PART THREE—
URBAN PROGRAMS: TRANSPORTATION AND HOUSING

Two of the federal domestic programs that have the greatest effect on urban areas are transportation and housing. Jose Gomez-Ibañez argues for a reduction of the federal role in transportation. Sherman Maisel sets an agenda for housing policy that includes a substantial federal role in providing aid to new home purchasers and to low-income renters. John Kain suggests that federal support for fair housing has helped to mitigate the effects of racial discrimination in housing and that a recent trend toward black suburbanization may provide a glimmer of hope for the future. In the Commentary, Michael Goldberg uses the Canadian experience to support his contention that the federal government's role in housing and urban affairs should be smaller than Maisel advocates. Theodore Keeler questions Gomez-Ibañez's conclusions on the grounds that the benefits of urban highways far outweigh the costs and that the costs of fixed rail systems are substantial. Richard Muth disagrees with both Kain and Maisel, maintaining, first, that fair housing policies are not responsible for reported declines in housing segregation and, second, that there is little need for a federal role in housing policy. Melvin Webber faults Gomez-Ibañez for concentrating on the auto-transit dualism, suggesting that the appropriate dichotomy in most modern metropolitan areas is that between large-vehicle and small-vehicle systems.

Chapter Seven—
The Federal Role in Urban Transportation

Jose A. Gomez-Ibañez

The federal role in urban transportation increased in size and scope after World War II. In the late 1950s and early 1960s, federal involvement was largely limited to grants for urban segments of the Interstate Highway System, but by the late 1960s and early 1970s it included aid for other types of urban highways and for mass transit as well. At first, federal assistance was confined to aid to capital expenditures, but it gradually broadened to include some operating and maintenance expenses. By 1980, federal aid accounted for as much as 40 percent of all government capital spending on urban highways, 80 percent of capital assistance for urban mass transit, and 30 percent of mass-transit operating subsidies.

The Reagan administration proposed to reduce federal aid for urban transportation significantly, by restricting highway aid largely to the Interstate Highway System, by phasing out operating aid for mass transit, by reducing transit capital aid, and by allocating transit capital grants by formula rather than project by project. Congress rejected most of these proposals in 1982, when it restored or substantially increased funding for many of the existing programs.

While some of the proposals made by the Reagan administration may be undesirable, the idea of forcing local governments to assume more responsibility for decisions about the level and mix of urban transportation spending has merit. The rationale for extensive federal involvement in urban transportation is weak, and several of the federal programs, particularly grants for Interstate Highway construction and mass transit capital projects, probably distort local transportation decisions. Reduced

federal aid for highways and mass transit might impose temporary financial hardships, but in the long run most urban areas would probably be better off.

Development of Federal Programs

Urban Highway Aid

Early Federal Aid

Federal highway aid began in 1913 but was largely confined to rural roads until after World War II.^[1] The first highway program apportioned federal funds among the states to cover up to one-half the construction costs of rural post roads, roads over which the mail was carried. In 1921 the federal program was restructured and federal aid was restricted to two types of highways: primary or interstate highways, and secondary or intercounty highways. Each state, in consultation with the federal government, could designate a small portion of their highway mileage as part of these federal aid systems. Highway aid was apportioned among states on the basis of a formula that considered population, land use, and rural-post-route mileage.

Until the Depression, urban highways were specifically excluded from federal aid mileage. Public works relief legislation in 1932 and 1933 provided some temporary funding for urban highway construction since many of the unemployed were city dwellers, and in 1934 the prohibition on including urban mileage in primary or secondary systems was eliminated. Not until the Federal Highway Aid Act of 1944 was specific authorization made for urban highways. This act established three distinct categories of federal-aid highways: primary and secondary rural highways (which were to receive 45 and 30 percent of all aid, respectively) and urban extensions of the primary and secondary routes (which were to receive only 25 percent of the aid).

The Postwar Construction Boom

Pressures to increase highway spending mounted after World War II as postwar prosperity fostered automobile ownership and use. During the 1940s and 1950s urban highway traffic grew at almost double the rate of the urban population, as higher incomes encouraged travelers both to increase the number of trips they made and to make more of them by automobile rather than mass transit (Table 7.1).

Federal aid for urban highways became available on a large scale in 1956, when the federal government responded to the rapid traffic growth by funding the Interstate and Defense Highway System. Under the Interstate program, federal grants paid 90 percent of the construction costs of a congressionally designated system of 42,500 miles of limited access highways. The system was initially conceived as serving intercity travel and bypassing major cities but, under pressures from urban interests, about 20 percent of the final system was located in urban areas. Urban mileage accounted for an even larger share of Interstate program expenditures, since the urban segments were often wider and more costly than their rural counterparts. Unlike the primary and secondary road programs, Interstate grants were to be distributed to the states on a project-by-project basis rather than by formula. In 1970 Congress also established a new federal-aid urban system, instructing the states to designate for the new system mileage that supplements (but does not overlap) the older urban extensions; aid for the urban system is allocated to states on the basis of urban population. Thanks largely to the Interstate program, the federal share of highway capital spending by all levels of government grew from only 10 percent in 1950 to around 40 percent in the 1960s. While in 1950 federal grants were almost exclusively for rural roads, by the 1960s and 1970s many federal highway dollars were spent in urban areas.

Disenchantment with Highways

During the 1960s many social and environmental costs of highway construction and automobile use became apparent. The first focus of concern was the social and aesthetic costs of expressway construction, particularly in urban areas. In response, Congress in 1962 required that metropolitan areas receiving federal highway aid prepare a comprehensive land use and transportation plan, which included consideration of mass transit alternatives. These regulations were substantially strengthened in 1969 by passage of the National Environmental Policy Act, which required the preparation, with public comments and hearings, of an environmental impact statement for any federally funded project. In 1973 Congress also allowed local governments to trade in the federal funds designated for locally contested segments of the Interstate System for use on mass transit improvements, as long as the United States Secretary of Transportation certified that the contested segment was not essential to the national Interstate System. By 1980, over 8000 miles of urban

Interstates had been opened to traffic, although in almost every metropolitan area controversial segments remained unfinished.

The 1960s and 1970s also saw the development of federal programs to regulate the safety, air pollution emissions, and fuel economy of new motor vehicles. In 1965 Congress established the National Highway Traffic Safety Administration, with authority to set safety standards for new cars and to order manufacturers to recall and correct defective vehicles. Air pollution was the next focus of concern, as the problems caused by automotive emissions of hydrocarbons, carbon monoxide, and nitrogen oxides became more noticeable in major urban areas. The Clean Air Act of 1970 required 90 percent rollback in the emission rates of these pollutants from new automobiles by the 1975 model year. High costs and technological uncertainties forced several extensions of the deadlines and a relaxation of the nitrogen oxide target to a 75 percent rollback, but by the 1982 model year the goals set by the Clean Air Act were achieved. Finally, the oil crisis of 1973–1974 stimulated Congress to pass a law requiring that average new-car fuel economy be improved in stages from 18 mpg in model year 1978 to 27.5 mpg in 1985. United States automakers were in compliance with the schedule as of model year 1984, although weakening gasoline prices were making it more difficult to induce customers to buy fuel-efficient models.^[2]

The level of highway construction and investment peaked during the late 1960s. Although nominal highway spending continued to increase during the 1960s and 1970s, capital outlays failed to keep pace with the rapid inflation in building costs (see Table 7.2). The decline in real spending was due in part to the heavy reliance on federal and state fuel taxes to finance highway improvements. Since fuel taxes are assessed in cents per gallon, real receipts fell with rapid inflation and the automobile fueleconomy improvements of the 1970s.

The scope of federal highway aid expanded in the 1970s, despite the spending decline. Highway maintenance became a major concern toward the end of the 1970s, as many of the highway pavements and structures built in the decades immediately following World War II began to reach the end of their useful lives. Until the 1970s, federal highway aid could only be used for new construction or for major reconstruction projects in which the road was widened or substantially rebuilt. In 1976 Congress responded to the maintenance crisis by expanding the definition of highway construction to include "resurfacing, restoration, and rehabilitation" (3R) and establishing a separate program of grants for Interstate 3R projects. In 1970 Congress also established a special aid program for

bridge repair and reconstruction on the federal aid highway systems, activities that had been exclusively state-financed.

Urban highway traffic has continued to grow despite the slowdown in highway construction, higher fuel prices, and increased regulation of new car pollution and fuel economy. Traffic growth was slightly lower in the 1970s than in earlier decades, probably largely because the growth in urban population also slowed during that decade. As in the previous decade, urban traffic increased about four times faster than the urban population during the 1970s (see Table 7.1).

Urban Mass Transit Aid

Almost as rapid as the growth in urban automobile use was the decline of urban mass transit ridership in the 1940s and 1950s, falling from a wartime high of 18 billion trips per year to 7 billion trips in 1960. In an attempt to stem the ridership loss, in the 1940s and 1950s many large metropolitan areas began to provide government aid to their mass transit systems, and several metropolitan areas planned new rail systems. The rapidly deteriorating financial condition of mass transit strained local resources, however, and by the early 1960s had led to strong pressure for federal assistance.

Federal aid began in 1964 with a small program of capital grants that would finance up to two-thirds of the cost of a capital project.^[3] Funds were allocated for specific projects at the discretion of the Secretary of Transportation. Funding for the capital grant program expanded rapidly during the 1970s, reaching levels of almost $3 billion per year, and in 1973 the maximum federal share was raised to 80 percent of costs. Federal grants were used to finance local government buyouts of ailing private bus companies, replacement or refurbishing of much of the existing transit rolling stock and equipment, construction of several new rail systems, and extensions to many of the existing rail systems.

In 1974 Congress established a new federal transit grant program to fund operating expenses, partly in response to pressure from metropolitan areas that felt they did not benefit greatly from the capital grant program, either because their rail systems predated federal aid or because they were too small to consider rail seriously. Funds from the new grant program could be used for capital as well as operating expenses, although in practice almost all was used for operating aid. The federal grants had to be matched by an equal amount of state or local aid if operating expenses were being subsidized, but the federal share could be as high as

80 percent for capital projects. In a significant departure from the original capital-grant program, the operating grants were allocated among urbanized areas according to a formula based on population and population density instead of at the discretion of the Secretary. Federal operating aid grew steadily from around $200 million in 1970 to nearly $1 billion in 1980.

Transit aid from all levels of government increased so rapidly during the 1970s that sometime in the middle of that decade government aid became a more important source of revenue to the industry than passenger fares (Table 7.3). Passenger fares had been enough to cover operating costs and make a small contribution to capital expenses through the 1950s, despite the fact that the industry was contracting. In 1964 passenger receipts fell below operating expenses for the industry as a whole and by the 1980s covered only about 40 percent of operating costs and made no contribution to capital expenses. Government aid grew from $518 million in 1970 to $7.8 billion in 1982, and the federal share of all aid grew from 26 to 53 percent, as shown in Table 7.4.

The influx of government aid was not enough to restore mass transit ridership to its former levels. The steady postwar decline in nationwide mass transit ridership was reversed between 1974 and 1980, when patronage grew by 15 percent, but in 1981 it began to fall once again. The most important urban travel market is commuting, since work trips dominate rush hour travel and are key contributors to traffic congestion and other urban transportation problems. Public transportation's share of commuting trips in all metropolitan areas fell steadily from 19 percent in 1960 to 12 percent in 1970 and 8 percent in 1980 (Table 7.1). As of 1980, mass transit captured more than 10 percent of the commuter market only in a dozen of the largest, densest, and oldest metropolitan areas, where high levels of traffic congestion and parking fees discourage auto ownership and use. Only in one metropolitan area (New York) does transit's share of the commuting market exceed 20 percent.^[4]

The performance of the first new rail systems—opened in San Francisco in 1972 and Washington in 1976—was particularly disappointing; construction costs soared far beyond original projections and the ridership gains were smaller than originally forecast. These results failed to discourage local interest in new rail systems, however, in part because federal aid was available to fund up to 80 percent of construction costs. To force local governments to consider smaller and more cost-effective alternatives, in 1976 the federal government required that cities applying for rail grants submit an analysis of non-rail alternatives, including low-

cost express bus systems. Federal regulations also restricted new rail starts in each metropolitan area to one major corridor at a time, rather than to an entire system of many lines. In 1978 these regulations were strengthened to limit federal liability in the case of construction cost overruns and to specify in more detail the types of low-cost, non-rail alternatives that had to be considered.

Policy under the Reagan Administration

By the time President Reagan took office in 1981, federal grants had increased to cover roughly 20 percent of all government highway aid and 50 percent of all transit aid (Table 7.4). The Reagan administration initially proposed dramatic changes in urban transportation policy as part of its "new federalism" initiative. The theme of the new federalism was to return responsibility for largely local programs, such as urban transportation, to state and local governments. The Reagan administra-

tion proposed both to reduce the level of federal highway aid and to restrict it to highways, such as the Interstate System, in which the federal interest was clearest. Federal gasoline taxes would be lowered (to the levels necessary to fund the smaller program) and state governments would be encouraged to increase their own fuel taxes as needed. For mass transit, the Reagan administration proposed to eliminate federal aid for operating expenses, reduce federal aid for capital projects, and allocate

capital grants on the basis of a formula rather than at the discretion of the Secretary of Transportation.^[5]

The administration achieved some of these goals in its first year in office. The FY1982 budget reduced federal transit aid below FY1981 levels and left highway aid unchanged despite inflationary pressures. The administration also imposed a moratorium on federal aid for starts of new rail mass transit systems or extensions, although funding would continue for projects already under construction.

These victories did not last long. Frustrated by the Reagan administration's moratorium on new rail starts, the House and Senate appropriating committees began to order the Urban Mass Transportation Administration (UMTA) to fund the planning, engineering, or construction of new rail systems in specific cities. The FY1982 appropriations conference report earmarked federal planning grants for six projects and construction grants for five others (two in Miami), while the FY1983 report earmarked funds for alternatives analysis of seventeen projects and engineering or construction of eleven. If all these new rail projects are built with federal aid, the cost to the federal treasury will be between $12 and $19 billion.^[6]

During 1982 pressure also grew within the administration and Congress to increase highway spending as a means to alleviate the deterioration of the nation's highways and bridges and to combat unemployment from the 1981–1982 recession. To gain support from urban areas for increased highway spending, Congress proposed to increase mass transit aid as well.

The Surface Transportation Assistance Act, passed in December 1982, greatly increased federal funding for highways and restored the cuts to mass transit. The law raised the federal gas tax from 4 to 9 cents per gallon, the first increase since 1959. Four cents of the increase is dedicated to the Federal Highway Trust Fund to finance a near-doubling of 1982 levels of highway aid (Table 7.5). The remaining cent of the new gas taxes is earmarked for a new Federal Mass Transportation Trust Fund, which will be used to finance the discretionary capital grant program. Additional transit funding is authorized out of general revenues for new operating and capital formula grant programs and total transit aid is restored to the 1981 funding levels (Table 7.6). The Reagan administration's only victories were in the composition of the spending authorized by the act. Only highway aid for Interstate and rural primary systems is increased, for example, while funding for rural secondary and urban extensions

was held constant. Transit operating assistance is also to be gradually lowered.

Evaluating Highway Aid

Benefits and Costs of Postwar Highway Expansion

Although some benefit-cost analyses have been made for isolated highway projects, there have been no recent and reliable attempts to measure the benefits and costs of the federal urban highway program as a whole.^[7] A comparison of highway user tax receipts with highway expenses suggests, however, that the benefits of the postwar boom in urban highway construction, partly financed by federal aid, have on the whole exceeded the costs.^[8]

User Payments as Benefit Measures

Motorists pay a variety of special taxes to use the highways, including gasoline taxes, tolls, and vehicle registration fees. Since motorists are willing to pay at least this much (and maybe more) for highway use, these user charges provide a minimum estimate of the benefits motorists receive from highway investments. In theory, a particular segment of the highway system might be deemed worthwhile if users pay enough in highway taxes to cover the costs that segment imposes on society.^[9]

Estimating the user component in highway taxes involves differentiating between special charges made for highway use and general taxes levied on consumption or personal property. The commonly used rule for making such separations is to regard a tax as a highway-user charge if collected from highway users or on motor vehicles but not on most other comparable goods or services. By this rule, highway-user charges clearly include federal gasoline, vehicle, spare parts, and excise taxes; state motor vehicle and driver registration fees; and special highway-user taxes on trucks. State gasoline and motor vehicle sales and property taxes are not necessarily included, since many states or localities levy sales or personal property taxes on a variety of other goods, although typically at lower rates. Estimates of United States highway-user-charge receipts derived by following this separation principle are shown in Table 7.7 in current dollars for the years 1956 through 1975.

Government Highway Expenditures

The costs of the urban highway system include both direct government highway expenditures, such as highway construction and maintenance outlays, and the externalities or social damages that highway users impose on others in the form, for example, of automobile air pollution and noise. Table 7.7 also shows estimates for 1956 to 1975 of direct highway expenditures paid by all levels of government, but excluding any external social costs of highway use. The estimates assume that all government highway capital expenses are amortized over the life of the facilities using a 5 percent real discount rate—thus, for example, the 1975 costs include charges for the depreciation of capital investments made in previous years. Capital outlays include all expenditures for the purchase of right-of-way (including relocation assistance), grading and drainage costs, construction of bridges and tunnels, and paving costs. Operating expenses include routine maintenance, snow removal, mowing, painting, and similar activities as well as highway administration, research, safety, and law enforcement. Outlays for all police assigned to traffic and highway safety duties, as well as correctional expenses attributable to enforcing traffic laws and curtailing automobile theft, are included as highway law enforcement costs.

Most highway expenses and user revenues can be attributed to particular parts of the highway system, such as urban or rural roads or Interstate and non-Interstate highways. Some costs and revenues, such as the administrative expenses of highway agencies or annual vehicle registration fees, are, however, not clearly assignable. These costs and revenues amount to about one-quarter of the total and are allocated, somewhat arbitrarily, to the different highway types according to the vehicle mileage occurring on those highways.

For the national highway system as a whole, highway user charges exceeded direct government costs by 50 to 80 percent in the 1950s and 1960s, but fell to rough parity with costs by 1975. If a more conservative 10 percent discount rate is used instead of 5 percent, revenues exceeded expenses by about 25 percent in the 1950s and 1960s but fell short of costs by about an equal amount in the 1970s.^[10] The ratio of revenues to expenses fell largely because total amortized highway outlays increased in real terms (although current capital investment has declined), while user revenues did not keep pace with inflation.

Most of the shortfalls between highway user revenues and direct gov-

ernment highway expenditures occur in rural rather than urban areas and on roads other than the Interstate and other federal aid systems. For the urban Interstates and urban extensions of the federal primary and secondary systems, user revenues were double or triple direct government expenses in the 1950s and 1960s and still exceeded expenses by 30 to 50 percent in 1975.

Even though collectively urban highway users may pay their capital and operating costs, the users of highways in the centers of large dense metropolitan areas probably do not. Costs of constructing and maintaining a lane-mile of highway are much higher in the downtowns of large metropolitan areas than they are in the suburbs or in small metropolitan areas.^[11] The number of lanes in an urban highway is dictated by peak period rather than off-peak travel needs, moreover, so there is substantial reason to assign most of the costs of constructing and maintaining that capacity to peak users only. While highway construction maintenance and administration costs in 1975 averaged only about 1.3 cents per vehicle mile on all urban Interstates, for example, studies suggest that the cost of serving peak-hour users on expressways in the central business districts of Boston and San Francisco were as high as 10 to 30 cents per vehicle mile in the mid 1970s.^[12]

While the capital and operating expenses attributable to peak-period and downtown motorists are much higher than average, the highway-user taxes paid by these motorists are only moderately higher. In 1975, tax receipts from all urban highway users—mainly in the form of the gasoline tax—averaged about 1.7 cents per vehicle-mile. Because of congestion, peak-period and downtown users probably consumed a bit more fuel and therefore paid perhaps 10 to 20 percent more taxes per mile than average. The differential between peak and off-peak highway-user tax payments is therefore far smaller than the differential between peak and off-peak highway capital and operating costs on major downtown highways.

In short, while urban highway users as a whole, and urban Interstate users in particular, supply more highway tax revenue than the government pays directly in highway expenditures, many peak-hour motorists in the centers of very large urban areas probably pay less than their share of highway costs. Since user taxes actually paid may be viewed as a lower-bound estimate of benefits (many highway users might well be willing to pay more rather than do without), it seems likely that the benefit/cost ratio for many urban highways exceeds unity (still ignoring social exter-

nalities for the moment). Even high-cost downtown urban expressways might be able to generate receipts in excess of costs if special tolls were charged, but this is speculative.

External Social Costs

These results probably are not substantially changed by including the external social costs of highway use, such as air and noise pollution, the aesthetic blight of urban highways, and the social costs of relocation and land takings. Although most of these external costs are extremely difficult to measure, estimates of the costs of motor-vehicle air and noise pollution have been made. A study of urban transportation in the San Francisco Bay Area estimated the property, health, vegetation, and materials damage of air pollution from the average car in the 1968 fleet at 1.29 cents per vehicle-mile (in 1973 dollars). Because of federally mandated reductions in new-car emissions rates, the same study estimated that a 1972 model automobile in San Francisco had air pollution costs of only 0.48 cents per vehicle mile. A 1982 model automobile that achieved the nearly 90 percent reduction in emissions over 1970 levels would generate air pollution costs of only 0.1 to 0.2 cents per vehicle mile.^[13] Similar estimates of the effects of traffic noise on neighboring businesses and residences suggest that noise costs probably averaged less than one-tenth of a cent per automobile mile in the mid 1970s.^[14]

Estimates are unavailable for several other social costs of highways, the most notable being uncompensated costs to households and businesses relocated by highway construction, and the visual or aesthetic blight of highways. These costs are probably substantial for some poorly designed and badly sited urban highways constructed in the 1950s and 1960s. Federal regulations mandating relocation assistance, environmental impact statements, and increased citizen involvement in highway planning have probably reduced these uncompensated damages substantially on highways built after the 1960s.^[15]

The external social costs of urban automobile use in the mid 1970s probably averaged about 0.5 cents per vehicle mile based on air and noise pollution alone, and perhaps as much as 1 cent per vehicle mile if an allowance for other unestimated social costs is included. Combined with the 1.3 cent-per-vehicle-mile average cost of constructing and maintaining highways, these figures imply that urban highway users probably on average only slightly underpaid their costs during the mid 1970s. The

rough parity of highway user receipts and costs suggests that many of the urban highways constructed with federal aid during the postwar period were worthwhile, especially when it is remembered that user fees are a minimum estimate of user benefits. The principal exceptions are likely to be extremely high-cost highways designed to accommodate rush hour traffic in the centers of major metropolitan areas.

