This new debate is obscured by the closely related and older debate between proponents of supply subsidies, that is, government subsidies to build housing or rehabilitate rundown housing, and the proponents of demand subsidies, which use government funds to provide housing allowances for renters and mortgage credits or downpayment assistance for homeowners.1 Housing allowances are the same in each case: they use the market and provide demand subsidies. The provision of social housing and the use of supply subsidies often overlap, as when funding is provided to a nonprofit corporation to build new housing or to rehabilitate a vacant property for low-income people, but they are not the same.
Housing assistance programs for low-income people have gone through three phases, and social housing advocates aspire to creation of a fourth. The first federal housing-assistance programs for low-income people financed the construction of public housing. These were both supply subsidies for construction of new housing and a means of creating housing that was outside the market, in this case owned by housing authorities of local government. After construction of about 1.3 million units of public housing, the total began to decline. A substantial part of public housing was in large developments occupied entirely by the very poor, and the management problems overwhelmed many big city housing authorities.
In the 1960s, federal assistance relied more on the private market, giving construction subsidies for about 1.7 million units of housing to be owned and operated by private, for-profit landlords, and for about 0.2 million units to be owned by private, non-profit organizations. The private, for-profit owners of subsidized housing were required to rent the new buildings to low-income tenants at affordable rents, but were allowed to opt out of the program after twenty years. As a result, beginning in the late 1980s, in areas where market rents had increased substantially the owners opted out and raised rents to market levels, displacing thousands of poor tenants. Where conditions had declined, owners simply undermaintained their property for additional, final profits and then allowed HUD to foreclose on deteriorated buildings.
In the 1980s, Federal policy shifted decisively away from supply subsidies of all kinds and towards the use of housing allowances (Low Income Housing Information Service 1994). By 1995 the number of Section 8 certificates and vouchers had doubled, to about 1.5 million. Improvements in the overall quality of the housing supply in the United States over the previous forty years made it plausible to argue that the housing problems of the poor were primarily problems of affordability and of constraints that prevented the market from responding to higher rents by producing more units. Subsidized new construction clearly cost more than simply providing housing allowances (Bradbury & Downs 1981; Dolbeare 1983, 30; Downs 1991). Supply-oriented funding persisted mostly as one of many uses of federal block grant funds and in the Low Income Housing Tax Credit, established in 1986. In 1995 the U.S. Department of Housing and Urban Development announced its intention to shift public housing operating subsidies into housing allowances, giving every public housing tenant the option either to stay or to use the certificate or voucher in the private market, a step likely to force local housing authorities to close down much of the least desirable public housing.
The triumph of housing allowances and market-based subsidies seems virtually complete at the national level; yet, simultaneously, a dramatically different policy approach has developed at the local level. The national disaster of expiring rent restrictions helped make clear the importance of alternative forms of housing ownership for providing affordable housing to low-income people, since only the small stock of subsidized housing that was owned by private nonprofit organizations remained a secure supply. Responding to the threat of extensive displacement, advocates of low-income housing worked to help nonprofit community corporations and limited-equity tenant cooperatives buy buildings whose rent restrictions were expiring and convert them to nonprofit ownership.2
The effort to preserve the affordability of "expiring use" properties by converting such for-profit housing to nonprofit ownership demonstrated the point that creation of social housing is an analytically distinct policy from use of supply subsidies, even though they may often be used together. We should, then, move beyond the debate about supply subsidy versus demand subsidy, to analyze the merits of housing allowances and social housing as alternative means of delivering housing affordability to low-income people. In the following pages I make a start on this analysis by looking at the two approaches, not from the usual policy-analysis perspective of an omniscient observer of national policy, but a viewpoint on decision-making by local government. This perspective is particularly appropriate for understanding a policy whose successes have largely come at the local level.
Social housing uses a combination of nonprofit ownership and capital grants to provide permanently affordable housing allocated on the basis of need rather than ability to pay. The defining economic mechanism of all project-based assistance, including social housing, is the use of capital grants for construction or purchase of housing in order to reduce monthly rents or ownership charges.3 This result is possible because housing is a particularly capital-intensive good, requiring a large initial investment in construction or purchase as well as the ongoing operating and maintenance costs.
