Skip to main content

Testimony: How to Promote Housing Integration and Choice through the Section 8 Voucher Program

before the National Commission on Fair Housing and Equal Opportunity

The Section 8 Housing Choice Voucher program is by far the nation’s largest low-income housing program. More than 2.2 million housing vouchers have been authorized by Congress and allocated to the approximately 2,400 housing agencies that administer the program. It is also the housing program that is the most targeted on the lowest income families: 75 percent of new families served each year must be extremely low-income.[1]

The voucher program does a better job than any other low-income housing program of enabling families to live in lower-poverty neighborhoods.[2] But there is mounting evidence that in many metropolitan areas it is not doing as well as it could at helping families to live in safer communities with better schools, services and access to jobs. As a result, it is falling short of its potential to improve the lives of the families it assists. Failing to provide voucher holders access to high opportunity areas may leave them concentrated in a small number of increasingly poor neighborhoods. (The perception that an influx of voucher holders undermines communities may need to be addressed, even where the causal link is missing.)

What can be done to make the voucher program more effective at increasing integration by race and class, while at the same time maintaining a commitment to the core principle of choice? I have spent much of the last 15 years trying to refine the answers to this question. I want to briefly highlight recommendations in three areas — and then address the important issue of housing stability raised by Xav Briggs in his testimony. Policy changes are needed that would —

  1. Empower families to make mobility moves;
  2. Create the right incentives for agencies; and
  3. Increase participation by owners.
  4. Or framed more broadly, we need to jettison the now-discredited assumption that the housing market would work for poor and minority families if they just had the means to pay the rent, and reform the program and its administration to overcome the imperfect knowledge, prejudices and other barriers that continue to limit access to opportunity.

    Empower Families To Make Mobility Moves

    Payment standards — Families may be willing to pay somewhat more than 30 percent of their income (the minimum contribution in the voucher program) to live in a better neighborhood, but poor families have limited discretion to shift spending within tight budgets. Program payment standards — the maximum subsidy payable in a particular area — must be sufficiently high to make it financially feasible for families to choose units in better neighborhoods. Conversely, the subsidy should not be overly generous in neighborhoods where voucher holders may otherwise concentrate, and that already are too poor. Several policy changes would help achieve these objectives.

  • Fair Market Rent areas should be smaller. Each year HUD sets “Fair Market Rents” for each metropolitan area and non-metropolitan county, using data for the entire geographical area. These FMRs are the basis of the voucher subsidy levels set by individual housing agencies: housing agencies have discretion to set payment standards only within 10 percent of the HUD-determined FMR. Rents typically vary substantially between opportunity areas and poverty-concentrated neighborhoods within each FMR area. If FMRs were based on smaller geographic areas they would be more accurate and not be overly generous in lower-cost areas, and the 10-percent flexibility would more frequently be sufficient to enable agencies to set payment standards high enough to promote access to opportunity areas. The more detailed data available through the Census Bureau’s American Community Survey makes it easier to regularly update FMRs for smaller areas. The Section 8 Voucher Reform Act (SEVRA) would require HUD to do this.[3] But HUD has discretion to make this change, and could do so when it proposes the 2010 FMRs in the spring of 2009.
  • HUD should set the FMR based on 50th percentile rents where needed to expand housing choice. In 2000, HUD issued a new policy by interim rule to expand housing choice in certain large metro areas where data indicate that voucher use is overly concentrated. To enable families to use vouchers in outside of these concentrated areas, the policy increases the metropolitan area FMRs by basing the FMR on the midpoint — or 50th percentile — of the rent distribution, rather than pegging it at the 40th percentile which has been the policy since the mid-1990s. The Bush Administration never issued a final rule. Initially, 39 metro areas with a substantial share of the nation’s population qualified for higher FMRs — and hence higher payment standards — under the interim rule. By 2008, the number of qualifying areas had fallen to 28 (in 2005, HUD dropped many areas to the 40th percentile but increased 10 areas to the 50th percentile). For 2009, HUD has reduced the number of qualifying areas to 14. Based on the last 8 years of experience, it is important to evaluate the statistical and other criteria in the interim rule and refine the policy in a final rule to better achieve the goal.[4]

It is also important to make it easier for housing agencies to adopt “exception payment standards” above 110 percent of the FMR, as discussed in the section on agency incentives below.