Future Highway Needs

Several forecasts of future urban highway spending needs have been prepared, the most sophisticated of which are the Federal Highway Administration (FHWA) biennial reports to Congress on highway needs. The FHWA estimates that highway capital spending on all roads (urban and rural) should be increased by about 50 percent over the levels of the late 1970s and early 1980s in order to restore the roads to what it terms minimum tolerable conditions; a 40 percent increase is needed merely to prevent further deterioration.^[16] Such increases would restore real highway capital spending to the peak levels of the mid and late 1960s (see Table 7.2). The composition of capital spending would shift to emphasize the repair and reconstruction of existing pavement and structures instead of major widening or alignment improvement projects.

The FHWA's highway needs forecasts are limited, however, in that they are based on common highway engineering practices or rules of thumb rather than detailed benefit-cost analysis. The FHWA presumes, for example, that additional lanes are needed on an urban Interstate highway segment when the peak-hour traffic volumes exceed 85 percent of the roadway's maximum hourly capacity. Similarly, the FHWA assumes that repaving is needed on an urban Interstate when the Pavement Serviceability Index (PSI), a physical index of pavement roughness, falls below 3.2. There is some evidence that the volume-to-capacity standard used by the FHWA may, on average, recommend more highway widening than benefit-cost analysis would call for, while the FHWA's pavement standard—again, on average—may recommend too little repaving. For urban highway segments with widening and paving costs much higher or lower than average, moreover, the simple uniform standards used in the FHWA forecasts are almost certainly misleading.^[17]

Despite the shortcomings of its forecasts, the FHWA was almost certainly correct in recommending a substantial increase in highway investment over late 1970s levels. Given the slowdown in real highway capital

spending in the 1970s, the net value of highway investments has undoubtedly been declining, despite continuing growth in urban (and rural) traffic volumes. If the postwar highway investments were, on the whole, worthwhile, then comparable—or perhaps greater—levels of investment are probably justified for today's higher traffic volumes.

The Merits of Federal Involvement

The Rationale for Federal Aid

Whatever the appropriate level of urban highway investment, one key issue is why the federal government should be so heavily involved. Since 70 percent of the United States population lives in urban areas, the majority of the country clearly has a strong interest in urban highways. At least in theory, however, our federal system reserves powers and responsibilities to state and local governments unless some compelling and distinct national interest is involved. This devolution of resonsibilities is based both on democratic ideals and the pragmatic argument that those who are closest to a problem often know best how to solve it.

The principal rationale for federal highway aid programs has been the national interest in an intercity transportation system that serves long-distance or interstate as well as local traffic. When federal highway aid began in 1916, the road system was largely unpaved and road construction and maintenance were the responsibility of county governments. The counties were notorious for their failure to cooperate in improving roads that served more than one county, perhaps because their dependence on property tax revenues made it difficult to finance improvements that served more than local needs. An interconnected road system would benefit all, it was argued, by promoting interstate commerce and reducing the social and political isolation of rural communities. The federal government gave highway aid directly to state governments, on the theory that states would have more interest than counties in promoting an intercity highway system.^[18]

While federal intervention may have been needed to promote an interconnected highway system seventy years ago, it may be unnecessary today. Thanks in part to early federal aid, each state now finances and administers its own system of trunk highways, leaving county and city governments responsible mainly for local or secondary roads. Federal aid may not be necessary even to induce states to build a coordinated interstate highway system. In the decade before the Interstate System was funded,

for example, many Eastern and Central states cooperated in the construction of an interconnected system of limited-access toll expressways that allowed motorists to travel between New York and Chicago or Boston and Albany without ever having to stop for an intersection or traffic light. Toll financing had eliminated the problem of using local taxes to support interstate travel and by 1956, when Interstate funding ended the boom, around 12,000 miles of toll expressways had been built, started, authorized, or projected.^[19]

To the extent that there is a distinct national interest in the highway system, it applies more clearly to roads that primarily serve long-distance and interstate rather than local travelers. Although Interstate System planners rationalized the inclusion of urban segments on the grounds that interstate traffic often originates or terminates in urban areas, urban expressways probably have a limited claim to federal aid, since their design is largely dictated by peak-hour local commuting traffic.

Perhaps the strongest argument for a federal role is in the areas of highway research and demonstration projects. Research on pavement durability, highway planning techniques, and highway safety measures is of potential benefit to all states. Since no single state captures all the benefits, there is little incentive for a state to fund research alone. The federal government, however, can consider the benefits to all states in designing its research program.

Incentives under Federal Aid

If state and local interests predominate in urban highways, federal highway aid may distort state and local highway decisions in undersirable ways. It is extremely difficult to predict what the urban highway system might have looked like without federal aid, much less whether the system would have been better or worse. The design of the federal aid programs offers some simple clues about the degree to which state and local choices might be altered. The broader the category of highway projects eligible for aid and the lower the funding levels, the more likely federal aid will merely serve to reduce state or local taxes and leave highway spending decisions largely unchanged. Conversely, the more restrictive the eligibility and the greater the funding, the more likely it is that federal aid, if accepted, will alter local choices.

According to this test, federal aid for the primary, secondary, and urban systems probably had little effect on postwar state and local highway decisions. Although states can designate only a fraction of their highway

mileage as eligible and federal funds must be matched by an equal amount of state or local aid, the level of federal aid probably falls far short of current expenditures on these systems. In 1980, for example, federal primary, secondary, and urban systems aid accounted for only $2 billion of the $9 billion spent for capital outlays on state-administered highways (excluding the Interstates).^[20] The ratio of spending to federal aid was probably even lower in urban areas, since the federally aided mileage is still disproportionately rural.

The Interstate program offers potentially strong incentives to distort state and local choices, however, and efforts to reduce or offset these incentives have been only partially successful. Only segments of the 42,500-mile congressionally designated Interstate System are eligible for aid, and each segment must meet strict federal design standards governing access control, lane width, curvature, ramp length, and other features. In return, the federal government will pay up to 90 percent of the cost of constructing the segment, no matter how expensive the project. (Only allowable, highway-related costs can be reimbursed, of course, and competitive bidding must be used.)

These strong financial incentives led early critics to fear that state and local highway agencies would overlook the environmental costs and social disruption of highway construction and build many unnecessary Interstate segments. Although blighting and disruptive highways were probably built, as noted earlier, the majority of urban Interstate mileage appears to have been worthwhile. The rapid rates of expressway construction and planning in the years immediately preceding the start of the Interstate program also suggest that some, though certainly not all, of the limited-access highways might have been constructed anyway. Finally, federal regulations requiring comprehensive transportation planning, environmental impact statements, and public hearings have helped to reduce the problem of blighting and disruptive highways by increasing the opportunities for citizens to block or delay highway projects.

Recently a new problem has emerged, due to the combination of the strong financial incentives of the Interstate program and the federal regulations mandating increased citizen involvement. To overcome local opposition to the construction of environmentally or socially sensitive urban Interstate segments, state highway officials are now encouraged to incorporate costly design features that are largely paid for by the federal government. Two extreme examples of this new problem are the West Side Highway in New York City and the Central Artery in Boston. Both of these downtown elevated expressways are now part of the designated Interstate system but were built before Interstate aid was available and

consequently do not meet Interstate design standards. Part of the West Side Highway collapsed in 1973, and the Central Artery deck is nearing the end of its expected structural life. Local groups oppose reconstructing the highways in their current form, since they are ugly and blighting; in any case, a reconstruction project would be ineligible for Interstate funding unless federal officials granted a design variance. New York's solution (which was approved by federal transportation officials but has been delayed by an environmental lawsuit) is to build four miles of new Interstate offshore in the Hudson River, with 230 acres of landfill for housing and parks to serve as a buffer between the highway and existing neighborhoods. Boston's proposal (which is pending federal approval) is to depress one mile of the highway along the original alignment, with decks to support overhead development. These offshore and depressed highways would cost around $1–2 billion each, four times more than reconstruction. The added costs appear to be far larger than the estimated environmental or aesthetic benefits to be gained.^[21]

If these two highways can be rehabilitated only at such high costs, they may not be worth maintaining at all. But the availability of 90 percent federal funding—coupled with complaints from local politicians and congressional delegations if federal officials attempt to review local designs more closely—have made local officials unwilling to consider less grandiose schemes.^[22] The Interstate program, in short, may have shifted from building some urban highways that were excessively disruptive and blighting to building some that have been overly beautified.

Financial Effects on Urban Governments

Federal highway aid may not only distort local choices, it also may fail to alleviate long-term financial pressures on local governments to any significant degree. The availability of federal highway aid reduces the need to raise state and local taxes, to be sure, but the aid is financed by federal highway user taxes paid in part by urban residents. Since the federal highway aid program is heavily oriented toward rural roads, federal highway-user tax payments probably significantly exceed federal highway grants in most urban areas. Although detailed urban/rural data are not available, data on federal highway receipts and expenditures by states for the period from 1956 through 1982 show that the ratio of federal highway grants to federal user tax payments was below the national average in virtually every highly urbanized state and that almost all the states with above-average grant-to-tax payment ratios were rural.^[23]

These simple calculations also ignore the fact that federal highway aid

may encourage state and local governments to build projects whose costs exceed their benefits. If the benefits from the West Side Highway project are less than the costs, for example, then the federal aid for that project is worth less to New York than its nominal value. In short, many heavily urbanized states would probably be financially better off if they could substitute their own state or local fuel and vehicle taxes for federal highway taxes and highway aid.

Evaluation Transit Aid

The Experience with Transit Subsidies

Three arguments are usually offered in support of government assistance to mass transit. First, mass transit aid is said to reduce the level of automobile use and thereby to alleviate traffic congestion, the demand for costly urban highways, and automotive air and noise pollution and energy consumption. Second, subsidies to mass transit are thought to promote more concentrated, downtown-oriented forms of metropolitan development, with possible attendant infrastructure or energy savings. Finally, transit assistance is expected to benefit low-income house-holds, the elderly, and children—those too poor, too old, or too young to drive or own an automobile.

Reducing Auto Use

The failure of transit aid programs in the 1970s to increase transit ridership significantly was discouraging, especially given the rapid increase in government aid during the decade. Where transit ridership did increase, moreover, only a fraction of the new riders were former auto users; thus, the cost per auto trip avoided was often extremely high.

The new rail systems provided the most dramatic examples of the difficulties of reducing auto use at reasonable cost. Both the San Francisco Bay Area Rapid Transit (BART), opened in 1972, and the Washington D.C. Metrorail, opened in 1976, had anticipated that fares would cover at least operating costs and, in the case of BART, make some small contribution to construction costs as well. Construction and operating costs on both systems proved to be higher than anticipated, however, even adjusting for inflation, while ridership was overprojected by nearly a factor of two.^[24] In 1976, the average fare on BART was 72 cents, while the operating cost per rider was about $1.95 and the total public subsidies—operating and capital—amounted to $3.76 per rider. Since only 35

percent of all BART riders were former auto users (the remainder being former bus patrons or persons making new trips), the subsidy per auto trip avoided amounted to at least $10.74.^[25]

Metrorail has fared slightly better, perhaps in part because it is still under construction and only some of the most productive segments have been opened to date. When the commitment to build Metrorail was made in 1969, the cost of the planned 98-mile system was set at $2.5 billion. The 31-mile Phase III system, which was in operation by late 1979, alone cost over $3 billion to build (in 1980 dollars), and the cost to complete the entire system (since expanded to 101 miles) is now climbing toward $12 billion.^[26] With the Phase III system, total costs amounted to $2.73 per rail trip in 1980, two-thirds of which was for capital.^[27] Metrorail Phase III did reduce rush-hour auto trips to the downtown core by 4 percent.^[28] Since only 36 percent of all Metrorail riders were former auto or taxi users, however, the Metrorail capital cost per auto trip removed from the road amounted to $5.21, while the total cost (operating plus capital) might have been as high as $7.58.^[29] As the outer extensions of the system are opened, the cost per rider and per auto trip diverted are likely to increase.^[30]

Operating aid also brought disappointing ridership gains, despite the fact that the average operating subsidy per rider increased from 5 cents in 1970 to 76 cents in 1982. The small ridership increase was caused in part by the absorption of much of the aid in reduced productivity and higher unit costs rather than in fare reductions or service improvements. Between 1970 and 1982 the average wage for transit drivers grew 40 percent more than inflation, for example, while labor productivity, as measured in annual vehicle hours of service per full-time employee, fell by 20 percent.^[31] The rapid cost increases meant that the average real fare declined by only 19 percent from 1970 to 1982, while the number of vehicle miles of service offered by the transit industry increased by only 13 percent.^[32]

Even where operating aid led to significant fare reductions or service improvements, the ridership increases were often relatively small. In the Boston metropolitan area, for example, government operating subsidies for mass transportation increased from approximately $80 million in 1971 to $233 million in 1980, while passenger revenues remained relatively constant at $50–60 million per year. The operating aid was used to maintain existing service and keep the fare at 25 cents, despite rapid inflation in operating costs. Inasmuch as the consumer price index more than doubled bwteen 1971 and 1980, retaining the 25-cent fare amounted to reducing the fare by half (in real dollars) over the course of

the decade. Despite this substantial decline in price, ridership remained stable or declined slightly throughout the 1970s. In late 1980 and 1981, when financial realities finally forced Boston officials to raise fares, the response in ridership and auto usage was also relatively small.^[33] When bus fares increased by 100 percent and rail fares increased 50 percent in August 1981, for example, ridership fell by only 10 percent and car usage in the metropolitan area increased by at most 0.5 percent.^[34] Boston temporarily retained some mass transit riders by avoiding a fare increase during the 1970s, but the cost of doing so amounted to around $2 to $3 in added subsidy per passenger trip saved.^[35]

About a dozen metropolitan areas significantly increased their transit ridership during the 1970s, sometimes at a relatively modest subsidy cost of 50 cents or so per added rider. In most cases, however, fare reductions and service improvements were not the only factors contributing to the ridership gains. A few cities allowed their transit service to deteriorate greatly before public subsidies began; thus the increase was significant only compared to the very small base. In several other cities, policies to discourage auto use—such as parking taxes or controls, priority for buses in traffic, or exclusive bus lanes—appear to have contributed greatly to increasing patronage.^[36]

In short, the experience of the 1970s showed that it is difficult to expand transit patronage—and reduce auto use—much beyond transit's traditional markets. Transis has always been best in serving commuters to the downtowns of large and congested metropolitan areas, such as New York, Chicago, San Francisco, and Boston. The combination of high downtown parking fees and slow traffic speeds makes transit competitive with the automobile for these trips, particularly if transit can be protected from the congestion with either an exclusive rail or bus right-of-way or bus priority in traffic. The irony is that where transit contributes most to the alleviation of auto problems, it is probably least in need of government assistance. Where traffic congestion is severe and parking charges high, transit can offer relatively competitive service without significant government aid.

Concentrating Urban Development

The supporters of the new rail transit systems hoped they would lead to more concentrated forms of urban development as well as reduced auto use. A review of the experience of new rail systems in San Francisco, Washington, Montreal, and Toronto and major rail extensions in other cities suggests, however, that rail service is not sufficient to insure intensi-

fied development around suburban or outlying stations. A variety of other complementary factors are needed, including rapid economic and population growth in the metropolitan area and support from the community and local land use policies. In Toronto, for example, clusters of high-rise apartment buildings and retailing appeared around outlying stations, in part because rapid metropolitan growth encouraged new construction and local officials offered powerful zoning incentives for development around stations. With BART and Montreal's Metro, by contrast, there was little development around suburban stations, due in part to slower growth, community opposition, and difficulties of land assembly.^[37]

The principal land-use effect of these new rail systems appears to be in facilitating downtown high-rise office development. San Francisco, Toronto, and Montreal have experienced rapid downtown development since their new systems were completed; Washington, D.C., and Atlanta, where partial systems are now in operation, show signs of similar trends. Other non-rail factors probably contributed, however, since the downtown office booms often pre-dated the subway openings and other cities without new rail systems experienced similar developments.

If new rail systems promote downtown offices but leave suburban development dispersed, their land-use effects are similar to those of the radial expressways that transit advocates often oppose. By increasing the accessibility of the downtown to the suburbs, the rail system promotes more concentrated employment in the downtown and the displacement and dispersal of residences into the suburbs. The overall effect is probably to increase rather than reduce the average length of the commuting trip and to turn the downtown and its immediate neighborhoods into areas that are active during the day and have few residents and associated nighttime activities.^[38]

Helping the Poor

The experience of the 1970s also suggests that transit aid is not always an efficient or effective way to help poor people.^[39] Transit accounts for only a modest fraction of the trips poor people make. Although poor households are more dependent on mass transit than rich households, in 1977–1978 mass transit accounted for only 6.9 percent of the trips made by metropolitan residents with household incomes below $6000, a figure not much different from transit's 3.4 percent share of trips by metropolitan residents of all incomes.^[40] While 85 percent of all metropolitan households owned at least one car in 1977–1978, a surprising 54 percent of

households with incomes below $5000 and 83 percent of households with incomes between $5000 and $10,000 owned autos.^[41]

Poor people make up a small fraction of all transit riders, moreover. Travelers with 1977–1978 household incomes below $6000 accounted for only 25 percent of all transit trips in metropolitan areas, for example.^[42] The share of subsidy that goes to poor transit riders is probably even smaller, since low-income ridership tends to be concentrated on services where the subsidy per rider is relatively low. Subsidies per rider are typically three to four times higher on commuter railroad services than they are on buses or subways, for example, yet only 9 percent of all commuter rail trips were made by persons with 1977–1978 household incomes below $6000, while 38 percent were made by travelers with incomes above $25,000.^[43] Within rail and bus systems, transit subsidies tend to be higher for long-distance and peak-hour travelers, because the costs are higher while fares typically are not. Poor people ride much shorter distances than average because they live closer to the city center; the poor also travel disproportionately in the off-peak hours since many are unemployed.^[44]

Transit subsidies could better help the poor, of course, if they were targetted directly at poor travelers or the services they use. Virtually every transit system offers discount fares for handicapped and elderly persons, but these discounts help only slightly: 75 percent of the metropolitan poor are neither handicapped nor elderly.^[45] A more promising possibility is to subsidize poor users directly; poor persons might be issued discount fare identification cards or, to avoid the possible stigma of an identification card, could be sold monthly transit passes or a supply of transit tokens at a discount price. Targetting the subsidies to services such as bus, off-peak, and short-distance that poor people tend to use is less efficient but would be an improvement over the existing system. The federal government has sponsored several successful demonstrations of the possibilities of direct subsidies to poor people, but so far the practice has been adopted only in a handful of cities.^[46]

The Problems of Federal Aid

The Federal Rationale

The rationale for federal involvement in urban mass transit shares many of the weaknesses of the rationale for federal aid to urban highways. The argument most often cited in the early 1960s debates over the

initial federal capital grant program was the need to counterbalance federal highway aid. The federal and state highway trust funds, all financed with dedicated gasoline taxes, were thought to have induced state and local governments to channel too much capital spending into highways and too little into mass transit. Transit had declined because of undercapitalization, the argument continued, and federal transit aid was needed to correct the imbalance.^[47]

The failure of the transit investments of the 1970s to increase ridership significantly suggests that undercapitalization was probably not a major cause of the decline of mass transit patronage. Rising real household incomes, suburbanization of jobs and residences, and other demographic trends probably played more important roles in the postwar patronage losses. Even if local governments had seriously overinvested in highways and underinvested in transit, a massive new transit aid program may not have been the correct answer. By subsidizing both the highway and transit modes the federal government might reduce the balance between transit and highways only at the risk of overcapitalizing transportation in general. Reducing or eliminating the federal highway aid program might have encouraged more balanced spending on all forms of transportation.

Discretionary Capital Grants and the Rail Starts Problem

Whether mass transit was ever undercapitalized, the federal discretionary transit capital grant program now appears to encourage overcapitalization. The transit capital grant program has many of the same features as the Interstate Highway program. The federal government will pay up to 80 percent of the cost of a transit capital project (after 1982, 75 percent). Since grants are allocated for specific projects at the discretion of the Secretary of Transportation, moreover, there is no clear limit to the assistance a city may receive; the use of aid for one project does not preclude aid for others.

State and local governments have an obvious incentive to use the discretionary capital grant program to substitute federally paid capital costs for locally paid operating expenses.^[48] For example, one early study examined the possibility that transit authorities might save on bus maintenance costs by using federal capital grants to retire buses earlier than they would otherwise. That study never investigated whether transit authorities actually engaged in this practice but estimated that if they did, the waste would amount to more than 20 percent of the value of the federal capital grants.^[49]

Since most federal capital grants have been spent on rail transit, a most troublesome possibility is that local authorities may substitute federal capital grants for local operating expenses by building rail instead of bus systems. Among the advantages often cited for new rail systems is the potential for reducing labor costs by replacing several bus drivers with a single train operator. There are some offsets, of course, such as the added costs of maintaining track, signals, power distribution systems, stations, and other facilities not normally found on a bus system. In some situations, however, rail may offer savings in operating costs even though total costs, including capital, are higher.

The availability of federal aid has been a strong factor in the recent revival of rail transit. Rail systems are often classified as either heavy or light: on heavy rail systems, such as a conventional subway or metro, passengers board from platforms and power is distributed by a third rail; on light rail systems, such as streetcars or trolleys, passengers may board from street level and power is distributed by overhead catenary. From the 1930s through the late 1950s, rail transit mileage declined steadily in the United States as many cities replaced their streetcar lines with buses.^[50] When the federal capital grant program began in 1964, five United States metropolitan areas had rail systems that pre-dated World War II, Cleveland had opened a new heavy rail system in 1958, and San Francisco had already begun the construction of BART. Between 1964 and 1984, almost every older rail system extended its lines, four cities opened new heavy rail systems, one opened a new light rail system, and another five cities began building new light rail systems. As of 1984 at least 13 more cities were in various stages of designing or planning new rail systems, while virtually every existing rail system was planning extensions to existing lines (Table 7.8). While most of the rail construction in the 1960s and 1970s involved heavy rail systems, almost all the cities building or planning new systems in the 1980s are considering light rail.