The rent that tenants pay to a landlord has two conceptually distinct components: first, the cost of operating and maintaining the building, usually 40 to 60 percent of the rent, and second, the net operating income (NOI), the remaining 60 to 40 percent of the rent.4 The purchase price of for-profit housing is based on the capitalized value of its future net operating income, so conversion to social ownership requires the nonprofit to buy out this income stream. Project-based subsidies in the form of capital grants eliminate the need to provide profits for investors or repay lenders of money used to construct or purchase the property. This allows the nonprofit owner who receives the subsidy to reduce the net operating income component of the rent so that residents need pay only the amount necessary to operate and maintain the property. (A prudent social housing organization will not entirely eliminate the NOI, but rather set rents at a level that maintains a fund for future major repairs or renovations.) Housing allowance programs are tenant-based subsidies that simply pay toward both parts of the market rent.
Social housing requires both the initial capital investment and a form of ownership that can keep housing costs permanently below market and will take the return on invested capital in the form of reduced rents allocated to people according to need, rather than as profits from market rents. Ownership vehicles commonly used for creation of social housing include nonprofit housing corporations, mutual housing associations, limited-equity cooperatives, and land trusts. Nonprofit housing organizations are private, charitable corporations run by a governing board that may or may not include residents. Mutual housing associations combine private, nonprofit ownership with extensive tenant participation in managing their buildings and tenant membership on the governing board. Limited-equity stock cooperatives are a form of home ownership in which residents jointly own and operate the property but cannot profit from increases in the market value of the property when they move. Land trusts use a private foundation to maintain permanent ownership of the land, in order to separate ownership and control of the building in which residents live from ownership of the land and the associated ability to profit from increased land values.
I will make the following points of comparison between the two model programs: first, tenant mobility and economic integration; second, the number of people assisted over time, looking at both the short and the long run; third, administrative complexity; fourth, program stability over time; and fifth, political support.
Tenant Mobility and Economic Integration. The first point of comparison concerns tenant mobility and economic integration. Tenant assistance through housing allowances allows tenants to choose where they want to live and to move when they want to. As the size of an assisted family changes, for example, the tenants can easily move into a unit of a different size. More importantly, with expanded ability to rent the unit of her or his choice, the tenant can choose among neighborhoods and thus escape areas of concentrated poverty and low economic opportunity.
HUD Secretary Henry Cisneros has stressed the importance of tenant mobility and its links to economic mobility, in continuing the shift in housing programs from project-based to tenant-based assistance. This policy is based primarily on the experience with the Gautraux Assisted Housing Program in the Chicago area, which has shown that mobility can yield important benefits when recipients of housing assistance are given intensive counseling and social service support in moving to and dealing with a new environment (U.S.HUD 1994; Rosenbaum 1995; Williams 1995).
There are serious limits to the economic integration of low-income tenants through housing allowances. Often tenants are restricted by economic or racial discrimination. Since many suburban areas have pervasive restrictions on construction of multi-family housing, rentals in these suburbs are limited. Some low-income people cannot afford to own a car or are unable to drive. They depend on transit, so they cannot move into suburban areas where a car is necessary. Others do not want to move away from family members on whom they depend for help in caring for their children. The housing allowance experiments showed that voucher programs had virtually no effect on local mobility (Rossi 1981; Varady 1982; Lowry 1983, 350). Landlords renting to tenants in the Section 8 housing assistance programs tend to constitute a limited submarket, and it is difficult for aid recipients to move outside it (Finkel and Kennedy 1992). It is easier for recipients of housing allowances to move only within low-income areas, where they face the least resistance and can obtain the most housing for the allowable rent subsidy. The Berkeley Housing Authority reports that although nearly one-third of its Section 8 recipients have moved to other cities, most go only to neighboring Oakland, a poorer city where it is easier to rent larger two- and three-bedroom apartments or houses.
Creation of a local housing allowance program to supplement the Federal Section 8 program in a smaller city like Berkeley, with its population of 105,000, raises additional issues. If the housing allowance can be used only in the home city, it greatly limits mobility. Yet if the program allows or even encourages tenants to move out of town, it will be controversial: some critics will ask why people who move to other communities should be assisted with local funds, and others will oppose the program as a strategy for removing the poor.
Social housing allows the residents mobility only after a very extensive pool of nonprofit housing has been created, and assists in deconcentration only if this housing is geographically dispersed. Political and economic pressures direct most social housing into already low-income areas. Political resistance to low-income housing is strongest in higher-income areas, the exception being housing for the elderly. In addition, housing is typically less expensive in low-income neighborhoods, so that projects appear more cost-effective there. Social housing is also more likely to gain political support when rehabilitation or new construction will help upgrade a neighborhood showing substantial deterioration. Furthermore, acquisition of buildings in predominantly low-income neighborhoods will immediately help the low-income tenants who live in them, but acquisition of middle-income apartment buildings can only help low-income tenants as current tenants move out. At best, in some cities like Berkeley where low-income neighborhoods are gentrifying, purchase of housing in the present lower-income neighborhoods can be considered buying into increasingly mixed-income neighborhoods to ensure that low-income people can remain there.
