1997) Growing Smart Legislative Guidebook: Model Statutes for Planning and Management of Change, Phase I interim edition, Chicago, IL: APA.

Cervero, R. and K-L, Wu (1997) "Polycentrism, commuting and residential location in the San Fransisco Bay Area, 1980-1990," Environment and Planning A, 29(5), 865-886.

Downs, A. (1994) New Visions for Metropolitan America, Washington, D.C. : The Brookings Institution and Cambridge, Mass. : The Lincoln Institute of Land Policy.

Dunphy, R. (1997) Moving Beyond Gridlock : Traffic and Development. Washington, D.C. : The Urban Land Institute.

Ewing, R. (1997) "Is Los Angeles-style sprawl desirable?," Journal of the American Planning Association, 107-126.

Garreau, J. (1991) Edge City : Life on the New Frontier, New York : Doubleday.

Giuliano, G. and K.A. Small (1991) "Subcenters in the Los Angeles Region," Regional Science and Urban Economics, 21(2), 163-182.

Gordon, P. and H.W. Richardson (1996) "Beyond polycentricity : The dispersed metropolis, Los Angeles, 1970-1990," Journal of the American Planning Association, 63(1), 95-106.

Gordon, P. and H.W. Richardson (1997) "Are compact cities a desirable planning goal ?", Journal of the American Planning Association, 63(1), 95-106.

Hoyt, S. and M.L. Rath (1999) New York State Assembly Memorandum in Support of Legislation.

Judd, D. and T. Swanstrom (1994) City Politics: Private Power & Public Policy. New York, N.Y. : Harper Collins.

Kain, J. (1968) "Housing segregation, negro employment and metropolitan decentralization," Quarterly Journal of Economics, 82, 175-197.

Kain, J. (1994) "The spatial mismatch hypothesis : three decades later," Housing Policy Debate, 3, 371-462.

Knaap, G.J. (1985) "The price effects of urban growth boundaries in metropolitan Portland, Oregon", Land Economics 61(1), 26-35.

Knaap, G.J. and A. C. Nelson (1988) "The effects of regional land use control in Oregon : A theoretical and empirical review", The Review of Regional Studies 18(2) :37-46.

Knaap, G.J. and A. C. Nelson (1992) The Regulated Landscape: Lessons on State Land Use Planning from Oregon, Cambridge, MA: Lincoln Institute of Land Policy.

Kunstler, J. Home from Nowhere: Remaking of Everyday World for the 21st Century. New York, NY: Simon & Schuster.

Mayo, S.K. (1998) "Land prices, land markets and the broader economy," in Land Lines : Newsletter of the Lincoln Institute of Land Policy, March, 4-5.

McDonald, J. and P.J. Prather (1994) "Suburban employment centers : The case of Chicago," Urban Studies, 31(2), 201-218.

McMillen, D.P. and J.F.McDonald (1998) "Suburban subcenters and employment density in metropolitan Chicago," Journal of Urban Economics, 43(2), 157-180.

Mieszkowski, P. and E.S. Mills (1993) "The causes of metropolitan suburbanization," Journal of Economic Perspectives, 7(3), 135-147.

Mills, E.S. and B. Hamilton (1994) Urban Economics. 5th edition, Harper-Collins.

Mills, E.S. and L.S. Lubuele (1997) "Inner cities," Journal of Economic Literature, 35, 727-756.

Moe, R. and C. Wilkie (1997) Changing Places : Rebuilding Community in the Age of Sprawl. New York, NY : Henry Holt and Co. Inc.

National Association of Home Builders (1997) "Urban growth plans headed for a tumble", Nation's Building News, November 24.

Orfield, M. (1997) Metropolitics : A Regional Agenda for Community and Stability, Washington, D.C. : The Brookings Institution and Cambridge, Mass. : The Lincoln Institute of Land Policy.

Preston, J. (1998) "Battling sprawl, states buy land for open space," The New York Times, June 9.

Rusk, D. (1993) Cities without Suburbs, Baltimore : Johns Hopkins University Press.

Tiebout, C. (1956) "A Pure Theory of Local Public Expenditure," Journal of Political Economy 64, 416-424.

Yinger, J. (1995) Closed Doors, Opportunities Lost : The Continuing Costs of Housing Discrimination. New York : Russel Sage Foundation.



USC Seal

Main Page | Subscribe | Submission Requirements | Editorial Board | Archive | Links

ISSN 1548-6036

Copyright 1999-2000
University of Southern California
Los Angeles, California 90089-0626