Several studies have compared the operating and capital costs of using buses and heavy rail to serve metropolitan commuting trips. To insure that the cost comparisons are meaningful, these studies attempt to make the quality of service on the two modes comparable. While it is difficult to make the services exactly equivalent, the cost studies usually insure that the ratio of seated to standing passengers is the same on rail and bus and that bus travel times are comparable to heavy rail by, for example, operating the buses on an exclusive busway or an expressway whose access ramps are controlled to prevent congestion. The cost studies generally show that the bus is less expensive than heavy rail in radial corridors with

weekday peak-hour travel volumes below 15,000 transit passengers. When the peak travel volumes exceed 30,000 transit passengers, heavy rail is less expensive, and between 15,000 and 30,000 passengers the choice is usually determined by local conditions, such as the availability of a relatively free-flowing expressway that buses can use.^[51] Radial corridor volumes above 15,000 persons per peak hour probably are found only in metropolitan area with populations of at least several million, with high population densities and a strong downtown.^[52] Of the metropolitan areas that opened or began construction of heavy rail systems in the 1970s, probably only San Francisco and Washington, D.C., have corridor volumes that exceed the minimum heavy rail threshold. Even in those metropolitan areas, moreover, a careful bus-rail cost comparison would probably suggest much smaller heavy rail systems than the ones opened or planned.

The high cost of the heavy rail systems built in the 1970s stimulated interest in light rail systems in the 1980s. Unfortunately, there has been very little analysis of the comparative costs of light rail and bus.^[53] Light-rail proponents argue that it can be much cheapter than heavy rail because it can operate safely on city streets where building an exclusive right-of-way or subway would be expensive. Because the right-of-way need not be grade-separated, the argument continues, a light rail system does not need nearly as high corridor volumes to bring its per-passenger costs down to reasonable levels. If the light rail system is not grade-separated, however, the comparison bus service probably can also avoid expenditures for exclusive busways or similar facilities. Whether the savings in rail costs outweight the savings in bus costs may depend greatly on local conditions. Many of the metropolitan areas now building or planning light rail systems have such modest populations and densities as to make construction of even light rail systems very risky.

The high construction costs of some of the new rail systems built to date lend credence to the view that the total costs of many of these systems are greater than those of comparable bus systems. Even more striking is the possibility that at least some of the new rail systems may have increased rather than reduced locally paid operating costs. San Diego's new light rail system, for example, is widely touted as a success because farebox receipts cover 80 percent of its operating expenses, compared to only 40 percent on the San Diego bus system. The rail system consists of a single sixteen-mile line connecting the downtown with the southern suburbs and the United States–Mexican border, and it offers faster terminal-to-terminal travel times than the two bus routes it replaced.^[54] The

original bus routes were among the most heavily patronized in the San Diego system, however, and their farebox receipts exceeded operating expenses. Had the bus routes enjoyed the partially grade-separated right-of-way of the rail system, their travel times probably would have been comparable to rail and their operating profits even greater.^[55] Comparisons of bus costs for other larger rail systems are more difficult, since it is harder to identify the specific bus services they replaced. The steady increases in operating deficits in many of the metropolitan areas that recently opened rail systems suggests that the operating savings over buses may have been small.

Federal regulations requiring analysis of low-cost alternatives and restricting new rail starts to one line at a time have not been effective in limiting new rail construction.^[56] The 1976 requirement for alternatives analysis is credited with eliminating Denver's heavy rail proposal and reducing the size of the Buffalo light rail plan (when bus systems were found to offer comparable service at a fraction of the cost). Although the analysis requirement probably has discouraged or reduced the scale of other proposals, since the Denver and Buffalo decisions federal officials have not turned down any rail proposals after alternatives analysis, and the number of cities engaged in rail planning has increased.

As in the case of highways, the complexity of urban transportation planning helps to limit the effectiveness of federal rail transit regulations. Forecasting rail and bus patronage is a difficult art, for example, since ridership projections are sensitive to the detailed spatial distribution of future jobs and residences, which is itself uncertain. The non-rail alternatives can also appear relatively costly and unattractive if they are poorly or unimaginatively designed. Although experience has shown that many of the rail analyses overestimated patronage and understated costs, it is often difficult to demonstrate conclusively before the fact that a forecast is unreliable. Given the subtle judgment involved in accurately forecasting ridership and costs or designing alternatives, it is difficult for UMTA's small staff to detect or correct the potential errors in the alternatives analysis.

These problems are often further compounded by strong political support for rail from local officials and their congressional delegations. Local officials displeased with UMTA decisions are often quick to bring congressional pressure to bear on the agency, particularly when the UMTA position seems petty or unreasonable. The FY1982 and 1983 congressional earmarking of rail funds for specific cities is perhaps only the most extreme example of this problem.

In May 1984 the UMTA administrator announced that future rail-start decisions would be based on the added cost per new transit rider attracted, using a low-cost bus alternative as the baseline for measuring added rail costs and ridership. The hope is that the figures will embarrass Congress and discourage future earmarking for particularly cost-ineffective rail projects. The cost-per-rider estimates will be based on data from the alternatives analyses, however, which will increase the need for UMTA to monitor the quality of these studies.^[57]

Incentives of Operating Aid

The principal debate over federal operating assistance focuses on the extent to which it either substitutes for state or local aid or distorts state and local decisions in undesirable ways. Federal assistance was originally restricted to capital expenses in 1964, for example, partly out of fear that operating grants would reduce incentives for transit managers to control labor costs and improve productivity. Capital grants do not eliminate these incentives altogether, of course, since they free local resources that otherwise might have been absorbed by capital needs for operating expenses. With operating aid, however, the incentives might be more direct and obvious.

The relationship between operating aid and efficiency has received increasing attention because of the rapid inflation in transit operating costs during the 1970s. A careful study of the real (net of general inflation) increase in United States bus transit operating deficits between 1970 and 1980 found, for example, that increasing real labor compensation rates or declining labor productivity accounted for 48 percent of the deficit growth, while real fuel-cost increases accounted for only 11 percent of the deficit growth, service expansions only 19 percent, and fare reductions only 22 percent.^[58] Several cross-section and time series studies of transit agencies have also found that higher levels of government operating subsidies are strongly associated with higher wage levels and lower labor productivity. Correlation does not prove causation, however, and none of these studies has been able to demonstrate conclusively that higher subsidies caused operating cost increases rather than vice versa.^[59]

A more recent concern is the possibility that federal operating grants may encourage expansion of transit service into markets where transit has little chance of capturing significant ridership. Many smaller metropolitan areas greatly expanded their transit service during the 1970s even though low levels of auto congestion and low population densities make

it difficult to attract ridership. Some larger metropolitan areas greatly increased suburban and crosstown services, where the possibilities for attracting additional patronage at reasonable subsidy costs is probably also limited.

The cost inflation and the expansion of underutilized services encouraged by federal operating aid may well be centered on small, low-density metropolitan areas. State and local government aid accounts for 70 percent of all operating assistance for the United States transit industry as a whole, far less than the 50 percent minimum match required for federal grants. The formulas for distributing federal operating aid also provide less assistance per rider to many of the largest and most transit-oriented metropolitan areas; in the big Eastern and North Central cities such as New York, Chicago, Boston, and Philadelphia, federal aid accounts for less than 20 percent of all operating assistance. In many smaller and low-density metropolitan areas, federal aid is at the 50 percent limit and its potential effect on service and efficiency decisions is much greater.^[60]

Federal officials have rejected suggestions that they monitor the services or efficiency of operating grant recipients, in part for lack of simple and reasonable standards. The best remedy would probably be to increase the local matching requirement or to change the formula to favor the larger, transit-oriented cities more. The operating aid formulas were changed in 1978 and 1982 to increase the large cities' share of funding, but the need for broadly based congressional support has insured that many small, low-density metropolitan areas still receive a disproportionate share of aid.^[61]

Financial Effects on Urban Governments

Federal transit aid is targeted far more heavily at urban areas than federal highway aid, and urban taxpayers almost certainly receive more in federal transit grants than they pay in federal taxes to support the program. In 1980, for example, 34 states—all predominantly rural—received less than $10 per capita in federal transit aid, while only 8 states—all with cities that were building or expanding rail systems—received over $20 per capita.^[62] Since all federal transit aid was financed out of general revenues in that year, the federal tax burden for the program is probably far more equally distributed and urban areas are the net gainers.

As in the case of highways, however, the benefits to urban residents from a federal transit grant may be smaller than the dollar value of the

grant. To the extent that federal grants encourage too many rail transit extensions, expansion of underutilized services, or reduced productivity, the value of the grant to the recipient may be significantly lower than its dollar value. Capital grants may cause additional financial strains for urban governments if the capital projects increase rather than reduce local operating costs. The experience of the San Diego light rail system suggests, for example, that the construction of some rail systems may burden local governments with increased transit operating deficits.

The financial effects of federal transit aid also probably should not be viewed separately from those of federal highway aid. If urban representatives in Congress trade their support for highway aid in return for rural support for transit aid, then it would be impossible to eliminate or reduce the federal highway program without making comparable reductions in the federal transit programs. One recent study of the state-by-state allocations of federal transportation grants and taxes suggests that the urban bias of transit aid is not enough to offset the rural bias of highway aid. Of ten highly urbanized states, only two received more in highway and transit grants than they paid in federal highway and other taxes to support the programs. Almost all the states that received more in federal highway and transit grants than they paid in federal taxes were rural.^[63]

The Federal Role Reconsidered

The Reagan administration's proposals to reduce the federal role in urban transportation represented a significant redirection in federal policy. Despite the unsympathetic reaction from Congress, a reduced federal role might have proven beneficial not only for the federal taxpayer but for urban residents.

The rationale for federal involvement in urban transportation is much less compelling now than when that involvement began. Federal highway aid began early in 1916 because of a national interest in establishing an interstate and interconnected highway system. This concern always applied more clearly to rural than urban highways and has been far less compelling since the development of strong state highway agencies than it was under the old country road system. Federal transit aid was rationalized in part as a needed counterbalance to federal highway aid, but in retrospect reduced federal highway aid might have been a more sensible remedy.

Several of the federal urban transportation programs begun in the postwar period appear to distort local transportation decisions in unde-

sirable ways. Interstate highway aid and discretionary mass-transit capital grants create the greatest problems, since both have low state and local matching requirements and are distributed for specific federally approved projects rather than by formula or for a broad range of potential uses. The Interstate highway grants probably encouraged the construction of some highways that were excessively blighting during the 1950s and 1960s and now may be encouraging the construction of highways that are excessively beautified. The transit capital-grant program appears to encourage excessive spending on transit, particularly in the form of new rail systems.

Some of the federal aid programs are relatively less distorting, particularly the federal urban-systems-highway grants and, to a lesser extent, the transit operating and capital formula grants. In both cases, the funds can be used for a relatively broad range of transportation activities, and state and local spending usually exceeds the minimum federal matching requirements. These less distorting programs account for an increasing share of federal transit funding but a declining share of federal highway aid.

A reduction in federal highway and transit aid might leave urban residents financially better off, as long as it was accompanied by a reduction in federal tax rates. On average, urban dwellers probably pay more in federal highway taxes than they receive in federal highway aid. Although the federal transit program reduces this imbalance somewhat, even when the transit program is included urban residents pay more in federal tax than they receive in transportation aid. This calculation exaggerates the benefits urban dwellers receive from federal aid, moreover, since it ignores the possibility that federal aid encourages at least some projects whose costs far exceed their benefits. If estimates of the benefits from federally supported urban transportation projects were available and could be compared with the federal taxes urban residents pay to support these programs, the case for a reduced federal role might be even stronger.

Chapter Eight—
The Agenda for Metropolitan Housing Policies

Sherman Maisel

Virtually every country, including the United States, has over the years developed a wide range of governmental housing policies. The stated goals are to insure each household the opportunity to live in a decent house in a suitable environment at a cost that would leave sufficient income available for other needs. Governments have also been concerned with the efficiency of housing production, the damage to the industry and to the housing market from cyclical swings in output, and the availability and use of credit. At the local level, deterioration of neighborhoods and central cities impose large-scale costs on individuals. Such changes cannot be reversed by private action. They require government programs.

In the past fifty years the United States government has built up an exceedingly complex, extensive, and expensive mélange of programs in such spheres as assistance to low-income families; community development and slum clearance; mortgage insurance and guarantee; secondary market institutions and operations; tax subsidies for homeownership and rental housing; policy development and research; equal opportunities; and consumer protection.

In seeking to plan an agenda for the future, we must examine the main factors that brought these programs into existence. Were they based on the peculiar institutional requirements of housing, or were they primarily the result of industry and bureaucratic pressures? Are they the remains of logical reactions to past needs that have far outlived their usefulness, or will they continue to be useful? How can their efficiency be improved?

Recent Developments and the Future Housing Agenda

When the Reagan administration came into office, it resolved to change existing housing policies. It was concerned that housing was taking too large a share of the government's budget, of national saving, of capital investment, and of the money available in the credit markets. It believed that "the genius of the market economy, freed of the distortions forced by government housing policies and regulations . . . can provide for housing far better than Federal Programs."^[1]

To implement its program, the administration called for reductions in several types of government expenditures on housing. It advocated curtailing or halting much governmental effort in credit markets. It appointed the President's Commission on Housing, chaired by William F. McKenna. The April 1982 report of this commission is frequently cited as the basic documentation and philosophical basis for the Reagan administration's housing agenda.

The commission identified major problems: (1) affordability of decent houses for low-income families and first-home buyers; (2) a shortage of rental units; (3) a cyclical shortage of jobs and disturbingly large losses and bankruptcies among building contractors; and (4) hard times, losses, bankruptcies, and market disruption in the housing finance industry, particularly among savings and loan associations and savings banks.

The commission recommended that government programs be reoriented in order to address these problems directly. Instead of federal programs, the country ought to rely on the private sector; encourage free and deregulated housing markets; promote an enlightened federalism with minimum government intervention; and recognize the government's continuing role in meeting the housing needs of the poor.^[2]

Its specific recommendations, on the whole, were far from earthshaking. For political reasons, both nationally and within the commission, it failed to recommend action with respect to the most expensive programs. Contrary to its stated philosophy, it urged increased federal intervention in others. Basically, it proposed:

1. That government programs aimed at increasing the supply of dwellings be replaced by a voucher program, which would subsidize people rather than structures.^[3]

2. That most government programs for homeowners' mortgage insur-

ance and in the secondary market be removed or reduced, but that the government aid in the development of new mortgage instruments.

3. That the thrift industry be restructured and both housing finance and building undergo deregulation.

4. That the tax-expenditure programs for homeownership be continued.

5. That the tax incentives for mortgage lending be expanded and new subsidies be introduced.

6. That federal intervention be reduced except when the federal government believed that states and localities were misusing their powers in spheres such as zoning, building codes, rent controls, and due-on-sale clauses. In such cases, the general concept of federalism should be put aside and the federal government should increase its intervention.

On the whole, the administration has adopted this agenda, although the eagerness with which it has pursued particular recommendations has varied greatly.

I believe that both the commission's proposals and the administration's agenda are faulty. It is true that the government urban programs are large and expensive and that in credit markets they have far-reaching effects. However, the administration's agenda does not deal with the most expensive programs—tax expenditures for homeowners and for mortgage lenders. Its suggestions with respect to neglecting construction and privatizing credit appear flawed. It aims at curtailing aid to communities without a careful evaluation of the costs and benefits. While it emphasizes decentralized decisionmaking, its recommendations and laws go in the opposite direction. Most of the suggestions for local action are rehashes of suggestions that have been made repeatedly for over fifty years. It is hard to explain why, if they are truly efficient proposals, they have never been adopted. The agenda is weak because it is based on insufficient analysis, poor economic theories, and perverse views of proper government actions with respect to income distribution. It neglects significant factors in the housing sphere.

In the following analysis I argue that there are major reasons why government intervention in the housing market may improve both individual and national welfare. Both economic theory and history argue against the view that a private housing market can do as well as or better

than one aided by government policies. In the housing sphere, market results may be improved by government policies.

Proper analysis must examine the areas of possible improvement. A logical agenda must consider whether the implementation of government programs is likely to be successful. The amount of government expenditure for each proposal must also be analyzed to judge whether the nation is getting its money's worth or whether particular programs should be cut back or abolished. The competition between housing and other investments should be compared to see whether any clear evidence exists that investment or use of financial or real resources has been excessive. The results of such analysis can be improved if they start and are carried forward from a neutral intellectual base.

From the following analysis, I conclude that some of the commission's proposals make sense, while others should be altered or forgotten.

1. The problem of the ill-housed should be recognized as one caused primarily by households that are poor because they are out of the employment market. On the whole, income too low to afford housing occurs mainly among a few specific groups in the economy, namely, the elderly, families with female heads, and some very large families, principally among minorities. In some of these cases, the type of housing needed can be so specialized that it may require new construction.

2. Currently a problem also exists for moderate-income first buyers and for some tenants just above the poverty line in high-cost localities. The problem arises from inflation, with a resulting mortgage tilt, and also from high real interest rates. These needs can be attacked through improved credit programs to shape cash flows to real costs, through better targeting of tax expenditures, and through more attention to community development.

3. By far the most expensive program is the aid to homeownership through tax expenditures. Not only are these programs expensive, they are perverse: the higher one's income, the larger is the public's contribution. Similarly, tax aids for existing rental units tend to be as large as or larger than for new ones. Only slight effort has been made to limit subsidies to new supply. Tax expenditures are an extremely difficult political problem; still, no agenda for the future should neglect the most expensive of housing programs.

4. The government's attack on many of the credit programs seems ill

advised. These programs perform an important function in reducing risks and increasing the efficiency of the market. Most are self-supporting or could easily be made so. An initial purpose of most of these programs was to aid in the stabilization of production. This logical goal appears to have been forgotten, but it should be reemphasized. Still, most of the pressure for abolishing these programs seems to arise from a misapplication of political dogma. Unless pragmatic, not theoretical, reasons are advanced for terminating successfully functioning programs, they should be continued. However, attempts should be made to improve their efficiency and to see that they serve the needs of the overall economy rather than a few industry groups.

5. Finally, the programs to aid communities and for housing renewal and rehabilitation should be given more complete analysis. Economic theory appears to support such expenditures. It is not obvious why they should be a principal target of budget-cutting. Whether cuts or increases are proper depends on an analysis of their efficiency and what may be gained from existing or better-structured programs.

Federal Expenditures for Housing

In the United States as elsewhere, the unique features of housing have led to numerous governmental programs aimed at solving capital cost and financing problems, reducing cyclical instability, and aiding in the maintenance and rehabilitation of declining communities. The Department of Housing and Urban Development (HUD) lists over seventy programs it currently administers.^[4] This list is far from complete; many of the most expensive and significant programs lie outside HUD's jurisdiction.

Table 8.1 shows that in 1984, the various housing programs entail budget and tax expenditures of about $60 billion. This is 5 percent of the nearly $1180 billion contained in the federal budget plus tax expenditures. Clearly, a reexamination of the housing agenda should play a significant role in any formulation of government policies. The expenditure numbers in Table 8.1 are not exact, nor are the categorical breakdowns; tax expenditures are subject to significant estimating errors, and many programs serve dual purposes. The relative orders of magnitude would not, however, change much if other existing procedures were used.

Similar problems arise in estimating the federal presence in credit markets. Table 8.1 excludes some of the credit transactions that relate to intragovernmental loans used to minimize the budget costs of certain assistance programs. It also excludes Government National Mortgage Association (GNMA) guarantees that duplicate other government insurance and guarantees. However, a fair amount of duplication still remains, because a considerable portion of the portfolios of the Federal National Mortgage Association (FNMA) consists of government-insured and government-guaranteed loans.

Housing Complexity and Expense

Housing remains a universal problem because it is an expensive necessity. It takes a major share of both national and individual resources. It plays a critical role in the ups and downs of inflation. The Reagan housing agenda was initially based on a fear that housing was playing a deleterious part in the national economy. Supply-side economics emphasized the lack of investment in plant and equipment as a key factor explaining the decline in growth during the 1970s, and blamed the unfair or undesirable competition by housing for economic resources for some of this failure to invest elsewhere.

Houses require a great deal of capital, and the annual payments for housing services are large compared to the income of many families. The cost of an average new dwelling unit is currently over $80,000, or more than three times the income of the average family. The amount of capital

required for housing competes directly with other demands on total national saving (capital), such as plant and equipment or lending to the government to fund the public debt. In 1983, residential construction was 27 percent of total private fixed investment, compared to a peak of 42 percent in 1950, 35.5 percent in 1963, and 34.5 percent in 1972. It reflected, however, a considerable recovery from the 20 percent share in the severe recession year of 1982.

Very few homeowners or landlords can accumulate the capital needed for such large investments; it must be borrowed from those who have saved. Lending on residential mortgages is by far the largest activity in credit markets. In mid 1984, outstanding residential mortgages exceeded $1.4 billion, or more than 23 percent of the total outstanding private domestic debt. The task of raising this capital and lending it on a sound basis to millions of individuals occupies a fair share of the time of financial institutions. The market is complex, consisting of both general-purpose and specialized institutions operating in both the primary and secondary markets. Its demand for funds affects—and in turn is influenced in differing degrees by—all other financing activities.

In addition, the size and individual location of each dwelling create specialized needs for the organization of the housing industry and its methods of production. Houses are large, heavy, complex, composed of a large number of materials and subassemblies. Dwellings must be attached to individual sites, which involves a difficult, time-consuming, decentralized development and construction process.