As a tool for increasing economic integration, housing allowances seem to work better than social housing does mainly in those larger metropolitan areas where rental housing is available in economically diverse areas with access to employment opportunities. As local programs operated within one small city, however, each type of program faces barriers to achieving more economic integration that are substantial, though not always insuperable, and neither has a clear advantage. The Berkeley experience has included encounters with rising neighborhood resistance to subsidized housing projects; such resistance generates severe political difficulties and increases program costs. On the other hand, the efforts of a single low-income, tenant household with a housing allowance to find a landlord who will rent them one unit in a mixed or higher-income area may pass unnoticed—though often also unrewarded. The mobility programs that provide the assistance, however, present a larger target, and have faced substantial resistance in some metropolitan areas (Moberg 1995). A program within one city that tries to direct poor tenants into higher-income neighborhoods is an even more likely target of political attack, since neighborhood residents can influence their city government directly.
Cities are less able than is the Federal government to ignore declining areas, and are thus concerned with the economic integration of whole neighborhoods occupied mostly by the poor. Social ownership is virtually the only way to keep housing viable in neighborhoods undergoing drastic disinvestment and abandonment by the private sector; moreover, social housing organizations typically strengthen the organizational skills of the residents, promoting community self-help and mutual aid as well as advocacy for a more equitable distribution of municipal resources (Barton 1977). The debate over the effectiveness of this approach is extensive; it can reasonably be argued that it may be better to enable poor tenants to leave such areas entirely. Yet local elected officials cannot overtly accept abandonment of neighborhoods, whatever the relative effectiveness may be of individual mobility compared with community development.5
The target group, for purposes of this analysis, is tenants living on SSI for the disabled or elderly, a group that receives just over $600 monthly in California. The records of the Berkeley Rent Stabilization Program show that the median 1994 rent for a one-bedroom apartment is $525 monthly. The majority of apartments in Berkeley are studios or one-bedrooms suitable for a single individual, and very few of these units are available for less than $400 monthly, so that unless tenants on SSI receive some form of rent subsidy they usually pay almost all of their income for shelter -- 88 percent for the median rent apartment, leaving only $75 a month for food, not to mention other necessities of life.
The objective of the model program is to enable a tenant living on SSI payments for the disabled or elderly of $600 monthly to rent the apartment whose market rent is $525 a month. Acquisition of this unit by a nonprofit organization using a capital grant can reduce the rent to $240 monthly. This rent is enough to cover normal operating costs including utilities, which are estimated at 41 percent of the median rent, or $215 monthly, and a replacement reserve of $25 per month to keep up the building as major components wear out.6 (While this is a small replacement reserve, the program requires a high level of initial investment in upgrading the building after purchase so later repair and renovation costs should be low.) Using a capital grant to cover costs of purchase, rehabilitation, and program administration enables the nonprofit owner to eliminate the remaining net operating income, and thus allows the rent to be reduced to as low as $240 monthly, which is 40 percent of the tenant's income.
To estimate the per unit costs of the social housing program, we need to know the average capital investment per unit necessary to reduce rents to that level. The cost of acquisition and rehabilitation of the average existing housing unit by a nonprofit housing corporation was estimated as follows: Sales data from county property tax records in 1993 and 1994 show an average price for rental property of just under $42,000 per unit. Not surprisingly, this falls within the range of values that would be estimated by applying the capitalization rates currently used by local appraisers (from 8.5 to 10 percent) on an average net operating income (from 55 percent to 60 percent) of the average apartment rent of $6,300 annually. Normal property acquisition is not usually accompanied by substantial rehabilitation expenditures; with acquisition of housing by nonprofit organizations, however, it is important to have a higher level of rehabilitation than would be undertaken by a private landlord, in order to reduce ongoing operating costs. This often leads to total costs for acquisition and rehabilitation that are at the upper limits of, or even greater than appraised value and are estimated to add an average of an additional 10 percent to the per unit cost of the property. In addition, acquisition and repair have administrative costs both to the City, which provides the subsidy, and to the nonprofit housing corporation, which buys the property, that are estimated to equal 15 percent of the purchase price. This results in a total expected average cost of acquisition of $52,500 per unit. (City of Berkeley policy is that city-assisted projects should pay property taxes, so that there is no ongoing loss of income to the City.)
For the housing allowance program to deliver the same amount of affordability would require monthly subsidies of $285 plus administrative costs of about seven percent, for an annual cost of subsidies and administration of $3,700.
Note that in order to make the unit affordable to a disabled or elderly person living on SSI at the current HUD standard of 30 percent of income for rent and utilities, the monthly rent would have to be reduced to $180. The housing allowance program can do this by increasing the monthly subsidy by $60, to $345. For the social housing program to meet this goal the cost of the acquisition must be fully written down and additional subsidy will be needed. This is the problem facing Housing Authorities; they normally require substantial operating and modernization subsidies, because their tenants are unable to afford enough rent to cover routine operating costs, let alone the additional amount necessary for maintenance and renovation reserves. If, however, the residents of social housing include a mix of low-income tenants, instead of being limited to the poor, it should be possible to provide the additional assistance needed through internal subsidies rather than by depending on government for additional support.7
Since social housing ownership with capital grants can only reduce net operating income and cannot eliminate operating costs, such programs can deliver the largest subsidies in those market areas where housing is most profitable and where net operating income is the highest proportion of rent. Social housing will do little to reduce housing costs where profitability is marginal, although in such areas it does serve a different function, of maintaining decent housing by avoiding landlords' profit-taking at the expense of maintenance.
In order to compare the number of people that can be assisted in both the short and the long run with the same annual commitment of funds to each program, I make projections that initially assume that tenant incomes, rents, and operating costs increase uniformly with the consumer price index, with no relative changes, so that there are no changes in the program environment from year to year. I will then comment on how the results would be affected by divergent changes in these trends.
Since the cost of a year of assistance to one tenant in the housing allowance program is one-fourteenth of the cost of acquisition and rehabilitation of one housing unit, in the first year a housing allowance program can assist 14 tenant households, while the acquisition program assists one tenant household. Figure 1 shows what happens over time if the same amount of money in constant dollars is spent annually on each model program. With the same amount of money, the housing allowance program continues to assist 14 households each year, while the acquisition program adds another unit to the program each year. Assisting a steadily increasing number of households, the social housing acquisition program surpasses the number assisted by the housing allowance program after 14 years.
Clearly the immediate advantage of a housing allowance program is that you can reach a lot more people right away with the same amount of money. The long-term advantage of the acquisition program, however, is that the number of units assisted increases every year, while the housing allowance program helps the same number of people each year. It takes fourteen years before the acquisition program helps as many people as the direct assistance program does, but from year fourteen on, the acquisition program helps more people than the housing allowance program does.
It can be fairly argued, however, that this analysis understates the advantages of the housing allowance program by failing to give appropriate weight to the greater number of tenants assisted by the housing allowance in the earlier years. We can create a measure for the total amount of assistance by treating each year of assistance to each household as one unit of assistance—a "household-year" of assistance—and add up the number of household-years of assistance provided over time. Figure 2 shows that the acquisition program does not provide more cumulative household-years of assistance until twenty-eight years later. After twenty-eight years, however, the acquisition program starts to provide vastly more assistance, since at that point, with the same annual expense, it is helping twice as many tenants annually. Nevertheless, it is clearly a sobering consideration that it requires an entire generation for assistance provided by an investment in social housing to reach the "break-even" point with the assistance provided through housing allowances. Many uncertainties arise over the course of a generation. Among them is the question of whether the social sector will prove itself able to manage and maintain housing consistently and well over the long run (Bratt et al. 1994).
Different future circumstances affect the utility of the two programs, depending particularly on whether rents increase more rapidly or more slowly than incomes do. If the incomes of low-income people generally rise faster than rents do, then the need for subsidies will decrease over time, and the benefits of long-term accumulation of a subsidized housing stock are reduced. For example, if incomes should increase by four percent annually in real terms while rents remain stable, then after 22 years there would no longer be the need for a rent subsidy to the tenant receiving SSI. Thus, if the income support system steadily became more adequate, by the time the social housing sector grew to adequate proportions there would no longer be any need for a housing subsidy program of either type, and a long-term investment in social housing would be unnecessary.