The demand for construction fluctuates widely. Because of the amount of capital required, most of which must be borrowed and repaid over long periods, demand is strongly influenced by changes in interest rates and the availability of credit. Because the additions to supply in any year are small compared to the existing stock, changes in the demand for existing units greatly affect construction demand. Demand is met through a highly competitive decentralized reaction of numerous builders. The supply process is quite lengthy, particularly when the planning, financing, and permit process is included. Thus supply can greatly exceed demand, or can fail to meet it.

As a result of these forces, the number of dwellings started per quarter has varied by well over 100 percent in the course of individual business cycles. For example, the rate of starts at the beginning of 1982 was only about 40 percent of the rate at the end of 1978. By spring of 1983, the rate had increased by over 100 percent. Similarly, residential construction

was over 5.9 percent of the GNP at the end of 1977, falling to about 2.8 percent in mid 1982.

These forces together affect the structure of the housing industry in two ways. First, the industry's organization allows comparatively rapid adjustment of production to the market. Firms have minimal investments in fixed plant or equipment. Relations with labor, material suppliers, and subcontractors tend to be temporary and to allow for large swings in the level of production. Resources are shifted in and out of the industry with comparative ease. The costs of this flexibility are high wages, high peak profits, and relatively unspecialized techniques and equipment. Most observers believe that costs are raised in direct proportion to the degree of fluctuation. Second, even if no cyclical requirements existed, the industry's organization would be heavily influenced by the complexity of its product and the need to produce houses attached to individual sites in disparate local markets. House-building primarily consists of transportation of large volumes of heavy materials to and on specific lots. The land must be specially prepared. The assembly process is complicated by the variety of materials and the physical dispersion of sites.

The degree of rationality with which the industry has adjusted to its unique tasks has been debated for over eighty years. The government has engaged in various minor programs aimed at changing the structure of the industry. President Reagan's commission placed great emphasis on such programs and urged that the dictates of federalism be pushed aside, if necessary, to enforce centralized deregulation on states and localities. I know of no evidence to support the popular belief that the industry has not adjusted in the most efficient manner. No one has been able to show where greater efficiencies would come from or why they have not been introduced if they exist.

The issue of cyclical stability is also debated. Many observers feel that if demand could be stabilized, the costs of production could be reduced. Some of the existing industry features are efficient only because of the need to adjust to wide swings in demand. Because of the need to build in flexibility and to avoid fixed costs, productivity is reduced. Wages and other payments to the factors of production are high to compensate for unemployment, which tends to reach levels well above those of other industries. Because prices shoot up in periods of excess demand, income becomes maldistributed. Excess factor prices recede slowly, if at all, in the following recessions. A more stable environment might lead to significant changes in these areas.

In contrast, some argue that it is good national policy to concentrate instability in the housing sphere. Fluctuations in construction have only minimal effects on the availability of shelter, since movements in vacancies offset most variations in the rate at which new units come into the market. Only an extremely long period of subnormal activity in construction leads to overcrowding.

While disagreements are possible, my own view is that the country would be better off if housing production were more stable. If the economy requires sectors that can release resources in periods of excess demand, it would be better if the instability were shared more equitably with other industries. Shelter costs would be lower and welfare higher with a more stable housing production, leading to improvements in productivity and a more even distribution of income.

Because housing capital requirements are so large and because the structure of the industry is unique, in most economies it has been thought of as a major problem. Over the years the industry's structure has been under constant attack. Because it differs so greatly from others, it has been assumed—but never proved—to be faulty. However, emphasis on housing's competition for both real and financial resources is a new issue. While the scarcity of resources was always of concern in developing economies, most industrialized nations in the past viewed the macroeconomic problem of housing as one of fluctuating rather than excessive demand. Traditionally, their primary concern was over affordability, either directly, that is, the costs were too large a drain on the income of individual households, or indirectly, that is, too many families lived in substandard dwellings since they could not afford the cost of adequate houses.

Table 8.2 illustrates one measure of the affordability problem. It shows the number and percentage of households in 1981 who reported spending more than 30 percent of their income for housing (mortgage payments, property taxes, and utilities for owners; gross rent for tenants).

The required payments for housing services are large, and they are far from constant either in absolute terms or relative to household incomes. Current service costs depend on the real mortgage rate (a function of real interest rates, the costs and risks of mortgage lending, and changes in relative housing prices). They also must cover depreciation (low, in most cases) and basic maintenance. In addition, utility expenses, which grew rapidly with the acceleration in energy prices, must be covered. Finally, houses are taxed for local services which may or may not be related to

their shelter function. Offsetting some of these expenses are potential subsidies, either direct or through income-tax deductions.

In the past decade, the cost of many housing services rose at a more rapid rate than family incomes. The price of housing units rose relative to other prices. In the 1970s, inflation and tax subsidies made real interest rates extremely low or even negative for many families. In the 1980s, underlying real interest rates shot up. Risks inherent in real-estate lending became more obvious, raising real mortgage rates even more. Energy prices rose extremely quickly to record levels. Income-tax changes lowered the tax subsidy to homeownerships, while raising it for rental units.

As real housing costs increased rapidly, real household incomes stagnated. Median real household incomes in 1984 are well below the amounts earned in past peak income years such as 1969, 1973, and 1978. Furthermore, the national poverty rate has been increasing. In 1983, it was at the highest level in eighteen years. Increased costs and decreased incomes have meant that affordability of housing has become a critical national issue.

The degree to which affordability affects households, however, varies widely depending on their income and also on whether they already own a dwelling or are entering the market for the first time.

Difficulties occur primarily among tenant families—particularly those who are poor. More than 45 percent of all tenants reported that they paid over 30 percent of their income for rent in 1981 (Table 8.2). The percentage of poor in this category was nearly 79 percent. While rents as a share of income fall as incomes rise, affordability was also a problem for moderate-income families.

Table 8.3 measures changes in affordability for home purchasers. It indicates the percentage of families who could meet mortgage payments and property taxes on the median newly built house if they allocated no more than 30 percent of their income for this purpose. Under this standard, the percentage of families who could afford a new house fell from about 60 percent in 1970 to 48 percent in 1975, and to 32 percent in 1980. It reached a low of 27 percent in 1981. With interest rates below their 1981 highs and incomes rising faster than house prices, the index returned to 32 percent in 1984–just more than half of its 1970 level.

Such simple indexes tend to be arbitrary. Definitions differ, as do the underlying data.^[5] Tastes change. With rising incomes families can spend more for housing. The percentage of income the average family devotes to housing has risen and can rise further. By definition, half of the houses produced cost less than the median, while existing houses cost less than

new ones. The number of potential purchasers is raised because of these lower-priced units. On the other hand, the index makes no provision for the fact that housing markets are local. For example, in California housing prices rose far more than the national average. In California, according to this index, the percentage who can afford a new house is only two-thirds of the national average, while in highest-cost localities it is less than half.

In contrast, Table 8.2 reports that even with a broader definition of expense, only about 19 percent of owner households reported spending more than 30 percent of their income for housing. This figure is much smaller because most owners bought their houses in earlier days at lower prices. On the average, owners occupy their houses for about eight years. Because mortgage expenses tend to be fixed at the time of purchase, rising incomes and inflation decrease the mortgage burden. With higher prices, owners who relocate have capital gains which reduce their mortgage requirements. In 1981, almost one-third of owner-households reported no outstanding mortgage.

This and similar analyses indicate that the problem of affordability can be divided. It differs somewhat for three separate groups: families in poverty or with low incomes and special housing needs; first-time buyers or those forced by circumstances to move from a low-cost to a high-cost locality; and the majority of families, who can afford adequate housing but who would prefer to have more income available for other uses. If the expenses of shelter were lower, they would be better off.

Rental Assistance for Low-Income Families

Table 8.4 estimates that the United States government currently spends about $12 billion annually on assisting low-income families to meet their rent bills; of this, only about $2 billion is spent in rural areas. About 3.9 million households receive aid. In this table, as in the preceding ones, estimates have considerable variance. The actual subsidies have been greatly manipulated for budget purposes, making exact estimates difficult if not impossible.

The Reagan administration has recommended that the number of assisted households be limited to current totals, that expenditures be reduced, and that in the future rent vouchers be gradually substituted for existing programs: "The purpose of Federal housing programs should be to help people, not to build projects."^[6] Congress has more or less acquiesced. The 1984–1985 programs call for minimum construction of

TABLE 8.4 Federal Housing Assistance Programs (billions of dollars)   SOURCE: U.S. Office of Management and Budget and Congressional Budget Office.

about 12,000 new units for public housing, Indians, and the elderly together with an expanded voucher demonstration program.

These proposals raise several issues. Why is a federal housing program necessary if it is not to deal with the supply of units? Since housing vouchers are an inefficient method of subsidizing incomes and exert a minimal impact on housing standards, what good does such a program hope to do? Table 8.2 showed that affordability is primarily a problem for the poor. In 1984, between 14 and 15 percent of the population were in poverty. Of this group, about 79 percent of renter households and 64 percent of owner households spent over 30 percent of their income for housing. This target group for housing assistance is enlarged by the need to include some families lying just above the poverty line.

Table 8.5 illustrates some of the details of the poverty class. Only about one-quarter were families with a male head under 65. Of these, a minority were fully employed. In fact, only about 10 percent of the over 12 million households in poverty had a fully employed head. The poor are concentrated among families with female or non-family heads and the elderly; these groups tend not to be in the labor force. Of the traditional families with fully employed heads included in the poverty group, most are unskilled minority workers with large households. (Data are corrected for family size.)

Initially housing assistance programs were aimed at the ill-housed one-third of the nation. However, many housing difficulties facing normal families in the past were solved by a rise in income. Current estimates are that about 7 percent of urban houses are substandard. For various tech-

nical reasons, this probably overestimates the problem. Substandard housing is hard to define statistically; the Census Bureau has worked on this problem for thirty years without much success.

Because they lack potential wage-earners, most households in poverty do not have incomes that rise with growth in the economy. They cannot afford standard housing now and will not be able to do so in the future. These people constitute the hard core of both the housing and poverty problems. However, when we examine government housing programs, we note that only from 15 to 20 percent of those in poverty receive direct housing assistance. (Probably one-third of the assisted lie above the poverty line.) How is this possible? What do the other 80 percent of the poor do to obtain standard housing? About half of the unassisted poor are homeowners. Their current housing costs may be low, or they may have additional capital. The majority of the remaining poverty households have their housing paid for through Aid to Families with Dependent Children (AFDC) or other welfare programs.

The fact that so much of housing assistance comes through general income assistance rather than direct housing aid raises the basic issue of why general income supplements should not take the place of housing assistance, especially since most observers believe income supplements are preferable to categorical grants. The question is usually answered in one of two ways: (1) active lobbies insure that aid to the poor through categorical housing grants will exceed that available through income supplements; or (2) the housing needs of the poor are specialized—if government programs did not directly increase the available supply, too

few standard units would be available at affordable rents. Neither of these answers furnishes a logical basis for a voucher program. If government assistance is not required to increase the supply of units available at low rents, then the basic reason for a separate housing program has disappeared. Money going to a separate voucher program might be better spent on other needed urban programs or on general relief.

Tax Expenditure and Credit Programs

While most debates have been over public housing and rental assistance programs, Table 8.1 showed that these programs make up only a small part of the cost of federal assistance. Most expenditure and credit programs benefit homeowners; only minor portions aid the owners of rental properties. Tax expenditure subsidies reduce costs. Credit programs make more mortgages available at reduced costs and with greater stability. They lower the costs of housing services by reducing interest rates and the costs of production. They also may make houses easier to buy by easing cash flow difficulties.

On the whole, expenditures have not been well targeted. Better structuring of tax expenditure and of credit aids could help, but the Reagan agenda proposes only a few minor improvements. It appears retrogressive in several critical spheres.

Table 8.6 shows the distribution of tax expenditures by type. As indicated in Table 8.1, these are by far the most expensive federal housing programs. Homeowners are allowed to reduce the amount of income tax they pay by deducting interest and property taxes from gross income. Tax reductions resulting from these deductions totaled $34 billion in 1983 and are expected to rise to nearly $41 billion in 1985. Special capital gains treatments, currently at nearly $3 billion, are rising steadily. The other three considerably smaller subsidies cover expenditures that go partly to homeowners and partly to reduce rents on tenant units.

Estimates of tax expenditures are subject to considerable variation depending on the assumptions used. In addition, some observers object to the entire concept of tax expenditures on the theory that the only taxes owed are those enacted into laws. They believe that exemptions should not be examined one by one. Failure to tax should not be thought of as a subsidy even if the tax law is shaped so as to aid a particular function. In addition some, incorrectly, argue that the deductibility of mortgage interest and property taxes on owner-occupied housing is no special advantage because they are deductible in any business. What this argument fails to

recognize is that in the other cases income from ownership must be reported and the deductions are legitimate expenses of creating the income. In the case of owner occupation, the deductions are against sources of income arising elsewhere. A subsidy occurs because we do not report the imputed income from renting the house to ourselves.

Both economic and social reasons have been advanced for subsidizing homeownership. An increase in the percentage owning and living in their own homes can increase both individual and the national welfare. Homeownership increases the efficiency with which housing capital is used: owners take better care of their own property; maintenance is cheaper because much is performed in what would otherwise be leisure time, thus increasing total available labor and the real GNP; and no management costs are incurred. The level of national savings is increased as required mortgage payments lead to forced savings, as do investments in rehabilitation and new improvements. Various social and political advantages are said to arise from the fact that owners are more stable. Homeowners tend to take a greater interest in their community. The desire to increase the value of one's house results in a more attractive and safer neighborhood.

On the other hand, the tax expenditures have been opposed on the grounds that they improperly aid housing at the expense of other investment, are not efficient, are too expensive, and are perverse in their effects on income distribution.

Government credit programs developed because households can invest in housing only by borrowing large sums of money. Furthermore, it has been clear for many years that some of the wide fluctuations in house-building arise from prior movements in the cost and availability of credit. Starting in the Great Depression, the government has offered programs that served to lower down payments and lengthen amortization periods. Other programs have decreased credit costs by widening the geographic markets for funds.

The magnitude of some of the credit programs is brought out in Table 8.7. The top part of the table shows insurance and guarantees for private mortgages and tax-exempt bonds issued by local public housing authorities. In recent years, the three government agencies involved have issued $75 to $80 billion of these guarantees per year. Guarantees outstanding total over $300 billion. The FHA programs are virtually self-supporting. The losses on VA guarantees are not large and are considered a veteran benefit offered in lieu of bonuses paid in earlier wars. Payments on the public housing bonds are part of the costs shown under housing assistance.

The second category shows portfolio loans made by the government-sponsored agencies—the Federal Home Loan Bank System (FHLB), the Federal National Mortgage Association (FNMA), and the Federal Home Loan Mortgage Corporation (FHLMC). In this period these agencies are expected to increase their portfolios by about $10 billion a year. Their outstanding portfolios total about $150 billion. In addition, these agencies plus the Government National Mortgage Association (GNMA) are expected to issue about $100 billion in bonds, certificates, and guarantees against private mortgages. Only part of these sums are in addition to those listed above, since all of those from GNMA plus parts of the others duplicate to a large extent guarantees already given by the FHA and VA and thus are already counted. Including these in the government total is double counting.

The reasoning behind each of these programs is quite straightforward. A reduction in the price of credit can make houses more affordable. Large cyclical swings in credit availability cause major inefficiencies as well as distress for the firms and labor employed in house-building. High credit costs and periods in which credit is rationed or unavailable cause the average level of housing production to fall below the optimum. If government programs could be properly devised and administered, they would serve to improve the nation's housing standard, increase economic stability, and lower housing costs.

The basic idea that the government could lower credit costs and increase credit stability has led to the growth of the programs shown in Table 8.7. The FHA–VA insurance and guarantee programs have at times underwritten nearly half of all housing starts. In recent years, they have also been prominent in helping finance the resale of existing units. The direct purchases, mortgage pass-throughs, and other funds furnished through GNMA, FNMA, FHLMC, and FHLB have dominated the mortgage market.

Before the introduction of the government programs, housing credit for individuals required large down payments and involved expensive risk premiums and high transaction costs. It depended on loans primarily from local institutions, whose source of funds fluctuated widely. This meant that the supply of mortgages was cyclical. The government programs introduced standardized instruments. They enabled shares in residential mortgages to be bought in a manner similar to the purchase of bonds. As a result, risks to lenders were reduced significantly and major credit markets were opened to mortgage borrowers. In theory, the government agencies also were to furnish credit so as to stabilize production. In fact, policies primarily attempted to increase availability as rapidly as possible. Stabilization was of secondary importance.

Ownership Costs

While expensive to the government, the programs to aid homeownerships were extremely successful through the 1970s. Homeownership rose steadily, from an estimated 44 percent in 1940 to 65 percent in 1981. The FHA–VA programs gave the initial impetus to increasing ownership. They reduced down payments to between 0 and 5 percent. The need to save for a down payment no longer inhibited ownership. The prime restraint became the amount of income available for mortgage payments. Initially these programs plus the growing secondary market helped solve the payment problem also. By reducing risks through better underwriting and pooling, the government was able to extend the amortization period to thirty years. Interest rates were reduced because additional funds were brought into the mortgage market and transaction costs were lowered.

Ability to meet mortgage payments gradually eroded, however, under the pressure of the mortgage tilt problem. In periods of inflation, although real interest rates and costs of shelter may not be high, many families cannot qualify for mortgage credit. This is true even though these families could afford to pay the real costs of owning and occupying a house. The

risk to lenders is low, but nominal mortgage payments are high—they are front-loaded. The share of a family's income required for mortgage payments in the first year of owning a home may be two or three times their real costs and well above the share of income to be paid after twenty years. The number of people disqualified by current interest rates depends on the rate of inflation and the way loans are structured. The net effect of the mortgage tilt, however, is greatly to exacerbate the affordability problem.

Furthermore, after 1980 the real costs of housing increased. Until then, the interaction of inflation and the tax subsidy meant that owners who could benefit from the tax deductions had negative real costs. Rosen, for example, shows that for the period 1975–1979, a typical California homeowner making average mortgage payments and whose property appreciated at the average rate had a negative cost of ownership of $2,237 per year.^[7] Everyone who could meet mortgage payments profited greatly by owning a home. Since this was obvious to most people, the desire to own spread rapidly.

Since 1980, nominal and real interest rates have increased. A typical family who bought between 1981 and 1984 would have had a positive real cost of $6,423 per year. Neither the tax benefits nor the various credit programs succeeded in reducing the housing burden. Most people entering the housing market without accumulated gains and savings found that affordability was a critical problem.

What, then, should the agenda be? I think the current administration's plan, continuing the tax expenditure program and reducing or abolishing most of the credit aids, is wrong. At the start of the Reagan administration, a strong argument was made that too much investment and credit were going to housing at the expense of other forms of investment. It was assumed that the housing subsidies had reduced costs by so much that expenditures were excessive. The argument appears quite weak. Housing's share of the total fixed investment has fallen, not risen. It was 37 percent of the total in the period 1950–1954 and had fallen to 24 percent for the period 1980–1983.

Tax Expenditures

The claim that the tax system unduly subsidized housing at the expense of other investment has not held up well under careful scrutiny. Fullerton and Henderson state that under the 1980 tax laws, marginal tax rates on owner-occupied housing, at 19 percent, were considerably higher than on equipment, at 9 percent, but were lower than average rates on corpo-

rate assets, at 34.5 percent, and on unincorporated rental housing, at 40 percent. The 1982 tax law did not alter the marginal tax rate for owner-occupied housing, but marginal taxes were reduced still further on equipment (they became subsidized), while the rate on structures fell by 10 to 20 percent. Estimates of marginal tax rates vary depending on assumptions concerning interest rates and the rate of inflation. It is difficult to argue that housing was either aided or hindered by the total tax system in comparison with all other investments. Currently a wide divergence exists in the marginal-tax effects on different types of investment goods.^[8]

Other problems with the existing tax expenditure programs are more important. While these programs did succeed in holding down real costs for most homeowners, they accomplished this goal in an extremely inefficient manner. It is not unreasonable to assume that the large amounts spent on subsidizing homeownership would bring greater benefits if the program were rationalized or if the amounts could be used to support other housing programs or, failing that, if they were used to reduce taxes or the deficit.

Tax benefits and inflation caused major disturbances in housing prices between 1970 and 1980. The negative costs of ownership increased demand. Lagging supply fostered excessive profits and undue increases in the prices of land, labor, and materials.

The tax subsidies through tax-exempt bonds and thrift institutions and those for rental housing are inefficient for other reasons. All suffer because they allow excess returns to inframarginal lenders or owners at the expense of the government, without equivalent benefits to borrowers or tenants. For example, the subsidy through excess depreciation allowances is reestablished on the sale of existing properties, yet this subsidy goes only to the previous owners. It is capitalized into owners' quasi-rents through a higher price-to-rent ratio. It does not affect rents, since these are primarily determined by the price and supply of newly constructed units.

Finally, the existing programs have a perverse effect on income, because benefits depend upon the homeowner's other income. The higher one's income, the higher the subsidy. Many elderly and moderate-income families are not able to benefit fully from the deduction because their itemized deductions prior to the inclusion of housing fall below the standard deduction. Gravelle, for example, estimates that individuals earning $17,000 in adjusted gross income would benefit from only half their potential housing deductions. She also shows that, on average, the amount of tax saving rises rapidly with income.^[9] These tax incentives

have another perverse feature: they are not limited to a single house, but apply equally to second or third houses, whether held for vacation or other reasons.

The form of the subsidy is unfortunate. Other countries cope with these difficulties either by counting the imputed income from owner-occupancy or, at a minimum, disallowing deductions. Reform proposals in this country have concentrated on abolishing deductions for second homes and on allowing deductions only against some standard tax rate such as 14 or 20 percent. Because a majority of voters profit in some degree from this subsidy, it has been politically sacrosanct. As a result, tax expenditures in this category have been increasing steadily. The Congressional Budget Office estimates that they will reach $68 billion by 1988.

The tax expenditure packages should be reshaped; the country is not getting its money's worth from these programs. While they are hard to alter politically, at least marginal adjustments should be possible. One approach would be to target first-time buyers more directly. Tax expenditures could be used to subsidize their savings up to a limited total. Higher tax credits could be given for those whose total deductions are limited and whose marginal tax rates are no higher than 25 or 30 percent. Actual payments or tax refunds could be authorized if the total tax expenditure for a family is below some limited amount.