If rents increase faster than incomes do, however, the benefits of the two programs depend on whether this discrepancy results primarily from increases in operating costs or in net operating income. If it results primarily from increases in operating costs, social housing cannot hold down the increases unless it receives operating subsidies. If, on the other hand, rents increase faster than both the increases in operating costs and the incomes of low-income people, then acquisition of housing constitutes an investment of increasing value in the future, and the break-even point in the total assistance it provides arrives somewhat sooner. Although social housing is removed from the market, the value of the investment in social housing is not entirely independent of the market.
In the best of cases, using the same amount of money, housing allowances help far more people right away, while social housing builds up a stock of subsidized housing that could assist more people over the course of a generation. This leads to an initial conclusion about the choice between housing allowances and social housing—that the desirability of the two programs depends to a large extent on one's optimism or pessimism about the future of the United States economy and the nation's social safety net over the next generation. If you expect both substantial economic growth over the next generation and broad sharing of the benefits of this growth with low-income people, then clearly the greater need is to disperse subsidies as widely as possible now, while people really need the help. If you are pessimistic about the economic future, and especially the future for low-income people in the United States, then building up a stock of affordable housing for the future clearly makes sense. Certainly the trends of the last twenty years support the pessimistic scenario, with decreasing real incomes for people dependent on public assistance and decreasing real wages for unskilled labor (Krugman 1992).
Administration of a housing allowance program is a long-term commitment, but a normal one. Many local governments run Housing Authorities, which administer the Section 8 certificate and voucher subsidies provided by HUD. As long as the funding is stable and the same rules apply, such programs are not a major burden. Indeed, many local housing authorities are able to operate their Section 8 programs for less than the administrative cost allowance provided by HUD.
This financial instability is a major potential problem for a housing allowance program. Housing allowances must provide a basic amount of money every month, or people will lose their homes for nonpayment of rent. A local government that creates an ongoing subsidy program requires a stable funding source dedicated to the program. One local rent subsidy program is funded from condominium conversion fees (Connerly 1993). These fees, however, will fluctuate drastically with the local real estate market for condominiums and can at best provide short-term and emergency assistance. Housing allowances require a long-term commitment that must be met month after month and year after year if the assisted people are not to lose their homes.
Social housing programs have an obvious attraction here: the number of units bought or built can be adjusted to the money at hand, yet once the housing is bought and the capital subsidy is put in place, the housing is and will remain nonprofit for the foreseeable future. In a constitutional system of government where protection of private property rights is fundamental, but the right to a decent place to live is at best highly contingent, social housing is endowed with the constitutional protections of private property and cannot be taken away if political changes bring in local administrations that oppose housing subsidies. This contrasts strongly, for example, with the reduced rents delivered by rent controls, a regulatory system that can be weakened or simply abolished by the local government—or by state legislation that preempts local regulatory powers, as has recently occurred in Massachusetts and California.
Political instability is a severe threat to a housing allowance program. While immediate termination of housing allowances to current recipients would result in obvious and immediate hardships, there are now serious proposals at the federal level to phase the assistance out by simply not providing assistance to new applicants, by reducing subsidy levels, and by setting time limits on assistance. In fact, public assistance levels throughout the United States have declined in real dollars over the past twenty years, and there is no reason for confidence that housing allowances will not be similarly weakened. It is even more likely that a local government that has been using its own money to provide housing assistance beyond that provided by the federal government will decide to phase out a supplemental housing allowance program.
There are potential long-term funding sources for ongoing housing subsidies. In San Diego, part of the hotel tax is dedicated to the Housing Trust Fund, and while hotel tax revenues fluctuate, surely some part of the amount could be considered secure (Calavita et al. 1994). In Burlington, Vermont and Seattle, Washington, small property tax increases were dedicated to housing use, and Seattle uses part of its housing trust fund for operating subsidies to housing for the very poor (Connerly 1993; Davis 1994). In San Francisco, an unsuccessful 1990 ballot measure proposed that an increase in the business license tax on rents be dedicated to use for housing subsidies. Receipts from such a tax would increase as fast as rents in San Francisco increased, and thus would provide a very stable base of support for any ongoing subsidy program.
Even unstable initial funding sources could be stabilized and used to support housing allowances, by creating a housing allowance foundation. Since the acquisition program is paying the capitalized value of the unit's net operating income up front, while the direct subsidy program is paying the net operating income monthly, we could get the same result with capital grants to a housing assistance trust fund that then invested the proceeds at an adequate rate of return and used the profits to provide rental assistance in the form of housing allowances. Such a foundation would have to be separate from the local government that created it in order to protect the foundation's capital, after political changes, from efforts to recapture the funds for other uses. Over time, if the value of its investments kept up with local rental housing values, a housing assistance foundation could help as many people as could a social housing program with the same resources. A foundation could also be a valuable supplement to social housing, rather than an alternative. For example, while not used for housing allowances, the Housing Trust Fund in Chicago uses interest on principle provided from a single large mitigation payment to fund operating subsidies to housing for very low-income people who cannot afford even enough rent to cover operating costs (Connerly 1993).