Credit Programs

Another major thrust of the administration's program has been an attempt to curtail most of the credit programs. The administration has argued that the government was reducing efficiency by subsidizing borrowers and by interfering with the private market. I find but slight evidence to support such arguments. The government, because of its size and credit standing, has been able to reduce risks to lenders and thus has increased market efficiency; it has greatly lowered transaction costs by standardization and its own efficiency. Probably more credit has flowed to housing as a result of lower costs, but the net result should be an increase, not a decrease, in public welfare. Ideological arguments that those services efficiently provided by the government should instead be supplied by higher-cost industries do not seem valid.

The existing programs can properly be criticized for paying too slight attention maintaining stability. They have at times been pro- rather than anti-cyclical. Builders, mortgage brokers, and real estate operators have wanted the cheapest and most available credit, no matter what the state

of housing demand. In response, the government programs have continued to operate at full speed even when available resources were insufficient to meet demands. When demand fell, some programs were helpful in supporting production. However, the programs' total cyclical impact has probably been negative.

An agenda for the future should pay greater attention to the needs of those who may not be able to afford housing when either nominal or real interest rates are high. This group does not include the poor; they need generalized income assistance and perhaps some construction for special-function units. It does not include most existing homeowners, since unless their income falls, they are able to meet these expenses. The program beneficiaries will be found primarily among existing tenant groups or newly formed households. Some find it difficult to purchase because their income is insufficient to cover their necessary expenses. Others want to remain tenants but find that rent increases exceed income increases. Some tenants find their dwellings deteriorating due to lack of maintenance or community decline. Finally, owners may be forced to remain in unsatisfactory units because they are unable to pay the much higher interest rates now current.

To benefit these groups, it appears logical to continue the basic government credit programs in the areas of insurance and guarantees and in the secondary market. In these spheres the government has major advantages over private entrepreneurs. Some risks, such as those of credit crunches or inflation, can best be assumed by the government. In other cases, because catastrophic events are likely to have extremely abnormal distributions, the amount of private capital needed to insure safety may be too large to be profitable. Since homeownership creates public as well as private goods, it may be worthwhile for the government to take unknown risks in order to reshape financial contracts. For institutional reasons, the reaction of financial markets to swings in credit differs widely among industries. All might gain if these swings in housing could be moderated.

Government programs have developed in the field of housing finance for these reasons. The programs have been successful in reducing required down payments and real interest rates. Now new instruments must be developed to shape payment flows more closely to expected incomes. Interest rates now carry high risk premiums, reflecting past variances and future uncertainty. These types of risks are best assumed by the government. The government should place more policy emphasis on the goal of increasing cyclical stability. Policies should not be used to make small cost reductions in boom periods; rather, the government should restrain

its own lending in periods of adequate private credit to be better able to supplement credit when private mortgage flows are curtailed. Finally, more effort could be made to insure the maintenance or renewal of values through community programs. These are the subject of the next section.

Externalities and Aids to Community Development

A third distinctive characteristic of housing is its dependence on external factors for its use and value. Because of its fixed location, the value of a dwelling is greatly influenced by its neighborhood, its environment, its public infrastructure and services, transportation, job locations, and social perceptions. The development of useful lots requires a considerable capital investment both by individual owners and the community.

The private market is simply inadequate to solve the problem of maintaining values, given the extent of external forces. Historically, a critical variable in neighborhood changes has been the migration and immigration of poor families. Central cities have grown and declined as transportation systems have altered and suburbs have been opened.^[10]

Urban problems have been exacerbated by the unequal fiscal burdens and capacities of different localities. Competition among localities for development is shaped by national forces having little or nothing to do with the basic value of individual houses. Because of their durability, dwellings and communities are slow to adjust to change. Much of value can be lost if the market is not aided in its adjustment to change.

The concept of externalities provides one of the clearest justifications for government intervention in the housing area. A family's housing standard may depend as much on its neighborhood and community environment as on its house. Deterioration of neighborhoods occurs for numerous reasons: the infrastructure may be outdated; traffic and smog may reduce livability; the structures may be old-fashioned; competition from newer areas and suburbs may reduce a neighborhood's desirability. While these and similar physical features are important, it is clear that many problems arise from poor people's need for affordable houses. Low rents are feasible primarily through overcrowding and under-maintenance of dwellings. The recent gentrification of many neighborhoods containing most of the same physical features as neighboring slums shows how significant nonphysical factors can be.

The principal local development grant programs have included Community Development Block Grants (CDBG), funded at 3.47 billion dol-

lars in fiscal year 1983, and the Urban Development Action Grants (UDAG), funded at 0.44 billion dollars. The levels of funding for CDBGs and UDAGs are estimated to be at the same levels, $3.47 billion and $0.44 billion respectively, through fiscal year 1985. Two new programs, Rental Rehabilitation and Rental Development (so-called Section 17 programs) were funded in 1984 at $0.15 billion and $0.20 billion respectively; funding estimates for 1985 are $0.15 billion and $0.12 billion.

These programs differ mainly in the rules under which the funds are distributed. Funds are available for use by local authorities on a wide range of eligible activities. Decisionmaking is primarily by the local units. The Community Development Block Grants are distributed on a formula basis, but most of the other funds are awarded through competitions based on applications for specific proposals from individual communities.

The Reagan agenda called for three types of change. The first was to reduce the funds available in real terms. This was accomplished by denying additional funds despite inflationary cost increases. The second was to reduce the amount of federal interference, specifically, to lower the pressure for earmarking funds for housing programs. The third change was to alter the programs for rental housing rehabilitation and development. Section 17 of the Housing and Urban-Rural Recovery Act of 1983 replaced Section 312 of the Housing Act of 1964 as the principal vehicle for aid to rehabilitation. Under the new act, funds could be used for new housing construction.

Some people question the use of federal funds for such specific local purposes. They ask why the federal government should spend its money for aid to localities. People could avoid problem areas by voting with their feet. If they wanted better schools or lower taxes, they could move from the central city to the suburbs, or from one state to another if they wanted fewer services and wanted to pay less. Both mobility and lack of mobility cause problems. Most of the poverty that brings about a need for development and renewal is not local; it is the result of national policies. Few localities can afford the required improvements. Raising taxes beyond some point causes a loss, not a gain, in revenue. Furthermore, it is costly to abandon useful houses and expensive urban infrastructures. The cheapest housing may result from saving existing communities. The aged and the poor lack flexibility in their choices. They may not be able to move, or the trauma may lead to illness and death. On the other hand, many of the high costs in particular localities may be created by refugees or by transients attracted to localities by federal military or other programs. Why

should individual communities be forced to pay costs not of their own making? Why should we force or allow destructive downward spirals to ruin potentially valuable resources? The nation should recognize that many externalities, both positive and negative, reach well beyond the legal limits of specific localities.

Attention has been concentrated on housing assistance and credit rather than on attempts to improve communities. It is far from clear that expenditures have been allocated optimally. It is certainly possible that additional expenditures on community development and the rehabilitation and renewal of existing dwellings would do more to solve housing problems than any other approach. Existing communities have numerous advantages. The land is already developed. An infrastructure, though it may be old and deteriorating, is in place. Transportation is usually adequate, although often not to newer plants in the suburbs. Dwellings are already built; the problem is to maintain their livability while not raising their costs to moderate-income tenants.

Under the Community Development Block Grants, localities have a great deal of flexibility in their use of funds, provided that most funds are used in ways that will aid lower-income families. The largest share of funds (over one-third) is used to rehabilitate houses. The other major uses are for public improvements, public services and facilities. Economic development and planning and administration are other important uses. We have too little information to rank these expenditures with others in the housing field. It seems likely, however, that these sums are used more efficiently than those of many other programs.

Conclusions:
The Agenda for the Future

Housing remains a basic problem because of the large amount of capital required, the burden on individual families, the complexity of producing new supplies, and the actual or potential deterioration of the existing stock. Little progress has been made on the side of production. Real costs have risen relative to other consumer expenditures. Fluctuations have increased rather than diminished in amplitude. No major breakthroughs have occurred in technology or in the organization of the industry.

The past twenty-five years have witnessed improvement in some spheres but retrogression in others. The costs of owning or occupying a house have risen faster than family incomes. The average family entering the market today has a harder time meeting housing expenses than was

true in the past. On the other hand, the percentage of substandard houses, as measured by lack of plumbing or other equipment, or need for repairs, has declined. What has happened with respect to external forces such as neighborhoods and public services is uncertain. For a period, older neighborhoods declined drastically. Some of the worst remain; some have been destroyed. Some have been renewed with public funds, and others have undergone private renewal and gentrification.

The mélange of existing housing programs can be attributed in part to the way the housing problem was visualized over the past one hundred years and in part to happenstance. The current programs are expensive and frequently inefficient. The fact that each type of program tends to have its vociferous supporters is no guarantee of effectiveness.

The new Reagan agenda has emphasized "allowing industry to solve its problems," while imposing spending limits and a reduction of expenditures in real terms. The number of poor aided was reduced. Under current policies most families would have their assistance cut. There would be a gradual shift from other forms of assistance to vouchers. Virtually no aid would be programmed for new dwellings. The supply of units for the poor would be left to the market. Assistance to housing would be limited and would occur mainly through community development and renewal grants. Government and government-sponsored lending programs would be cut back, to leave the field free for the private market. States and localities would be urged to deregulate land and housing. They would be coerced to do so if they failed to meet federal standards.

If we were to start with a clean slate, but able to utilize most of the large sums shown in Table 8.1, how would we structure a logical program? Our program, unlike the administration's, would recognize that some of housing's most enduring problems cannot be solved by turning them over to the private market. Income distribution, reduction of credit risks, and maintaining or increasing the supply of public goods are needs best met through the interplay of market and government forces. For the poor and others who are either not able to afford market rents or who have special non-market needs, we would probably have a mixture of vouchers and new construction of specialized units for large families, for the elderly or handicapped, and in particular localities. We might attempt to substitute general income supplements for the voucher programs.

We would continue most insurance and guarantee programs. The government and government-sponsored agencies reduce risks and transaction costs and, therefore, the real costs of credit. They can meet specific

needs not addressed by the private market. They can hold an open channel for housing to the money markets if normal channels close down, as has periodically been true.

Some aid would be continued to homeownership on the theory that it does increase the national welfare. Tax expenditures for homeownership should probably be a flat amount per family, or at most proportional to income. Special programs might well be established for moderate-income families, particularly to aid in their purchase of a first home.

Money saved by reducing tax expenditures might be spent to improve the urban infrastructure and to renew the existing stock, using block grants. Each locality may well be the best judge of the most effective way to maintain the existing stock: through spending on housing, on transportation, or on other types of public services. Unfortunately, as in many social spheres, we have few good measures of whether the nation is getting its money's worth. The national welfare is almost certainly raised, but we do not know the actual payoffs.

Chapter Nine—
Black Suburbanization in the Eighties:
A New Beginning or a False Hope?

John F. Kain

Roughly half as many black households lived in the suburbs of America's metropolitan areas in 1980 as would be predicted on the basis of household income and family structure. And the black residents of central cities and of those few suburban communities where blacks live in significant numbers are intensely segregated. Segregation of ethnic and nationality groups rapidly decreased after they had attained levels approaching that of blacks today, but the segregation of American blacks has remained at a high level, and by some measures has actually increased.

Current and past discrimination is the principal explanation for black households' segregation. Prejudice of individual white sellers is a contributing factor, but without the active support of real estate brokers, renting agents, insurance companies, lenders, developers, homebuilders, and their associations, these individual acts would have been insufficient first to create and then to maintain this situation. Federal, state, and local governments initially supported the creation of segregated living patterns, then took a neutral stance, and only recently have begun to assist blacks in obtaining housing on a nondiscriminatory basis.

The intense residential segregation of black households with its adverse impact on black incomes, educational opportunities, housing conditions, and general welfare has long been recognized as among the most, if not the most, pressing of the nation's social problems. This recognition, and the growing political power of minorities, caused Congress to declare racial discrimination in the sale and rental of housing to be illegal. But discrimination is still pervasive. Data from the 1980 census of population

and some earlier intercensal data, particularly school enrollment statistics, however, point to the emergence of a new and hopeful pattern of minority residence—the movement of small but significant numbers of black households to a large number of widely dispersed suburban communities. This paper examines these developments using 1970 and 1980 census data for the Chicago, Cleveland, San Francisco—Oakland, and San Jose metropolitan areas.

History of Government Policy

Government support of fair housing is relatively new. Early in the twentieth century, many American communities enacted zoning ordinances requiring block-by-block racial segregation. State governments, having delegated zoning powers to local governments, supported the enactment of these ordinances, which were upheld in state courts. Though declared unconstitutional by the Supreme Court in Buchanan v. Warley in 1917, these ordinances frequently remained in force and as late as the 1950s, attempts to enforce them were still being made (U.S. Commission on Civil Rights 1975, p. 4).

After the Supreme Court's decision in Buchanan v. Warley , restrictive covenants became the preferred method of enforcing racially segregated living patterns. Although they were private agreements, racial covenants were judicially enforced until 1948, when the Supreme Court ruled in Shelley v. Kraemer that their enforcement was a violation of the Fourteenth Amendment. They were still widely employed, however, and many individuals continued to observe them.

Numerous state-chartered organizations also supported racial segregation. In 1950 the National Association of Real Estate Brokers (NAREB) Code of Ethics still stated that:

A Realtor should never be instrumental in introducing into a neighborhood, by character of property or occupancy, members of any race or nationality, or any individual whose presence will clearly be detrimental to property values in the neighborhood.

In Twenty Years after Brown: Equal Opportunity in Housing, the United States Commission on Civil Rights (1975, p. 39) charged the federal government with being "most influential in creating and maintaining urban residential segregation," noting that "for nearly 30 years after the first Federal housing programs were initiated, the Federal government either actively or passively promoted racial and ethnic discrimination." The commission pointed out that for fifteen years, the FHA

Underwriting Manual warned of the "infiltration of 'inharmonious racial groups' into neighborhoods occupied by families of a different race" (p. 40). Furthermore, until December 1949, following the Supreme Court's decision in Shelley v. Kramer , the FHA actively promoted the use of a model racially restricted covenant by builders and owners whose properties were to receive FHA insurance. The commission determined that the "policies of the four federal financial regulatory agencies responsible for the supervision and regulation of mortgage lenders also endorsed overt racial and ethnic discrimination in mortgage lending until passage of the 1968 Fair Housing Law" (p. 41). The commission reported that these agencies permitted mortgage lenders to consider minorities less desirable risks than whites, regardless of their personal or financial worth, and routinely to refuse to provide home mortgages for minorities in nonminority areas. The policy—in force until recently—of discounting all or part of a wife's income in determining eligibility for loans also worked to the particular disadvantage of blacks, since black females have higher labor-force participation rates than white females and thus make proportionately larger contributions to family incomes.

Federal administration of public housing programs also contributed to the maintenance of racially segregated living patterns. The federal Public Housing Administration (PHA) permitted local public-housing authorities (LHAs) to maintain either "separate but equal" or "open occupancy" policies. Most LHAs chose "separate but equal" and, with federal approval, created management offices for projects occupied by whites separate from those for projects occupied by blacks, and made up separate waiting lists based on race. The PHA continued to sanction "separate but equal" policies through the 1950s and into the 1960s, despite the enactment of a growing number of state and local laws prohibiting discrimination in public housing, and several court decisions that found state-enforced segregation in public housing unconstitutional.

LHAs also selected, and the PHA approved the selection of, separate locations for the units to be occupied by white and minority families. In some localities, the policies pursued by LHAs created segregated residential patterns and concentrations where they had not existed previously. In virtually all metropolitan areas, the location of public housing accentuated the concentration of minorities in central cities.

Federal Support for Fair Housing

Significant federal support for open housing began in November 1962, when President Kennedy signed Executive Order 11063 prohibiting dis-

crimination based on race, color, creed, or national origin with respect to the sale, leasing, rental, or other disposition of residential property and related facilities. Although couched in broad terms, the order covered only housing provided through Federal Housing Administration (FHA) mortgage insurance or Veterans Administration (VA) loan guarantees and federally assisted public housing; conventionally financed housing was excluded from coverage. The U.S. Commission on Civil Rights (1979) estimated that the order covered less than one percent of the nation's housing. The order also stipulated that builders and owners of housing could be barred from participation in federal programs if they were found to discriminate. Federal aid agreements executed before 20 November 1962, however, were generally exempt.

Title VI of the Civil Rights Act of 1964 broadened the coverage of Executive Order 11063 and led to important changes in the administration of federal housing programs, particularly public housing. All housing in urban renewal areas was made subject to Title VI, as was public housing regardless of the date of the contract for assistance, as long as federal financial contributions were still being received. However, housing provided through FHA mortgage insurance and VA guarantee programs outside urban renewal areas, as well as Farmers Home Administration housing, remained exempt.

Title VIII of the Civil Rights Act of 1968, widely referred to as the Fair Housing Act, greatly expanded the federal commitment to open housing. It prohibited discrimination in the sale or rental of all housing except single-family homes and owner-occuped one-to-four-unit structures sold or rented without use of a broker and without publication, posting, or mailing of any advertisement indicating any preference, limitation, or discrimination based on race, color, religion, or national origin. Title VIII covers more than 80 percent of all housing and encompasses the activities of all segments of the real estate industry including real estate brokers, builders, apartment owners, sellers, and mortgage lenders as well as all federally owned and operated dwellings and dwellings provided by federally insured loans and grants. Prohibited activities include:

1. Refusal to sell or rent a dwelling.

2. Discrimination in the terms, conditions, or privileges in the sale or rental of a dwelling.

3. Indicating a preference, limitation, or discrimination in advertising.

4. Representation to a person or persons that a dwelling is unavailable.

5. Denial of a loan for purchasing, constructing, improving, or repairing a dwelling.

6. Discrimination in setting the amount or other conditions of a real estate loan.

7. Denial of access to or membership in any multiple-listing service or real estate brokers' organization.

Title VIII also explicitly prohibits blockbusting —convincing owners to sell property on the grounds that minorities are about to move into a neighborhood—and steering , the process of directing a racial, ethnic, or religious group into a neighborhood in which members of the same group already live and directing other persons away from this neighborhood.

While Title VIII is very comprehensive in both its coverage and the range of prohibited activities, its effectiveness has been limited by lax enforcement. The Department of Housing and Urban Development (HUD), which is responsible for the overall administration of Title VIII, may only investigate and conciliate complaints of housing discrimination. The Justice Department may initiate litigation, but only when there is a pattern or practice of housing discrimination or where issues of housing discrimination are of general public importance. In its March 1979 report on the federal fair-housing enforcement effort, the U.S. Commission on Civil Rights gave the Justice Department's Housing and Credit Section generally high marks for its litigation efforts but noted that it had brought only 300 cases since its creation in 1969 (p. 70).

In spite of its legal and symbolic importance, Title VIII has several glaring deficiencies: it does not authorize the Secretary of HUD to initiate investigations unless a complaint has been made, even if the Secretary has reason to believe the law is being violated; it makes no provisions for third parties to file fair housing complaints; it does not provide HUD with enforcement authority; it makes no provision for regular collection of data on those subject to its prohibitions; it does not explicitly authorize the use of testers; it includes too narrow a litigation authority to the Department of Justice; and it limits the awarding of attorney's fees in private litigation to those plaintiffs who are "not financially able to assume said attorneys' fees." Several efforts to strengthen the Fair Housing Act have been defeated, as have several conservative attempts to weaken it.

Decisions by federal courts have been at least as important as congressional legislation and Executive Orders in combating housing market discrimination. In Jones v. Alfred H. Mayer Co. , the Supreme Court held

that in the Civil Rights Act of 1866 Congress intended "to prohibit all discrimination against Negroes in the sale or rental of property—discrimination by private owners as well as by public authorities." The Supreme Court thus, unlike Congress in the Civil Rights Act of 1968, allowed no exceptions. Subsequent decisions have established that a plaintiff can recover both punitive and compensatory damages as well as attorney's fees. Rulings supporting the awarding of attorney's fees have been crucial in enabling minorities to exercise their rights to rent and buy private housing, although use of the courts to secure these rights requires that black households know their rights under the Fair Housing law and depends on the willingness of private attorneys to bring fair-housing cases.

The Carter administration exhibited considerable ambivalence toward racial discrimination and segregation. The first drafts of Carter's 1978 urban policy statement completely ignored these issues, and the final version was not much of an improvement. In 1979, however, the Carter administration supported amendments to the fair-housing law that would have added an administrative enforcement mechanism, extended the statute of limitations from 180 days to two years, made attorney's fee awards available to all victims of discrimination, and authorized court proceedings to hold a house or apartment off the market while the discrimination claim was being decided. The bill passed the House of Representatives but failed in the Senate in December 1980, when its proponents fell five votes short of the two-thirds required to end a filibuster. In the same month the Carter administration belatedly sent to Congress an extensive set of Title VIII regulations. The proposed regulations—which did not become operative—would have provided the first comprehensive interpretation of Title VIII, covering both private market and federally assisted housing and including definitions of what constitutes discrimination and who may sue and be sued.

The Reagan administration's first fair-housing initiative was to recall the Title VIII regulations that had been submitted to Congress in the waning days of the Carter administration. Although HUD gave assurances that it would resubmit revised regulations after it had an opportunity to review them, no regulations had been proposed by the forty-fourth month of the Reagan administration.

During the Reagan administration's first thirty months, the Justice Department filed only six fair-housing cases, whereas forty-six cases were filed during the first thirty months of the Carter administration (Miller 1984). Selig (1984), Miller (1984), and others accused the Reagan admin-

istration of deliberately bringing a large number of low-impact cases, abandoning efforts begun during the Carter administration to focus the Justice Department's limited resources on a small number of high-visibility, precedent-setting cases. In addition, the Citizens Commission on Civil Rights contended that Reagan administration fair-housing cases were settled on terms that imposed much less significant penalties and fewer obligations than agreements secured in earlier cases (Citizens Commission on Civil Rights 1983).

The Reagan administration has also refused to employ the "effects test," even though federal courts and Congress have repeatedly stated that a violation of the Civil Rights Act may be proven by showing either that an intentional act of discrimination has been committed or that the actions complained of have had a discriminatory effect. Commenting on the administration's failure to use the effects test, the authors of an American Civil Liberties Union review of the Reagan civil rights record stated that "in defiance of the clear legal standard, the Reagan Administration has taken upon itself to redefine the law to its own liking" (1984, p. 7). Speaking more generally of the Reagan administration's civil rights record, these authors concluded:

The actions described above reveal a pattern of disregard by the Executive Branch for the rule of law and the mandates of the legislature. . . . Since 1803, amendments to the Constitution and dozens of congressional and judicial decisions have reaffirmed the principle . . . that the officers of the Executive Branch do not operate with unfettered discretion, particularly when constitutional rights are at stake. Their clear duty is to carry out the laws enacted by Congress, so long as those laws are consistent with the Constitution. To an unprecedented degree, the officials of this administration have flouted that duty.

The Citizens Commission on Civil Rights added that the Reagan administration has attempted to eliminate regulations requiring localities to certify that they will comply with civil rights laws and Executive Orders on equal housing and with the nondiscrimination provisions of the Community Development Act itself (p. 63); with diluting voluntary affirmative marketing agreements; with using the Paperwork Reduction Act and regulatory reform as excuses for eliminating collection of essential data on the race and sex of beneficiaries of federal programs; and with using budget cuts and the allocation of this smaller budget to low-impact programs that further weakened HUD's already inadequate enforcement efforts.

The Extent of Racial Segregation and Its Causes

Racial segregation in America has two dimensions: first, vastly disproportionate numbers of blacks live in the central cities of the largest metropolitan areas and blacks are seriously underrepresented in suburban communities; second, blacks are intensely segregated within both central cities and suburban areas. Evidence on the first point is provided by Table 9.1. The mean (unweighted) proportion of SMSA nonblacks residing in the suburbs of all thirty-one SMSAs in 1970 was 3.3 times as large as the mean (unweighted) proportion of SMSA blacks. The weighted mean was nearly twice as large as the unweighted mean, indicating that the concentration of blacks in central cities is greater in larger SMSAs than in smaller ones. The rapid growth of suburban black populations in a number of metropolitan areas during 1970–1980 (discussed below) reduced but did not eliminate the gap in a number of metropolitan areas; by 1980 the weighted-mean black/white ratio had declined to 2.54 and the unweighted ratio to 3.98.

Table 9.2 shows that black households are intensely segregated within both central cities and suburban areas. The segregation indexes measure the extent to which observed patterns of residence location by race differ from proportional representation: a value of zero indicates a completely uniform distribution—the proportion of blacks residing in every block or census tract is equal to their proportion of the entire city or metropolitan area population—while a value of 100 denotes complete segregation—all blocks or tracts are either 100 percent or 0 percent black. Table 9.2 contains both block and tract indexes. Block indexes are generally regarded as better indicators of the extent of segregation, but block statistics are unavailable for many parts of metropolitan areas. Moreover, the heavy concentration of black households in central cities and other areas covered by block statistics and the growth in coverage of block statistics over time means that care must be taken in using block indexes to study trends.

The segregation indexes in Table 9.2 confirm the widespread impression that racial segregation is particularly persistent in Chicago and Cleveland; the indexes for 1980 are insignificantly lower than the indexes thirty and forty years earlier. By comparison, the San Francisco—Oakland indexes are both substantially lower in 1980 and have declined steadily since 1950. Comparisons of the Bay Area with Chicago and Cleveland, however, may be confounded by the presence of Asian- and Spanish-

origin minorities in predominantly black neighborhoods. Statistics on the percentages of central city blacks living in census tracts that were more than 50 and 90 percent black in 1970, however, also suggest that the segregation of blacks is much less intense in the Bay Area than in Chicago and Cleveland. Using the lower threshold, 93.9 percent of Chicago's and 93.7 percent of Cleveland's but only 55.5 percent of San Francisco's black population lived in census tracts that were more than 50 percent black in 1970. Furthermore, no San Francisco blacks lived in tracts that were 90

percent or more black in 1970, as compared to 77.7 percent of Chicago's and 67.4 percent of Cleveland's blacks.

Persons seeking to justify the high levels of black segregation often point to the tendency of various ethnic minorities to cluster in identifiable communities as evidence of the normality of this behavior, suggesting thereby that the residence patterns of black households are explained by blacks' preference for living with other blacks. Studies comparing the residence patterns of blacks and various ethnic groups, however, demonstrate that the other groups exhibit much less segregation than blacks. Unlike the segregation of black households, which has remained at high levels or even increased for seventy years or more, the segregation of ethnic minorities has declined steadily over time (Lieberson 1963; Taeuber and Taeuber 1964; Hersberg et al. 1971). Attitudinal surveys also fail to support the preference argument. Seventy-four percent of black Americans interviewed in a 1969 Newsweek poll responded that they would rather live in a neighborhood that had both whites and Negroes than in a neighborhood with all-Negro families; only 16 percent chose an all-black neighborhood (Pettigrew 1973). Moreover, between 1963 and 1969 the percentage preferring all-black neighborhoods declined, while the percentage preferring integrated neighborhoods increased. Pettigrew (1973), after reviewing eleven surveys conducted between 1958 and 1969, found that "when presented with a meaningful choice between an all-black neighborhood and a mixed neighborhood, black residents overwhelmingly favored the latter."

Sheatsley (1966) and Pettigrew (1973) also documented a steady increase in white racial tolerance. Pettigrew, for example, compared responses to identical questions included on seven National Opinion Research Corporation (NORC) polls administered between 1942 and 1968 and a second set included in five Gallup polls conducted between 1958 and 1967. In the NORC surveys, the percentage of whites indicating that it would make a difference to them if a Negro, equal to them in income and education, moved to their block had declined from 62 percent in 1942 to 46 percent in 1956, to 35 percent in 1965, and to 21 percent in 1968 (Pettigrew 1973). Similarly, 48 percent of whites interviewed in 1958 stated that they definitely or might move "if colored people came to live next door"; nine years later the percentage had declined to 35 percent. An even more recent Gallup survey revealed that the fraction of white households who said they would move if a black family moved next door had declined to only 18 percent by 1978 (American Institute of Public Opinion 1978).

A more elaborate survey of 743 white and 400 black households in Detroit (Farley et al. 1978) obtained results that are generally consistent with those summarized above. Farley and his associates found that, "freed of racial hostility, . . . most Detroit area blacks would select neighborhoods which are about one-half white and one-half black," but concluded, after finding that the whites surveyed expressed a preference for neighborhoods with smaller concentrations of blacks, that "the prospects for residential integration seem quite slim" (p. 336). The interpretation of these preference data, as well as those of other analysts with similar views, has been strongly influenced by Schelling's (1972) theoretical analyses of the relationship between black and white preferences and racial segregation. Yinger (1978, 1980) emphasizes the need to consider more than preferences in applying Schelling's findings to real-world situations.

The low incomes of blacks are an even more common explanation proffered for their underrepresentation in suburban areas and their high levels of segregation within central cities. Noting that the central-city housing stock is older and of poorer quality than the suburban housing stock, this position asserts that blacks must occupy the cheapest central city housing and cannot afford expensive suburban housing. Several careful empirical studies have examined this proposition and have found little or no support for it (Kain 1977; Pascal 1967; Schnare 1977; Taeuber 1968). Nonetheless, this myth continues to find wide acceptance.

The estimate that only half as many black households lived in the suburbs of United States metropolitan areas in 1980 as would be expected on the basis of their incomes and family characteristics is confirmed by a more detailed analysis which calculates the ratio of SMSA black households to SMSA total households in 1980 for 135 household categories (see Table A in the appendix); multiplies these proportions by the number of suburban households of each type to obtain a predicted number of suburban black households of each type; and sums the 135 categories of suburban households. The 135 income–household type categories are defined in terms of nine household-income intervals; three household types—married couple families, male householder; no wife present; and female householder, no husband present—and five age-of-householder categories. The actual number of black households residing outside the central cities of all SMSAs in 1980 was 1,846,303; the predicted number, obtained using the model described above, was 3,631,702, or 1.97 times the actual number.

The gap between the actual and predicted numbers of black households

is even greater for individual metropolitan areas. Analyses for Chicago using 1970 public-use sample data, for example, indicate that the actual black population of Chicago's suburbs is only 30 percent as large as would be expected in the absence of racial discrimination (Kain 1984). The Chicago estimate was obtained using the methodology described above, except that family size was added to the list of explanatory variables.

A similar analysis for Cleveland in 1970, based on both tract and public-use sample data, used 384 types of households and roughly the same methodology to predict the residence locations of black households and nine individual nationality groups that are present in significant numbers in the Cleveland metropolitan area (Kain 1977). This analysis, prepared for the Justice Department as part of its fair-housing litigation against Parma, Ohio, obtained estimates of the actual and predicted racial and ethnic composition for each of the 440 census tracts in the Cleveland SMSA in 1970 and thereby provides a more precise indication of the extent of black segregation.

Figure 9.1 illustrates the distinctive character of black segregation. Virtually all Cleveland SMSA tracts in 1970 were either more than 70 percent or fewer than 4 percent black; 313 of the SMSA's 414 tracts were less than 4 percent black and only 61 were between 5 and 79 percent black. Most of the 61 "integrated" tracts, moreover, were actually "transitional" tracts located at the periphery of the ghetto.

In contrast to the distributions of actual percent black, 302 of Cleveland's 440 tracts were predicted to have between 10 and 19 percent black households. Only 11 tracts had this racial composition in 1970. No Cleveland SMSA tract had higher than 59 percent blacks predicted in 1970, while 70 tracts had actual black percentages greater than this fraction. The predicted proportions, then, in contrast to the actual percentages, cluster tightly around the SMSA mean proportion of blacks.

The Cleveland analysis of ethnic residence patterns demonstrates the vast difference between the tendency of various ethnic minorities to cluster in identifiable ethnic communities, and the intense segregation of blacks. While ethnic neighborhoods exist in every American city, members of ethnic minorities live in all parts of the metropolitan areas. The model used to predict the residential patterns of Cleveland's blacks, described above, is quite successful in reproducing the actual residential distributions of Cleveland's German-, Polish-, Czech-, Austrian-, Hungarian-, Yugoslavian-, Italian-, and Russian-American households (Kain 1977). Yet, as we have seen, the model has virtually no success in predict-

Figure 9.1
Actual and Predicted Number of Census Tracts by Tract Percent Black, Cleveland
SMSA in 1970 Source: John F. Kain, "Race, Ethnicity, and Residential Location,"
in Ronald E. Grieson (ed.),   Public andUrban Economics  (Lexington: Lesington
Books, D.C. Heath and Co., 1976), Table 17-2, p.278.

ing the residence locations of blacks. Instead of providing a rationale for the segregation of blacks, the current and past experience of America's ethnic and nationality groups provides strong inferential evidence that the intense segregation of America's black households is due to a systematic and continuing denial of choice.

Black Suburbanization During the '70s

During 1970–1980 the black populations of the nation's 239 metropolitan areas increased by 1.8 million within constant 1980 boundaries, and the black share of the suburban population of these areas increased from 4.6 to 6.1 percent (Grier and Grier 1983). Statistics on the black suburban population of the nation's large SMSAs (1980 populations of more than 1 million) (Table 9.3) indicate that the growth of suburban black populations was widespread. The suburban black populations of four of these large SMSAs grew by over 100,000, those of four more by

50,000–100,000, and of seven more by 20,000–50,000. The mean increase in the suburban black populations for all 31 SMSAs during the period 1970–1980 was 41,218, and the mean (weighted) growth, which averaged 66.8 percent, was 4.3 times as great as the growth rate of their suburban white populations, or 15,5 percent. Finally, both the unweighted average growth rate of the suburban black population and the ratio of black to white suburban growth rates are substantially greater than the weighted rates, indicating that suburban black populations grew particularly rapidly in smaller metropolitan areas.

Grier and Grier (1983) and others have correctly observed that most black suburbanization during the period 1970–1980 was attributable to the growth of established black suburban enclaves or peripheral expansion of the central city ghetto across central city boundaries. In these instances, the traditional pattern of ghetto expansion seems to be continuing, with the initial entry of black residents being followed by the withdrawal of white demand, rapid growth of the black population, and finally by the area's consolidation with the central city ghetto (Karlen 1968; Kain and Quigley 1975). This form of ghetto expansion may benefit central cities by breaking what long appeared to be a central city monopoly on black poverty, but it does little to change the expectations of either blacks or whites about the consequences of black entry on the eventual racial composition of formerly all-white neighborhoods and communities.

At the same time, analysis of recent census data reveals a quantitatively small but potentially highly important pattern of dispersed black suburban residence—a movement of small but significant numbers of black households into formerly all-white suburban communities. The finding that a fundamental change may be occurring in the residence patterns of black households is based on a fairly detailed investigation of the pattern of black suburbanization in four metropolitan areas, along with a general impression of similar developments in a number of other areas. Chicago, Cleveland, San Francisco–Oakland, and San Jose are examined in detail; statistics for San Francisco–Oakland and San Jose are combined and are reported as the Bay Area.

Table 9.4 shows that 69 of Chicago's 116 suburban communities (59.5 percent) and 16 of Cleveland's 41 suburban communities (39 percent) had fewer than 5 black households in 1970. Ten years later only 15 Chicago and 2 Cleveland suburbs had fewer than 5 black households. Similarly, between 1970 and 1980 the number of Chicago suburban communities with 10 to 49 black households more than doubled and the

number with 50 to 999 black households more than tripled. The number of Cleveland suburbs with 10 to 49 black households tripled, although the number with 50 to 999 increased only from 12 to 14.

Suburbs in the Bay Area also experienced impressive growth in black population, although smaller percentages of Bay Area suburbs had fewer than 5 or between 5 and 9 black households in 1970; combining these two categories, only 17.2 percent of Bay Area suburbs had fewer than 9 black households in 1970, compared to 72.4 percent for Chicago and 56.1 percent for Cleveland. These data support the view that racial segregation is less intense in the Bay Area than in Chicago and Cleveland and suggest that the movement of small numbers of black households to a large number of dispersed suburban communities, evident in Chicago and Cleveland during the period 1970–1980, may have begun a decade earlier in the Bay Area.

While the data in Table 9.4 point to an important qualitative change in black residence patterns, the quantitative impact of these changes remains small. In 1980, after a decade of rapid black suburban growth, 80.0, 73.6, and 67.4 percent of all black households in the three metropolitan areas still lived in central cities (Table 9.5). Most of the remaining black residents of the three areas—11.2 percent, 10.1 percent, and 13.5 percent of all SMSA black households—lived in suburban communities with 5,000 to 15,000 black households, either well-established suburban enclaves or "transitional" communities on the boundary of the central city ghetto. Suburban communities with 10 to 49 black households accounted for only 0.2, 0.3, and 0.2 percent of black households, and communities with 50 to 999 black households for 2.0, 2.2., and 8.5 percent of the total in the three areas.

Nearly all the black suburbanization in the Chicago and Cleveland SMSAs and in the Bay Area took the form of growth of established black suburban enclaves or peripheral expansion of the central city ghetto across central city boundaries. The number of suburbs with fewer than 5 black households declined substantially, however—from 92 to 17 for the three areas combined—and the number of suburbs with 10 to 999 black households increased from 87 to 146. The 1980 black populations of newly integrated suburban communities located at some distance from the central city ghettos were only a small fraction of the total 1980 black populations of the metropolitan areas studied, but the significance of the appearance of even token numbers of black households in so many, widely dispersed suburban communities cannot be overemphasized. The difference between none and several is enormous. Each of the thousands

of black residents of these communities in 1980 has dozens of black relatives, friends, and co-workers. Thus their success in overcoming the continuing discriminatory barriers provides information and encouragement for thousands or possibly hundreds of thousands of other black families, who even now may be considering a move to the suburbs. Further, the appearance of nontrivial numbers of black children in previously all-white suburban schools must have begun to change the attitudes and expectations of both blacks and whites about the likely effects of black entry into formerly all-white schools and neighborhoods.

While black entry into all-white communities adjacent to the ghetto may still signal a nearly certain transition to all-black occupancy, there are simply too few blacks to produce this outcome in the large number of widely dispersed communities that by 1980 had between 26 and 600 blacks. As both blacks and whites cease to believe that black entry necessarily leads to the creation of an all-black slum, this outcome will occur less often. Whites will continue to buy or rent housing in neighborhoods with black residents, and blacks will no longer feel they can obtain improved housing only in a few communities and neighborhoods that have come to be recognized as acceptable for black occupancy. The resulting changes in black and white expectations could produce significant decreases in racial residential segregation within a surprisingly short period.

The changes in black residence patterns discussed above represent the first hopeful signs of a solution to what until recently seemed an intractable problem. If the small gains suggested by the preceding analysis can be exploited, considerable progress could be made in solving one of the nation's most vexing and dangerous problems. The trends identified in this paper are rooted in long-term, and very likely irreversible, changes in black and white attitudes. Even so, the rate at which discriminatory practices are eradicated and blacks obtain access to previously closed segments of metropolitan housing markets will depend critically on government policy.

Policy Recommendations

It is impossible to determine how much of the newly emerging pattern of black suburbanization identified in this paper is attributable to the fair-housing legislation enacted during the past two decades and to federal and state courts' growing support, during most of this period, in defending the civil rights of black Americans. Still, it is clear that these efforts

have had an effect and, perhaps even more critical, that several administrations' support of civil rights has encouraged white Americans to moderate their attitudes and practices relating to racial integration. These conclusions open questions about the likely effects of the second Reagan administration.

Few would disagree with the finding that the Reagan administration has shown less support for fair housing and civil rights generally than the administrations that proceed it had demonstrated. While long-term decreases in white prejudice are likely to withstand even four more years of a Reagan administration, the extent of federal support for fair housing during the next four years could have a substantial impact on how much racial integration occurs in the next decade or two. Regardless of the outcome of the election or of the enthusiasm of the incoming administration for civil rights, efforts to devise policies to reduce housing market discrimination must be made with a clear understanding of the respective roles of income and racial discrimination.

In spite of clear evidence to the contrary, civil rights advocates and policymakers have continued to formulate policies based on the mistaken belief that the low income of black households is the primary cause of racial segregation and the serious underrepresentation of black households in suburban areas. Joel Selig, former Deputy Chief of the Justice Department's Civil Rights Division, argued that the Justice Department should concentrate on lawsuits challenging exclusionary land-use practices (1984, pp. 445–504); similarly, most legislators and policymakers would agree with the position set forth by the authors of a recent Yale Law Review note on racial steering:

While the supporters of the 1968 legislation understood that only a small number of ghetto dwellers would be able to afford property in suburban areas, they believed that the fair housing title was important as an expression of the nation's concern for the well-being of its black minority. (Notes 1976).

The conclusion that "only a small number of ghetto dwellers would be able to afford property in suburban areas" was and still is incorrect. The statistics for the city of Chicago and the Chicago suburbs (Table 9.6) once again demonstrate how little the level of black income has to do with the underrepresentation of black households in suburban areas. The six categories of suburb are defined by whether the area's median household income in 1980 was above or below the SMSA median, and by the area's percentage of blacks (over the SMSA average of 18 percent; between 10 and 18 percent; or less than 10 percent). Of Chicago's suburban areas, 21

had 1980 median incomes below the SMSA average and 101 had incomes above the average; the median incomes of all 122 exceeded the central city median. These same data show that only 7 areas (4 with above-average and 3 with below-average incomes) had a higher percentage of black households than the metropolitan area as a whole; 93 areas had above-average incomes and fewer than 8.1 percent black households.

The predicted distributions of black households in the 123 suburban areas, assuming residential location depends only on household income, are very different from the actual distributions. These estimates indicate that while blacks made up only 1.9 percent of all households in the 93 areas with above-average median income in 1970, they would have accounted for 14.8 percent of the households in these communities if household income had been the sole determinant of residential choice. Analyses published elsewhere show similar results for 12 sub-areas within the central city (Kain 1984).

The last three columns in Table 9.6 show that the percentage of metropolitan-area black households living in the city declined from 91.2 percent in 1970 to 85.1 percent in 1980. Even so, the number of black households living in the central city grew by 21.3 percent between 1970 and 1980, or 65 percent of SMSA growth in black households. The central city share of SMSA black households in 1980 is 33.6 percentage points greater than would be expected from a knowledge of household incomes alone. Finally, the percentage of all black households living in the 12 areas with below-average incomes and fewer than 8.9 percent black households would be 4.9 percentage points greater, and the share living in the 93 communities with above-average incomes and fewer than 8.1 percent black households would be 30.9 percentage points greater, if housing market discrimination had no effect on residence patterns.

The view that racial integration can only be achieved by forcing high-income communities to accept subsidized and other low-income housing (Selig 1984) is not supported by the data in Table 9.6. These and similar analyses suggest that priority should be given to measures that would strengthen enforcement of existing fair-housing laws and assist black households at all income levels to learn about and obtain housing in the nation's increasingly heterogeneous suburban areas. This policy emphasis happens to accord with the Reagan administration's reorientation of Justice Department enforcement efforts toward private discrimination, although not with its tendency to select low-visibility cases and to accept lenient settlements, or its decision not to use the effects test.

Appropriations to enforce fair-housing laws need to be increased substantially. The fiscal year 1979 budget provided only $17.4 million for fair housing altogether, compared to more than $300 million budgeted to support equal employment efforts. Appropriations for fair housing were further reduced by the Reagan administration. The highest priority should be given to providing the Justice Department with the staff and resources required to identify, document, and bring lawsuits against a much larger number of private and public actors. The department should be encouraged to use testers and other systematic methods of detecting discrimination and of insuring compliance with consent decrees and judgments. A major increase in Justice Department fair-housing litigation would encourage fair-housing groups to renew their efforts to achieve integrated residence patterns, strengthen the resolve of minority home-seekers to search for housing in predominantly white areas, and legitimize the efforts of private attorneys to bring Title VIII cases.