As we have seen, evaluation of the advantages and disadvantages of social housing and housing allowances depends a great deal on one's optimism or pessimism about the national economy and the social safety net. In the local politics of housing affordability the key players will include nonprofit housing organizations, tenant activists, for-profit developers, landlords, real estate brokers, neighborhood organizations, low-income tenants, and elected officials. Except for the landlords, each of these housing constituencies is more likely to support social housing than housing allowances. Housing and tenant activists are usually pessimistic about the nation's commitment to ending homelessness and poverty, and thus are likely to support social housing as an ideal. In addition, nonprofit housing development corporations create interesting jobs that are more consonant with activist ideals than is working for private developers or processing Section 8 applications in a local housing authority. Creation of a social housing sector not only transfers resources into affordable housing, but creates organizations that are capable of political mobilization on behalf of both themselves and their residents, a countervailing force to the political power of the private, for-profit real estate industry. Many nonprofit housing organizations try to organize the residents in advisory councils, which can then provide testimony and a political presence when such support is needed. The fact that these organizations are owners of substantial real property gives them a stability and permanence not found in voluntary organizations. The fact that they are private frees them from constraints on political action that restrain public housing authorities. At a time when the traditional sources of political support for economic redistribution, such as labor unions, have greatly declined in importance, the proponents of a social housing sector aspire to build a new and enduring political force dedicated to ending poverty and homelessness.
The preference at the local level for discrete subsidized housing projects as opposed to ongoing responsibility for housing allowances does not mean that local governments will necessarily choose to help create social housing. Often local governments use one-time subsidies to provide one-time assistance to home buyers (see, for example, Stegman & Luger 1993), or rehabilitation assistance to private landlords in return for only a few years of affordability, or emergency grants to people on the verge of eviction and homelessness. In such cases, there is no cumulative gain in subsidized units, since the subsidies generally expire as fast as new ones are created. Moreover, although local elected officials often prefer physical improvements they can point to over routine monthly social assistance, they can point just as readily to projects built by for-profit developers and can benefit from their campaign contributions as well.
Supporters of the competing uses of local government funds do not, however, always directly oppose social housing as such. In fact, members of competing interest groups often have tactical reasons to support some social housing. Local real estate brokers and agents usually will lobby for programs to assist first-time homebuyers, and they often oppose long-term affordability requirements on the grounds that low-income people should be able to benefit from increased equity, thus reassimilating subsidized housing into the market. But they are not strongly mobilized to oppose such requirements, and some real estate agents also support social housing as a form of charity, directed towards people who are unlikely ever to become homebuyers. For-profit developers simply want supply subsidies and prefer to avoid the long-term affordability requirements of social housing, but they will often work in partnership with nonprofit developers. Partnerships between for-profit and nonprofit developers are often politically valuable to both parties, since they can claim both the efficiency usually attributed to private corporations and the charitable purposes attributed to nonprofit organizations.8 Although private landlords can benefit from rehabilitation loans and housing allowances that pump more rent into local housing, they are also eager to keep property taxes down and avoid or eliminate rent controls and other tenant protections. In Oakland, California the local apartment owners' association has supported nonprofit housing development organizations as a way to divert housing activists from efforts to strengthen Oakland's very weak rent controls. Many neighborhood organizations, especially in middle-income neighborhoods, oppose creation of subsidized housing in their neighborhoods, but they sometimes support it in cases where acquisition or replacement of a vacant or dilapidated building can improve the neighborhood.
There is no visible political constituency for housing allowances at the local level. The San Francisco proposal to increase the gross receipts tax on rents, for example, was defeated by the combined opposition of most landlords, who opposed the tax increase, and most tenant organizers, who opposed a provision that would have eliminated the tax in the event that vacancy control was added to the City's rent control ordinance. Landlords do support such subsidies if they do not have to pay for them, but such programs are not a high priority. The low-income tenants who might benefit from such a program and might prefer the individual mobility provided by housing allowances are not well organized, and tenant organizers generally support social housing. Housing Authorities are required to help tenants in public housing form residents' councils; however, no one brings together Section 8 recipients, let alone those who are on the waiting list for Section 8 certificates or vouchers.