Recent court rulings providing for the recovery of attorney's fees and the growth in size of fair-housing judgments are beginning to attract the attention both of firms and individuals engaged in discriminatory practices, and of private attorneys. Greater activity by private attorneys in bringing fair-housing cases could make a major difference: a single Cleveland attorney, Avery Friedman, has brought more fair-housing cases in the last eight years than the Justice Department brought for the entire country. Friedman's cases involved private discrimination and typically resulted in small settlements, but he was instrumental in bringing an important pattern case against Ohio's largest residential broker (Heights Community Congress v. Hilltop Realty ) and he has obtained fair-housing awards of up to $50,000 for his private clients. Aided by the efforts of this private attorney and by those of the Cuyahoga Plan and other fair-housing groups, increasing numbers of Cleveland's minority households are learning of and demanding their rights to housing on a nondiscriminatory basis. While it is impossible to determine to what extent the appearance of growing numbers of black households in formerly all-white Cleveland suburbs are due to these efforts, there is good reason to believe they have indeed made a difference.

Awards in fair-housing cases have continued to increase in size; in July 1984, an upstate New York jury awarded a black plaintiff $545,000 in damages for discrimination encountered in attempting to rent an apartment (Grayson v. Rotondale and Son Realty ). This award is on appeal and is likely to be reduced somewhat, but the appeals court did recently

approve $120,625 in damages in another fair-housing case (Phillips v. Hunter Trails Community Association ), a reduction of an original award of $252,675.

The analyses presented above suggest that efforts to eliminate racial segregation should emphasize litigation and other measures to eliminate steering, discrimination in mortgage lending, and discriminatory marketing practices by developers and lenders. The Justice Department should also be encouraged to take vigorous action against local governments engaging in practices that limit black Americans' access to housing in their jurisdictions. In addition, a strong argument can be made for requiring, as a condition of receiving federal and state aid, that those suburban communities with many fewer black households than would be expected on the basis of their incomes and other characteristics be required to develop affirmative policies to attract minority residents. Local governments should be given wide latitude in developing programs and policies, but they should be held strictly to numerical goals. Because of the symbiotic character of residential and school segregation, moreover, communities with few black residents should be strongly encouraged to participate in voluntary busing, and, in assessments of their progress in reaching their occupancy goals, they should be credited for their participation in such programs.

With greatly increased staffing and strong support from top policy-makers, the affirmative marketing procedures instituted by HUD in February 1972 could make a major contribution to providing minorities with housing opportunities outside the ghetto. Every developer and sponsor participating in a HUD housing program must submit an affirmative marketing plan before the application can be approved. These plans require developers and sponsors to "carry out an affirmative program to attract buyers or tenants of all minority and majority groups" and to state in their plans the number of percentage of units they will sell or rent to minorities. These requirements undoubtedly assisted some minority households in obtaining rental housing outside of black neighborhoods. HUD's affirmative marketing requirement could be a powerful tool for achieving greater residential integration, but to date it has been little more than a paper requirement: HUD has had neither the staff nor the inclination to monitor these agreements.

Although it is true that subsidized housing accounts for a relatively small part of the nation's housing stock, governmental failure to administer these programs in a nondiscriminatory manner, as well as govern-

ment decisions about the siting of public housing projects, has contrib uted to segregated living patterns. When they design and administer subsidized housing programs, governments should consider the program's effect on racial segregation.

Housing vouchers, which are less visible and less disruptive than large projects, could be a powerful tool in achieving less segregated living patterns. Still in the absence of meaningful efforts to assist minority households in locating and obtaining units outside areas of minority concentration, vouchers are likely to reinforce existing patterns of segregation. It is essential that such programs embrace the entire metropolitan housing market and that local housing authorities (if they administer the programs) should be required to show that the programs reduce racial segregation.

While very little of the existing pattern of racial residential segregation can be attributed to income, several studies (Kain and Quigley 1975; Schafer and Ladd 1981) have shown that black households are much less likely to own their homes than whites of the same income and household characteristics, due to past restrictions on black residential choice and discrimination by mortgage lenders. Indeed, estimates (based on methods similar to those described previously for predicting black occupancy in the absence of housing market discrimination) indicate that more than 800,000 black households living in the nation's metropolitan areas have been deprived of the benefits of home ownership by housing market discrimination. These minority families did not obtain the inflation-induced increase in property values and wealth that benefited similarly situated white families during the past quarter-century. As a result, they and their children have much less wealth than similarly situated whites and are less able to trade up to better housing or to become homeowners. There is much to be said for a program that would recognize the effects of past housing-market discrimination on black homeownership and wealth and would assist larger numbers of black households in obtaining home-ownership. At minimum, strenuous efforts must be made to insure that blacks are not discriminated against in their attempts to buy suitable units outside the ghetto and that discriminatory lending practices are not employed to frustrate their efforts to obtain financing.

Appendix

References

American Civil Liberties Union. 1984. "In Contempt of Congress and the Courts: The Reagan Civil Rights Record." 27 February. Washington, D.C.: ACLU.

American Institute of Public Opinion. 1978. The Gallup Opinion Index. Princeton: American Institute of Public Opinion.

Citizens Commission on Civil Rights. 1983. A Decent Home. Washington, D.C: Citizens Commission on Civil Rights.

Farley, Reynolds, et al. 1978. "Chocolate City, Vanilla Suburbs: Will the Trend Towards Racially Separate Communities Continue?" Social Science Research 7:319-44.

Farley, Reynolds, Suzanne Bianchi, and Diane Colasanto. 1979. "Barriers to the Racial Integration of Neighborhoods in the Detroit Case." Annals of the American Academy of Political and Social Science 441:97-118.

Feins, Judith D., Rachael C. Bratt, and Robert Hollister. 1982. "Final Report of a Study of Racial Discrimination in the Boston Housing Market." Cambridge, Mass.: Abt Associates.

Friedman, Avery S. 1974. "Federal Fair Housing Practice." Practical Lawyer 20, no. 8 (December): 15-26.

Grier, George, and Eunice Grier. 1983. "Black Suburbanization in the 1970s: An Analysis of Census Results." Bethesda, Md.: Grier Partnership.

Hersberg, Theodore, et al. 1971. "A Tale of Three Cities: Blacks and Immigrants in Philadelphia, 1850-1880, 1930, and 1970." Annals of the American Academy of Political and Social Science 441:55-81.

Kain, John F. 1977. "Race, Ethnicity and Residential Location." Pp. 267-92 in Ronald E. Grieson, ed., Public and Urban Economics . Lexington, Mass.: D. C. Heath.

———. 1980. "National Urban Policy Paper on the Impacts of Housing Market Discrimination and Segregation on the Welfare of Minorities." Paper prepared for the Assistant Secretary for Community Planning and Development, U.S. Department of Housing and Urban Development, Cambridge, Mass. (mimeo.).

———.  1984. "Housing Market Discrimination and Black Suburbanization in the 1980's." Presented at Chicago Urban League Conference on Civil Rights in the Eighties: A Thirty-Year Perspective." Chicago.

Kain, John F., and John M. Quigley. 1972. "Housing Market Discrimination, Homeownership, and Savings Behavior." American Economic Review 62, no. 3:263-77.

———.  1975. Housing Markets and Racial Discrimination: Micro-economic Analysis . New York: National Bureau of Economic Research.

Kantrowitz, Nathan. 1979. "Racial and Ethnic Residential Segregation in Boston, 1830-1970." Annals of the American Academy of Political and Social Science 441:41-54.

Karlen, David H. 1968. "Racial Integration and Property Values in Chicago." Urban Economics Report 7, University of Chicago, April.

Lake, Robert W. 1981. The New Suburbanites: Race and Housing in the Suburbs. New Brunswick, N.J.: Center for Urban Policy Research.

Lieberson, Stanley. 1963. Ethnic Patterns in American Cities. New York: Free Press of Glencoe.

Miller, James Nathan. 1984. "Ronald Reagan and the Techniques of Deception." Atlantic Monthly 253, no. 2: 62-68.

Notes. 1976. "Racial Steering: The Real Estate Broker and Title VIII." Yale Law Journal 85: 808-825.

Pascal, A. H. 1967. "The Economics of Housing Segregation." Memorandum, RM-5510-RC. Santa Monica: RAND.

Pettigrew, Thomas. 1973. "Attitudes on Race and Housing: A Socio-Psychological View." Pp. 21-84 in A. M. Hawley and Vincent P. Rock, eds., Segregation in Residential Areas. Washington, D.C.: National Academy of Sciences.

Saltman, Julie. 1978. "Cleveland Heights: Housing Availability Survey, February-June 1978, Final Report." Cleveland Heights Community Congress (mimeo.).

Schafer, Robert, and Helen F. Ladd. 1981. Discrimination in Mortgage Lending. Cambridge, Mass.: MIT Press.

Schelling, Thomas. 1969. "Modes of Segregation." American Economic Review 59, no. 2: 169-85.

———.  1972. "A Process of Residential Segregation: Neighborhood Tipping." Pp. 157-84 in Anthony H. Pascal, ed., Racial Discrimination in Economic Life . Lexington, Mass.: D. C. Heath.

Schnare, Ann. 1977. "Residential Segregation by Race in U.S. Metropolitan Areas: An Analysis across Cities and over Time." Urban Institute Working Paper 246-2. Washington, D.C.: Urban Institute.

Selig, Joel L. 1984. "The Justice Department and Racially Exclusionary Municipal Practices: Creative Ventures in Fair Housing Act Enforcement." U.C. Davis Law Review 17: 445-504.

Sheatsley, Paul. 1966. "White Attitudes toward the Negro." Daedalus 95, 1: 217-38.

Taeuber, Karl E. 1968. "The Effect of Income Redistribution on Racial Residential Segregation." Urban Affairs Quarterly 4: 5-15.

Taeuber, Karl, and Alma Taeuber. 1964. "The Negro as an Immigrant Group." American Journal of Sociology 64: 374-82.

———.  1965. Negroes in Cities: Residential Segregation and Neighborhood Change. Chicago: Aldine.

Taylor, William L. 1984. "Farewell Civil Rights Commission." The Nation 238, no. 4: 113, 128-31.

U.S. Commission on Civil Rights. 1974. The Federal Civil Rights Enforcement Effort--1974. Volume 2. To Provide . . . For Fair Housing. Washington, D.C.: U.S. Commission on Civil Rights.

———.  1975. Twenty Years After Brown: Equal Opportunity in Housing. Washington, D.C.: U.S. Commission on Civil Rights.

———.  1979. The Federal Fair Housing Enforcement Effort. Washington, D.C.: U.S. Commission on Civil Rights.

Van Valey, Thomas L., Wade Clark Roof, and Jerome E. Wilcox. 1977. "Trends in Residential Segregation, 1960-70." American Journal of Sociology 82, 4: 826-44.

Wienk, Ronald, Clifford E. Reid, John C. Simonson, and Frederick J. Eggers. 1979. "Measuring Discrimination in American Housing Markets: The Housing Market Practices Survey." Paper prepared for the Department of Housing and Urban Development, Office of Policy Development and Research.

Yinger, John. 1978. "Racial Transition and Public Policy." John F. Kennedy School of Government Policy Note P-78-1. Cambridge, Mass.: Harvard University.

———.  1980. "On the Possibility of Achieving Racial Integration through Subsidized Housing." John F. Kennedy School of Government Policy Note P-80-2. Cambridge, Mass.: Harvard University.

———.  1982. "Evaluation of the Final Report of a Study of Racial Discrimination in the Boston Housing Market." Cambridge, Mass. (mimeo.).

Chapter Ten—
Commentary

Michael A. Goldberg

I am struck by the pivotal role the United States federal government plays in any proposed agenda for metropolitan America. Virtually overlooked are the states and the metropoles themselves, including the local governments of which they are comprised. An important, and remarkably widespread, theme concerns the economic efficiencies associated with such a federal role. There is broad agreement that the large federal presence has distorted the allocation of resources in such areas as housing, urban transportation, and environmental protection.

Federal involvement seems to have another frequent deleterious effect: pressure to meet perceived national and regional problems in the short run often leads to long-run inefficiencies. For example, in attempting to assure adequate supplies of residential mortgage funds, deposit-rate ceilings of various kinds were instituted. They in turn led to periodic and rapid flows of funds among financial institutions, creating the need for various liquidity-producing federal institutions such as Fannie Mae, Ginnie Mae, and Freddie Mac. In the end, a rather complicated web of institutions had to be created to handle the side-effects of well-intentioned federal residential mortgage policies. Similarly, the federal aid to urban transportation, while providing relief from traffic congestion and improving access for peripheral lands, had significant longer-run consequences for urban form and particularly for the economic and residential viability of the central city.

Above these specific themes lies perhaps the most striking aspect of all:

the assumption that federal intervention is a completely normal, accepted, and perhaps even preferred state of affairs.

I would like to address this assumption about the desirability of federal urban roles, sketching out in the process an alternative approach to metropolitan federalism and policymaking that has rather different results. I draw my examples from the Canadian context. Canada is in many ways an excellent source of alternatives since it, like the United States, is organized about federal principles, is large in land area, is a democracy based on English legal principles, is an advanced capitalist economy and, finally, is a next-door neighbor. These broad similarities make the differences I will touch on more compelling, and possibly more relevant.

Alternative Policies

Canadian Examples

The government of Canada has largely been excluded from direct involvement in metropolitan matters.^[1] The Canadian constitution assigns responsibility for cities and local institutions exclusively to the provinces. In 1977, there were a mere CDN$209 million of direct transfers from the federal government to local governments, compared with United States federal transfers to cities (excluding special-purpose district transfers, which were also excluded in the Canadian case) amounting to U.S. $11.4 billion in 1978, or more than five times as much on a proportional basis.^[2] Canadian federalism generally can be characterized as much more adversarial than its United States equivalent, with the Canadian provinces having a considerably stronger role than United States states.

Against this backdrop, I want to look at examples from two areas where Canadian policy has diverged markedly from that in the United States. The first relates to housing finance and the second to urban transportation.

The Canadian Housing Finance System

Canada's mortgage market provides a direct contrast to the situation that emerged in the United States during the 1960s and 1970s. At that time, United States federal involvement and attempts to pump money into the residential mortgage market set in motion a chain of distortions requiring a series of governmental and institutional responses to get liquidity and funds back into the financial sector. Canada followed a

rather simpler path that led to a reduction in federally induced distortions with a corresponding increase in both the stability and the magnitude of lendable mortgage funds.^[3]

Up through the 1960s Canada and the United States pursued similar strategies. Broadly speaking, the two federal governments sought to remove imperfections in the mortgage market; to make the mortgage instrument more attractive to lenders, by improving liquidity of mortgages and reducing default risk; and to stabilize supplies of mortgage funds available to homeowners and builders.^[4] These goals were achieved in Canada initially through the creation of the Dominion Housing Act in 1935, whereby the federal government got involved as a direct and joint lender and changed the mortgage instrument from the earlier balloon-payment mortgage to the present fully amortized mortgage instrument, while increasing lending ratios from the prevailing maximum of 60 percent to up to 80 percent and extending the term from a five-year period to ten years renewable for another ten years. Other federal legislation followed in 1938, 1944, and 1945, culminating in the National Housing Act (NHA) of 1954, which brought in mortgage insurance and enabled the government of Canada and its mortgage institution, the Central Mortgage and Housing Corporation (CMHC) to get out of direct and joint loans and instead insure qualified conventional lenders. Initially, CMHC set the mortgage rate at which it would insure loans. However, by the 1960s it became apparent that setting below-market rates was decreasing the amount of available mortgage credit. Therefore, in 1966 the NHA rate was set by formula as some floating percentage above the long-term Canada bond rate, and by mid-1969 the NHA was freed completely to be set by market forces. In 1967 the Bank Act was amended to allow banks to lend at interest rates above the previous 6 percent ceilings. This led to a massive increase in bank mortgage-lending activity, since the banks had largely been precluded from mortgage lending during the 1960s when mortgage rates rose above the permitted 6 percent level. The final major hurdles toward mortgage-market efficiency were cleared in 1969 when the Canada Interest Act and the National Housing Act were changed to allow borrowers to prepay any mortgage loan after five years, with a three-month interest-rate penalty. This effectively reduced minimum contract periods from 25–30 years to 5 years, although mortgages were still being amortized over periods ranging from 25 to 40 years.

These 1969 amendments were to prove particularly important during the inflationary period of the 1970s, since they created the five-year Canadian rollover mortgage whose interest rate was set entirely by mar-

ket forces. The five-year rollover made it possible for financial institutions to match the terms of their assets (the mortgages) with the terms of their liabilities (savings certificates), thus greatly reducing lender interest-rate risk. Since deposit rate ceilings did not exist in Canada, this meant that rates for both assets and liabilities could float freely in the marketplace, leaving the equilibration process unencumbered to allow and absorb price changes with relative stability. This served to increase the amount of funds available to borrowers.

Inflation in the 1970s brought new imperfections to light. First, the notorious tilt problem pushed lenders to seek alternative mortgage instruments. Second, the five-year rollover caused considerable difficulty for many borrowers whose mortgages came up for renewal during periods of rising interest rates. Moreover, as inflation increased, loan terms were progressively shortened—all the way down to three months, and, for variable-rate mortgages, to an effective weekly term. Borrowers were therefore being asked to bear interest-rate risk that they were unable to lay off elsewhere.

Recent federal policies have sought to remove these two imperfections. During the 1970s, CMHC experimented with a form of graduated payment loan and also, by insuring such loans, encouraged private lenders to use the GPM instrument. Finally, in 1984 CMHC initiated a mortgage-renewal insurance program to protect borrowers from adverse interest-rate shifts on renewal of mortgage loans.

The net result of these policies has been to remove government-induced distortions from the Canadian mortgage market and greatly to increase the efficiency of that market. Residential mortgage funds have been abundant and relatively stable and in turn have stabilized housing starts, while greatly increasing starts during the 1970s, a period when the United States housing and housing-finance industry suffered so critically from the volatility of interest rates.^[5] Additionally, Canadian mortgage-lending institutions experienced none of the liquidity and cash flow crises that plagued the American industry during this era.

My point in describing this experience in Canada is to put the United States situation in sharp relief by demonstrating that it is possible for a federal government to put into motion a series of policies over a three-decade period that reduce market imperfections in a rather straightforward manner, rather than producing the distortions and concatenation of secondary market institutions and instruments that have come to be taken as an unavoidable concomitant of the federal role in the United States.

Urban Transportation in Canada

Where the foregoing example pointed up a federal presence that reduced market distortions over time (my second theme), the case of Canadian urban transportation provides a striking contrast with that in the United States because the federal government is almost totally absent from the picture (my first theme). Urban transportation is a provincial, regional, and municipal function in Canada, and the lack of federal presence has not impeded either the planning or functioning of the urban transportation system. On the contrary, it could be suggested that the absence of a dominating federal presence results in a corresponding absence of the kinds of distortions and inefficiencies highlighted in Gomez-Ibañez's paper.

Not unexpectedly, the Canadian urban transportation system is quite

different from that in the United States. Much of this difference can, I think, be attributed to the very different behaviors of the federal governments in the two countries with respect to urban transportation. Tables 10.1 and 10.2 provide evidence of a Canadian urban transportation system that, relative to that in the United States, is much more reliant on public transit and correspondingly much less reliant on private automobiles and freeways. Accordingly, automobile ownership in Canada is roughly two-thirds and freeway lane miles per capita are less than one-quarter of what they are in the United States.^[6]

Given such differences in urban transportation and travel, we should expect Canadian cities to be denser and more compact than American cities. This turns out to be the case, as the data in Table 10.3 illustrate.^[7]

All of this suggests that it is possible to develop and maintain a perfectly efficient urban transportation system in the absence of large-scale federal spending. In many instances, the Canadian systems have been developed and operate without any significant provincial assistance either.^[8]

There is an alternative, then, to the continued massive American federal contribution to urban transportation, although it is not clear at this

point how the United States government can withdraw, given its past role and the urban form that emerged as a result of its role.^[9] My purpose here is not to suggest that the Canadian model is necessarily workable in an American setting but, rather, to demonstrate that successful urban transportation systems can be developed without a federal role. Indeed, there are some benefits of excluding large-scale federal intervention in transportation aside from the efficiency losses brought out in the study by Gomez-Ibañez. Specifically, the more compact urban form that resulted from the Canadian model has considerable advantages in terms of the health of the central city and the efficiency of providing services in the metropolitan region.^[10]

Conclusions

As economists, we have understandably viewed the agenda items for metropolitan America through "economic"-colored lenses. However, what comes out of these articles is a need for revised policies not just with respect to economic policy instruments, but more fundamentally with respect to the conduct of federalism itself in the United States. The problem of federal arrangements appears quite general in this context. Essentially, there are various levels of responsibility and concern. National defense, national transportation systems, and other issues of truly national scope fall within the purview of the federal government. Urban land-use controls and urban parks appear to be best dealt with on a local scale. So far so good. However, lying between these poles are hosts of issues that have both a national and a local component. The national

defense Interstate Highway System has had enormous effects on the localities through which it passes. Similarly, locally produced air and water pollution extend beyond the borders of local jurisdictions. In short, there is no unambiguous way to assign each responsibility to one level of government or the other.

This problem is repeated in federal-state interactions, state-local relationships, municipal-neighborhood dealings, and even in neighborhood-household and household-individual dualities. Instead of some optimal mix of higher-level and lower-level responsibilities, it appears that there is a dynamic balance that occurs in the needs and responsibilities of the various levels of government over time. I suggest that it may be time to reevaluate the balance that has emerged in the United States. Not only do federal-state-local relationships need to be reexamined, so do the interactions and dealings among the state-local, metropolitan-local, and state-metropolitan levels. The balance in the United States has tipped well over to the side of the federal government and far from the metropolitan regions and the local governments that comprise these regions. By presenting examples from Canada's federal system I have hoped to demonstrate that quite a different balance of federal-state-local powers and resources is possible, though the final mix that evolves in the United States would have to be congruent with American values, institutions, and needs.

Commentary

Theodore E. Keeler

Gomez-Ibañez's informative and insightful paper concludes that benefits from the federal urban highway program have exceeded costs; that benefits from investments in transit, especially fixed rail systems, have seldom equalled costs; that federal programs have probably siphoned considerably more funds away from urban highway users than they have returned in investments and that federal programs bias state and local highway investments toward capital-intensive programs which may not be justified; that federal transit programs have biased incentives toward large, capital-intensive rail systems which are difficult to justify economically; and that federal urban transportation programs overall have not acted in the interests of the areas concerned.

Regarding both the effects of the federal aid program and the undesirability of maintaining the transit portion of it, I am in complete agreement; indeed, I believe there is evidence not mentioned in his paper which strengthens his case.

Regarding the benefits versus costs of urban highways, I believe there is evidence that benefits are even higher than is stated. I have conducted the accounting exercise of comparing highway revenues and expenditures (Keeler, Cluff, and Small 1974). One problem is that many urban roads are financed through subdividers and never show up in the government expenditure accounts. Some additional evidence on willingness to pay is provided by the effects of OPEC and of federal regulation of automobile safety and emissions standards over the past few years. Both have increased the cost of owning and operating a car by much more than anyone

has seriously contemplated through increases in highway user charges. Yet auto vehicle-miles traveled have increased strongly and steadily over the past decade. Another indication of very high willingness to pay for the benefits of the private auto in urban transportation comes from travel-demand models in an indirect way. Most urban travel-demand models assume that auto users do not perceive the indirect costs of using their cars (e.g., depreciation, interest, insurance). Yet at least some of these costs are indeed variable with use. When Kenneth Small (1976) included all costs of owning and operating a car in his travel-demand equation, he found extremely high values of travel time—in the neighborhood of $40 per hour for walking and waiting time. What he seems to have identified, once again, is a very high willingness to pay for the benefits of the private automobile, and the evidence implies that whether or not people perceive these costs, it is quite clear that they are willing to pay them. Finally, Timothy Hau (1981) made a detailed benefit-cost analysis of the I-580 corridor in the San Francisco Bay Area, and found that for some bottlenecks of that corridor, additional lanes of highways for auto use could pay for themselves in one year alone, with no discounting. There is, thus, mounting evidence that highways are underbuilt in many metropolitan areas of the United States, and that the highways that exist should not be allowed to deteriorate.

I agree completely with the strong evidence presented by Gomez-Ibañez on the costs and virtues of rail transit versus highway transit and the private auto. But there is also some contradictory evidence; the truth may be more ambiguous than this article implies. Federal programs do divert money from urban systems and do spend it on rural systems; to use an economic term common in the study of regulated industries, the federal highway program cross-subsidizes rural roads with revenues from urban roads. I am inclined to agree that this is a bad thing, but more analysis and evidence are necessary before strong conclusions are possible. We need to go back to subdivider-built urban streets (maintained by local property taxes). These roads do not get federal (or, in many cases, state) money. But they do generate fuel-tax revenues, revenues which are used to cross-subsidize other roads. Thus, to the extent that revenues from users cover expenses for these roads, because they are supported by local property owners, these revenues are siphoned off to pay for rural and suburban roads.

Does this make cross-subsidization bad? There are good reasons to believe that it is bad in this context—it represents a possible misallocation of resources. And the most commonly asserted argument against cross-

subsidization, "every tub should stand on its own bottom," is based on a well-founded fear that cross-subsidization will misallocate resources to areas in which benefits do not equal costs. Nevertheless, economic-efficiency arguments can be made for cross-subsidization from urban to rural roads. For example, Walters (1968) suggests that urban roads are subject to constant or decreasing returns to scale, whereas rural roads, because of their indivisibilities and low utilization rates, are subject to increasing returns to scale. Under these circumstances, an efficiency-maximizing pricing scheme could very well generate a surplus from urban roads and use it to subsidize rural roads. It is rather unlikely that the present pricing and investment scheme represents an efficient (or even second- or third-best) policy. Nevertheless, it could be superior in efficiency to a program that did not involve such cross-subsidization.

In terms of distribution, it is unclear whether such cross-subsidization is good or bad. As Meyer, Kain, and Wohl (1965) pointed out, the residents of cities do benefit from the superior intercity transportation generated by improved rural highways. If, as Gomez-Ibañez argues, however, main intercity roads generate positive net revenues, then most of the cross-subsidization goes instead to low-density rural roads whose primary function is local service to rural residents. In that case, the incomes of urban dwellers suffer.

In short, then, although I agree that cross-subsidization is a mistake, I believe that more research is needed on the efficiency of such federal programs.

References

Hau, T.D. 1981. "Cost-Benefit Analysis of Urban Highway Pricing and Investment: An Explicit Expenditure Function Approach." ITS Dissertation Working Paper Series, UCB-ITS-DS81-3. Berkeley: Institute of Transportation Studies, University of California.

Keeler, T. E., G. S. Cluff, and K. A. Small. 1974. "On the Average Costs of Automobile Transportation in the San Francisco Bay Area." Berkeley: Institute of Urban and Regional Development, University of California (processed).

Meyer, J. R., J. F. Kain, and M. Wohl. 1965. The Urban Transportation Problem. Cambridge, Mass.: Harvard.

Small, K. A. 1976. "Bus Priority, Differential Pricing, and Investment in Urban Highways." Ph.D. dissertation, University of California, Berkeley.

Walters, A. A. 1968. The Economics of Road User Charges. Baltimore: Johns Hopkins.

Commentary

Richard F. Muth

It has always been a matter of some anguish to me that I have had to disagree so often with my good friends John Kain and Sherman Maisel. I am therefore delighted to report that there is much in their papers with which I agree. Kain's paper makes many important points about the residential segregation of blacks. First, the segregation of blacks is surprisingly persistent in many cities, despite the progress that has been made during the past two decades in other areas of racial relations in this country. Moreover, to the extent that suburbanization of blacks has occurred it has all too frequently been nothing more than an expansion of central-city black areas beyond the central city's boundaries. Surburban blacks, in my experience, are as highly segregated from whites as their central-city brothers are.

Kain is quite correct, I believe, in rejecting the hypothesis that black segregation is the result of black preferences. What little market data I have seen indicate that blacks pay higher prices for comparable housing in areas adjacent to white residences than in areas remote from them. This suggests to me that blacks prefer integration and are willing to pay for it. Nor can one account fully for black segregation by the lower incomes of blacks. Segregated areas of low- and of high-income whites often exist along with similar black areas in United States cities. Moreover, studies that have controlled for income differences find that segregation indices are only somewhat reduced by doing so. In any case, income differences can hardly account for the concentration of blacks in central cities, for while average suburban incomes do exceed average

central-city incomes, the differences are small. In 1979, for example, the average income of white central-city households was $20,103, that of white suburban households $23,720—a difference of only about 16.5 percent.^[1]

Kain is, however, mistaken, I think, in arguing that white preferences cannot account for black segregation. He bases his conclusion both on the wrong question and on the wrong kind of data. The appropriate question, it seems to me, is whether a white family would move into a dwelling in a black area, not whether it would move out if a black family moved next door. If whites will not move to an area where blacks live, the area will gradually become black.

Moreover, I tend to be distrustful of attitudinal surveys. I would far rather consider what people have actually done when confronted with an event than their response to a "what if" situation. Casual observation over the years supports the conclusion that few whites move into largely black areas. In addition, the few empirical studies of which I am aware find that whites pay lower prices for comparable housing when adjacent to blacks than when their housing is remote from black areas. This is precisely what one would expect if whites preferred segregation from blacks.

This last point is important because of its implications for Kain's policy recommendations. One fault I find with his principal recommendation is that Kain nowhere provides an explanation for the residential segregation of blacks. He appears to believe that it results primarily from the actions of real estate agents, renting agents, lenders, and others. I find it convenient to characterize his implicit explanation as a villain hypothesis, though Kain does not use this term. Nowhere, however, does he give us any explanation for the villains' behavior. It is possible to explain this behavior as being in the villains' economic interest. If a renting agent, for example, were to have no personal aversion to dealing with blacks or preference for dealing with whites, he might well turn away potential black renters, acting on the well-founded belief that at least some white tenants would move out if a black were to move in. Until the agent could replace those departing with new tenants, a higher vacancy level and thus lower net income would result. Moreover, if whites prefer to live with other whites, it would be difficult to replace white tenants who departed for whatever reason. The building or development would thus tend to become black-occupied.

If my explanation for villains' behavior is correct, then stronger enforcement of so-called fair-housing laws would have little long-run effect.

To be sure, by making the penalties great enough one could certainly change the behavior of rental agents and other villains. One could perhaps even devise incentives to make it less likely that white tenants would move out of a building into which a black family had moved, though I doubt such would prove either practical or politically feasible. I see absolutely no way through fair-housing legislation, however, to insure that whites would rent or buy in largely black residential areas. If I am correct, the residential segregation of blacks arises because blacks will offer less of a premium to live in the vicinity of whites than will other whites. To eliminate black segregation, an increase in the amounts blacks would offer relative to white offers for living among whites would be required. This could be accomplished by subsidizing blacks to live in white areas, by subsidizing whites to live in black areas, or by taxing whites who live in white areas. Of the three, subsidizing blacks might be most acceptable politically.

Certainly, there are strong reasons for preferring integration to residential segregation of whites and blacks. It is much less clear to me what is to be gained from the suburbanization of blacks per se. Casual observation, the small white income differential cited above, and the only study of which I am aware^[2] all suggest that suburban blacks are no better off economically because of their location. It is true, of course, that some central cities are now so largely black that some blacks would have to be suburbanized in order to achieve integration. Suburbanization would also tend to divide between central-city and suburban governments the fiscal burden associated with the low incomes of blacks. The latter problem might be handled even more easily, however, by collecting taxes for income redistribution at a higher level of government. Thus, I can see little reason for black suburbanization as distinct from white/black integration.

Maisel's paper, like Kain's, is highly critical of the practices of the Reagan administration. Examined carefully, however, it is not the actions of the administration so much as the recommendations of the President's Commission on Housing that Maisel opposes. It is important to distinguish between the two, I think. The administration has indeed succeeded in cutting back sharply on new production programs, as the commission recommended. The voucher program recommended by the commission, however, is currently no more than a fifteen-thousand annual unit pilot program. The Secretary of HUD, moreover, publicly disagreed with the commission's recommendation for a partial federal preemption of local rent control laws. Interestingly enough, the 1982 federal preemption of

so-called "due-on-sale" legislation and court decisions was based upon the same rationale as the rent control recommendation.

One of Maisel's principal justifications for government support of housing is that it has become less affordable in recent years. The data he cites, however, are based on mortgage payments. It is now well understood that these are highly misleading as measures of the costs of occupying a house. Even the Bureau of Labor Statistics has shifted to a rental-value measure in its consumer price index. Because of capital gains, the implicit rental value of homeownership fell sharply in the 1970s. Moreover, the income distribution of new house buyers changed very little during the 1970s and their median age actually fell.^[3] It was not until late 1979 that the implicit rental value of homeownership rose sharply, and then the rise resulted from the sharp increase in real interest rates in the economy generally and had little to do with housing per se.

I find Maisel's insistence that a housing program necessarily implies the need for government production to be curious indeed. After all, we do not require the recipients of food stamps to spend them in government grocery stores. Nor do we limit treatment received under Medicare and Medicaid to Veterans Administration or Public Health Service hospitals. Public production might be indicated if increased housing demand from lower-income families would lead to higher housing prices, but evidence from the housing allowance supply experiment refutes this supposition. Likewise, virtually every evaluation concludes that housing produced under public programs has been considerably more costly than comparable housing available on the private market. It therefore seems that more low-income families may be assisted at a given resource cost through voucher programs than through public production.

Regarding housing vouchers, Maisel asks—I presume rhetorically—whether money spent on vouchers might be better spent on other programs. My answer would be, yes and no. As a general rule, income transfers are preferable to in-kind subsidies, and I can think of no compelling reasons why housing is an exception. Most low-income families today live in standard housing or in housing that can be brought up to standard at relatively little cost. Their principal problem, now as it has always been, is low income. Housing vouchers provide them additional funds for the purchase of housing. Moreover, under the proposed voucher program, low-income families need not move to receive assistance if their current dwelling meets program standards. The voucher payment then acts as a straight income transfer. Since I prefer income

transfers to in-kind subsidies, I am delighted to call an income transfer a housing program if that is necessary to gain political acceptance.

I fully agree with Maisel's criticism of the President's Commission's stand on the tax issues related to housing. Indeed, I did everything I could as a member of the commission to change, and failing this to moderate, its position on these questions. It was anticipated that savings and loan associations would reduce their mortgage lending as they made use of their newly broadened lending powers. The mortgage-interest tax credit was thus proposed as a means of inducing other lenders to make additional mortgage loans to compensate for the reduction. In my judgment, however, such an inducement is not needed. Mortgage yields will rise and the returns to other kinds of lending fall as savings and loans associations redirect some of their lending. Others will then find mortgage lending relatively more profitable and increase theirs. The principal effect of the mortgage-interest tax credit would be to reduce the yields on mortgages relative to those on other securities. While Maisel, judging from his remarks on federal credit programs, might find this effect attractive, I do not. In my judgment, we have already reduced the cost of homeownership too far relative to the prices of other commodities, principally through the tax treatment of income from owner-occupied housing.

The commission's stand on this last issue is Maisel's strongest indictment against it, and rightly so. He questions, though less strongly than I would, the propriety of calling this treatment a tax expenditure. Given the level of other expenditures, the subsidy to homeowners is clearly paid by taxes collected from renters. Members of the former group living in SMSAs had average incomes of $26,229 in 1979, while similarly located renters' incomes averaged only $15,114.^[4] Not only do lower-income families subsidize higher-income ones, but the size of the subsidy grows proportionately with the homeowner's income and thus his marginal tax rate. I find it impossible to conceive of a subsidy/tax scheme that operates in a more perverse manner than the treatment of income from owner-occupied housing.

Given his recognition of the nature of the tax treatment of homeowners, I find it difficult to understand Maisel's denial that our economy has overinvested in housing. His comparison of housing built recently with that in the 1950s is surely flawed because of the postwar housing shortage. Most economists appear to believe that the price elasticity of housing demand is at least .75, and I believe it to be unity. At marginal tax rates of 20 percent, the tax treatment reduces the cost of a dollar's worth of

owner-occupied housing to about 85 cents, since interest costs and property taxes are about three-quarters of housing costs. Thus, because of the subsidy homeowners consume about 12 to 15 percent more housing than they otherwise would. And, as several writers have observed, the extent of the subsidy increases with inflation. It should not be surprising, then, that we built housing at about a 20 percent higher cost in the late 1970s than we did in 1965.

It is by no means clear to me that it is desirable to subsidize homeownership. In any case, the tax treatment of homeownership is a highly inefficient subsidy. Most studies suggest that tax treatment increases the extent of homeownership only on the order of 5 percentage points, or from about 60 to 65 percent. At the subsidy rate of 1982, $50 billion in aggregate, it thus cost about $15,000 annually per additional homeowner. Surely measures that are more cost-effective and less objectionable on distributional grounds could be found to stimulate homeownership.

Although I agree with the substance of Maisel's criticism of the commission and the administration on this score, his criticism strikes me as unfair. To my knowledge, no other administration or presidential commission has spoken out against it. Indeed, not long ago the Senate passed a resolution favoring the current system by something like a vote of 85 to 0. The only surprising thing about this vote to me was that those few senators not voting for it did not manage to have their votes recorded as favoring it. Nothing presents a more challenging political problem than doing away with a program that benefits two-thirds of the voters.

Yet I see an economically sensible and politically possible way to change matters. It is, to use that hideous Washington verb, "to grandfather" existing dwellings when requiring homeowners to report, for tax purposes, income received from their house. The great political hurdle to removing the current tax treatment is the fact that doing so would result in capital losses for two-thirds of the electorate. If it were not required to report implicit income for dwellings in existence at the time of the change, these dwellings would be worth more than new ones. The cost of the latter is fixed by conditions of supply, however. The owners of existing dwellings would thus not suffer capital losses, whether they continued to live in them or not.

Economically, such a proposal makes perfect sense. Regardless of tax changes, we cannot get the excess materials and labor expended on existing dwellings out of them and into other uses. The best we can hope to accomplish is to prevent excess investment in housing in the future. Moreover, since buyers would be willing to offer more for a tax-favored

dwelling than for a new one, its owner would be forced to consider its opportunity cost to society when deciding whether to remain in it. It seems to me that there are few tasks an economist might take up that would be better for our country or for our profession than to work for the repeal of the current tax treatment of owner-occupied housing.

Commentary

Melvin M. Webber

Gomez-Ibañez has presented a precise analysis of a transport sector's behavior and of the policy implications to be inferred. Despite its analytic elegance, the paper is flawed for having been cast in the antiquated mold of the auto-transit dualism. There is of course a long and distinguished pedigree that lends the credence of tradition to that dualism, but it is no longer a useful distinction for contemporary urban transportation analysis or policy.

Transit vehicles come in many shapes and sizes nowadays, and they support increasingly diverse service modes. Rubber-tired types now range in size from hundred-passenger monsters to vans that carry fewer than a dozen to automobiles with only four seats. Each travels on city streets and highways; each carries paid passengers; each is free-wheeling and able to change route flexibly. Even though not all are commonly understood to be transit vehicles, they all perform public transit services. Vehicles that follow fixed guideways also come in many shapes. There are old-fashioned suburban railroad trains like the Long Island Railroad, newfashioned suburban railroad trains like BART, trolley cars of various vintages and fashions from the old South Line interurban in Indiana to the new Boeing cars in Boston, so-called people-movers like those at the Dallas–Fort Worth airport and Morgantown, and numerous others on the drawing boards. That heterogeneity and those automobiles used as public transit vehicles have vitiated the transit-auto dichotomy. The tenable and useful distinctions are the operating characteristics of various system types. And then, if there are differences among vehicle types that

are now germane, they are the ones between small-vehicle systems and large-vehicle systems.

The strangest feature of contemporary urban-transit policy is that the flood of investment in transit systems is pouring into large-vehicle systems with operating characteristics fitted to batch passenger loads rather than small-vehicle system suited to steady passenger flows. It is especially strange because most of the planned investment is targeted to the large metropolitan areas of the United States West and South, settlements that are new, built largely since World War II, and built at low residential and employment densities. The counterpart of low density in these places is spatial dispersion of homes and jobs, hence relatively small numbers of travelers making trips having the same origins, destinations, and schedules. A further counterpart is a ubiquitous roadway network, such that all locations are linked directly to all others. When fitted out with sufficient vehicles, that network provides the connectivity that has helped transform once-tranquil urban settlements into some of the most productive centers of economic and social activity.

The waves of modal-choice studies of the past two decades have by now made it unquestionably clear why automobiles have become so popular in these places. Travelers select travel modes with an eye to reducing the combined dollar and time costs of travel. Those modes that come closest to offering door-to-door, no-wait, no-transfer service at tolerable monetary outlay are preferred over those that may be more comfortable, prestigious, or in other ways seemingly more attractive or more efficient. It is now clear that automobiles have become the dominant transport mode, especially in suburban and low-density areas, precisely because they most closely match both travelers' economizing criteria and the spatial dispersion of origins and destinations. Buses, especially small buses, come close to meeting these tests within some locations, but the automobile clearly meets them best. Together, and in combination with the ubiquitous road network, cars and buses provide random access—direct connection from anywhere to anywhere.

In this respect, the auto-highway system is analogous to the telephone system, which connects virtually all places directly to all other places, assuring random access. The decentralized locational structure of the modern metropolis effectively requires unconstrained connectivity for single travelers and single callers, wherever they may be located.

The contemporary metropolis of the American West and South differs structurally from nineteenth- and early twentieth-century cities of the Eastern seaboard and Europe. In those older settlements, large concentra-

tions of houses and jobs permit large-batch transit of travelers who move on identical routes and schedules. Those older cities built the successful subway and suburban rail systems that are tied in close symbiosis to their concentrated business districts. They are also the apparent sources of recent crusades to build large-vehicle systems in the South and West. Civic boosters in the new metropolises have heard that subways made the older cities great. Flushed with civic pride and expecting boosts in their local real-estate markets and their local economies, they now want subways too. With encouragement from the planning-and-engineering consultants and the generosity of their congressional delegations, the boosters have by now succeeded in acquiring billions of dollars earmarked for underground rail construction. It matters little whether the subways will accomplish the urban development and transportation objectives that are claimed for them, for these projects have themselves become the ends in view. They are no longer merely transportation-improvement projects; they are monuments fostering civic pride, pawns in complex political games, and sources of considerable monetary gain for some.

If, instead, the real objective were to improve accessibility and mobility, quite different strategies would be appropriate. Improvements to public transit service are mandatory components of such an urban-transportation-improvement strategy. For most metropolitan areas the search for transit-service improvement must inevitable lead to small-vehicle systems, specifically to those whose operating characteristics match the spatial dispersion of homes and jobs and hence spatially and temporally dispersed travel patterns.

Those transit system types will surely include buses of various kinds and sizes and automobiles used in public transit modes—in taxi, jitney, and ride-sharing arrangements. Dramatic reductions in traffic congestion should be readily attainable if car occupancy were to be increased. It is evident that much can be done at low capital cost, simply by offering motorists incentives to share their cars with others, for a vast transport capacity in America is going unused. There are enough front seats in the present automobile fleet to carry the entire population at the same time, but of course most of those seats move about empty. Where incentives are great enough, as they are on the San Francisco Bay Bridge, car-occupancy rates can be hiked appreciably. During the morning westbound peak, cars across the bridge are now carrying an average of slightly more than two persons each, owing at least in part to incentives offered commuters. Motorists can save up to 15 minutes and the 75-cent toll if they carry two passengers, and so they collect two or more strangers at bus stops and

BART stations in a win-win arrangement that benefits all parties, including the solo motorist who experiences lessened congestion.

Small vehicles, such as automobiles and vans, have a large latent utility as public transit and quasi-transit vehicles. The distinctions that cry out for analysis now are multidimensional, touching on the varied capabilities of transport systems operationally to match different land-use patterns and different travelers' assessments of costs. The old dualism between auto and transit will no longer serve. It would have been more informative to have read an analysis by Gomez-Ibañez of the consequences stemming from operating characteristics of various large and small vehicle systems, rather than more on the auto-transit debate.