It seems safe to predict that the social housing sector will continue to grow and to place increasing amounts of housing outside of the market. With the loss of ongoing subsidies by public housing authorities, it is possible that a substantial part of the most useable public housing could be converted to social ownership, greatly increasing the size of a fledgling social movement. As long as there are no prospects for creation of adequate income supports for low-income Americans, social housing offers the hope that increasing amounts of permanently affordable housing can grow to eventually meet a substantial part of the need.
At the same time, it is essential to see the inherent limits to the social housing sector. The poorest people in society do not have incomes sufficient even to pay operating costs, so that for social housing to be affordable to the very poor, either it must receive additional subsidies or the residents must be of sufficiently mixed incomes that higher rents for some can subsidize the others. Much of the local support for social housing arises out of concern with particular problems, such as the expiration of federal affordability requirements on subsidized for-profit housing, or the desire of residents in low-income neighborhoods to rehabilitate or replace dilapidated and vacant buildings, and not from support for the broad vision of a social housing sector. While the social housing sector will continue to grow, it is likely to remain a small fraction of the housing stock occupied by the poor, fitting within particular niches in local housing policies and providing a voice for those who otherwise have difficulty making themselves heard. Indeed, it may prove necessary to create a strong social housing sector at the local level in order to provide a voice for housing assistance for the poor, if federal housing allowances are to survive politically at the national level.
Barton is a senior planner in the Planning and Development Department of the City of Berkeley. He is co-author of Common Interest Communities: Private Governments and the Public Interest (Institute of Governmental Studies Press, Berkeley, 1994) and has taught at the University of California at Berkeley, where he received his PhD, and at San Francisco State University.
AUTHOR'S NOTE
I would like to thank Steven Sekhon, my former planning intern, for his assistance in gathering data for the report that led to this paper, and Ken Baar and an anonymous reviewer for helpful comments on earlier drafts. The views expressed in this article are the author's own and are not necessarily those of the City of Berkeley.
NOTES
1. For a review of the history of this debate, see the section entitled "Forty Years of Policy Debate" in Struyk and Bendick 1981b, 24-34.
2. See, for example, the regular reports in Preservation Update, published by the California Housing Partnership Corporation.
3. An alternative form of project-based subsidy is a guaranteed income stream that is used to borrow capital and repay it. Such subsidies are simply an alternative means of providing a capital grant, but paying it over a period of years.
4. NOI is roughly equivalent to what economists call economic rent—the cost of outbidding others for the use of that particular location and that particular housing unit under conditions of inherent scarcity of supply.
5. There is an extensive but inconclusive literature on the use of community development corporations to bring economic and organizational resources into very poor neighborhoods. A recent critique is Nicholas Lemann, "The Myth of Community Development," New York Times Magazine, January 9, 1994. The housing and community development journal Shelterforce devoted most of its January-February and March-April 1994 issues to responses.
6. This is based on the annual cost studies of the Rent Stabilization Program and on operating cost studies of several buildings recently acquired or examined as potential acquisitions by local nonprofit housing organizations.
7. This subsidy could come from a supplementary housing allowance given directly to the tenants, from operating cost subsidies given to the nonprofit owner, or from internal redistribution if the nonprofit owner rents to some people with higher incomes who can afford to pay a rent that is high enough to include some net operating income over and above the operating expenses. One tenant with an income of $1,000 a month could afford to pay $300 in monthly rent, thus providing an average $240 operating cost between the two units.
8. A similar effect has been noted for public-private partnerships, although in some cases the negatives win out, so that the partnership is tarred with both the greed of private corporations and the inefficiency of nonprofit organizations (DeNeufville and Barton 1987).
REFERENCES
Achtenberg, Emily P., and Peter Marcuse. 1983. Towards the Decommodification of Housing: A Political Analysis and A Progressive Program. In America's Housing Crisis: What Is To Be Done?, edited by Chester Hartman. Boston: Routledge & Kegan Paul. 202-31.
Barton, Stephen E. 1977. The Urban Housing Problem: Marxist Theory and Community Organizing. Review of Radical Political Economics 9,4: 16-30.
Bradbury, Katharine L., and Anthony Downs. 1981. Do Housing Allowances Work?. Washington, DC: The Brookings Institution.
Bratt, Rachel G. 1989. Rebuilding a Low-Income Housing Policy. Philadelphia: Temple University Press.
Bratt, Rachel G., Langley C. Keyes, Alex Schwartz, and Avis C. Vidal. 1994. Confronting the Management Challenge: Affordable Housing in the NonProfit Sector. New York: Community Development Research Center, New School for Social Research.
Brooks, Mary E. 1994. Housing Trust Funds. In The Affordable City: Toward a Third Sector Housing Policy, edited by John E. Davis. Philadelphia: Temple University Press. 245-64.
Calavita, Nico, Kenneth Grimes, and Susan Reynolds. 1994. Zigzagging Toward Long-Term Affordability in the Sunbelt: the San Diego Housing Trust Fund. In The Affordable City: Toward a Third Sector Housing Policy, edited by John E. Davis. Philadelphia: Temple University Press. 265-94.
Connerly, Charles E. 1993. A Survey and Assessment of Housing Trust Funds in the United States. Journal of the American Planning Association 59,3: 306-19.
Davis, John E. 1994. Building the Progressive City: Third Sector Housing in Burlington. In The Affordable City: Toward a Third Sector Housing Policy, edited by John E. Davis. Philadelphia: Temple University Press. 165-200.
DeNeufville, Judith I., and Stephen E. Barton. 1987. Myths and the Definition of Policy Problems: An Exploration of Home Ownership and Public-Private Partnerships. Policy Sciences 20: 181-206.
Dolbeare, Cushing N. 1983. The Low-Income Housing Crisis. In America's Housing Crisis: What Is To Be Done?, edited by Chester Hartman. Boston: Routledge & Kegan Paul. 29-75.
Downs, Anthony. 1991. The Advisory Commission on Regulatory Barriers to Affordable Housing: Its Behavior and Accomplishments. Housing Policy Debate 2,4: 1095-138.
Dreier, Peter, and J. David Hulchanski. 1993. The Role of Nonprofit Housing in Canada and the United States: Some Comparisons. Housing Policy Debate 4,1: 43-80.
Finkel, Meryl, and Stephen D. Kennedy. 1992. Racial/Ethnic Differences in Utilization of Section 8 Existing Rental Vouchers and Certificates. Housing Policy Debate 3,2: 463-508.
Krugman, Paul R. 1992. The Rich, the Right, and the Facts: Deconstructing the Income Distribution Debate. The American Prospect (Spring): 19-31.
Low Income Housing Information Service. 1994. The 1995 Housing Budget. Washington, DC.
Lowry, Ira S. 1983. Experimenting with Housing Allowances: The Final Report of the Housing Assistance Supply Experiment. Cambridge, MA: Oelgeschlager, Gunn & Hain.
Moberg, David. 1995. No Vacancy! Moving to Opportunity Program Loses in Suburban Baltimore. Shelterforce 17,1: 15-6.
Rosenbaum, James. 1995. Changing the Geography of Opportunity by Expanding Residential Choice: Lessons from the Gautreaux Program. Housing Policy Debate 6,1: 231-69.
Rossi, Peter H. 1981. Residential Mobility. In Do Housing Allowances Work?, edited by Katharine L. Bradbury and Anthony Downs. Washington, DC: The Brookings Institution. 147-84.
Stegman, Michael A., and Michael I. Luger. 1993. Issues in the Design of Locally Sponsored Home Ownership Programs. Journal of the American Planning Association 59,4: 417-32.
Stone, Michael. 1993. Shelter Poverty, Philadelphia: Temple University Press.
Struyk, Raymond J., and Marc Bendick, Jr. editors. 1981a. Housing Vouchers for the Poor: Lessons from a National Experiment. Washington, DC: The Urban Institute Press.
Struyk, Raymond J., and Marc Bendick, Jr. 1981b. Origins of an Experimental Approach. In Housing Vouchers for the Poor: Lessons from a National Experiment, edited by Raymond Struyk and Marc Bendick, Jr. Washington, DC: The Urban Institute Press. 23-50.
U.S. Department of Housing and Urban Development, Office of Policy Development and Research. 1994. "Residential Mobility Programs." Urban Policy Brief 1 (September): 1-6.
Varady, David P. 1982. Indirect Benefits of Subsidized Housing Programs. Journal of the American Planning Association 48,4: 432-40.
Walker, Christopher. 1993. Nonprofit Housing Development: Status, Trends and Prospects. Housing Policy Debate 4,3: 369-414,
Williams, Kale. 1995. Neighborhood Choice: A Way Out for the Poor. Shelterforce 17,1: 12-4.