Housing search policies and assistance — Housing agencies are required by federal law to set a time limit on the period families can search for housing in which to use their vouchers, and may allow extensions. Federal law does not prescribe a limit on the search period, however, and when agencies set it too short — the minimum is 60 days — families may not have sufficient time or be willing to take the risk to search in unfamiliar areas. Flexible search policies are essential to increasing the mobility potential of the voucher program. During the Clinton Administration, HUD made the rule changes necessary for housing agencies to be flexible. The policy challenge now is to encourage more agencies to use this flexibility, or to require longer search periods under certain circumstances.

Even with more flexible search periods, however, many families will need assistance to look in unfamiliar areas. Housing agencies rarely provide such direct assistance unless they have special funding for this purpose. Provision of detailed briefing materials is more common, but it is unclear whether this more economical approach is effective. HUD should investigate the effectiveness of such information-oriented search assistance.

Additional funding for direct search assistance could be more cost-effective than broad scale increases in voucher subsidy payments, and could potentially be designed so that programs could be relatively short-term but with lasting benefits. HUD never published the evaluation of the specially-funded Regional Opportunity Counseling programs, but anecdotal evidence indicates that those programs had mixed results. Before seeking funding for a new generation of ROC programs, HUD should redesign the program based on the experience of ROC grantees and other agencies that have operated mobility programs to implement litigation settlements and for other reasons.

Portability policies — The administrative geography of the voucher program — its balkanized operation in most metropolitan areas — creates substantial barriers to families moving from poorer and more racially concentrated areas to areas with greater opportunities. Boston is one of the worst examples: more than 60 agencies administer section 8 vouchers in the metro area. There are more section 8 agencies in Massachusetts than in California! When different agencies administer the voucher programs in different parts of a metropolitan area, families need to use the “portability” feature of the voucher program to move from one area to the other. Families may not understand their right to make such moves, and agencies may discourage such moves through lack of information or other policies. Unfortunately, HUD rules governing portability moves create financial and administrative disincentives for agencies, as they may incur higher costs and receive less administrative funding when families “port.” This problem can be addressed in two ways: by reducing the administrative barriers, or reducing the number of agencies.

SEVRA takes the first approach, by transforming “portability” policy so that families can move to areas served by other agencies and the “receiving” agencies generally must accept them (and receive additional funding for this purpose). The House and Senate bills differ somewhat on this key policy, and the approach in both bills remains controversial in some quarters, in large part due to a lack of trust in how HUD would administer it. In my opinion, statutory changes are not necessary for HUD to revamp its rules so that portability would work more efficiently and effectively.[5] This is a key area for leadership and initiative by a new HUD. But Congress should act if HUD fails to do so.

The alternative would be to reduce the number of agencies operating within metropolitan areas and expand their geographic scope, but this would pose a far more difficult political challenge. Moreover, there is little evidence at this point to help policy-makers decide whether enlarging agencies’ service areas would lead to more families moving to opportunity areas. HUD commissioned exploratory research on this issue in the mid-1990s. The report found that virtually the only agencies operating across metropolitan regions were state agencies.[6] Of the roughly 30 states that administer voucher programs, however, Center research has found that many are small and areas of operations are often geographically constrained by state law. Reforming portability policies, as challenging as it will be, is a far easier means to the goal of promoting mobility than attempting to consolidate a number of small agencies with longstanding political allies into a single entity serving an entire state or metropolitan area. If portability policy changes combined with the additional recommendations discussed below do not produce adequate results, however, this option should be reexamined.

Keeping the bulk of vouchers tenant-based, while increasing flexibility to project-base vouchers in opportunity areas — Currently, agencies may “project-base” up to 20 percent of their vouchers — that is, attach them to particular projects selected by the agencies (though families retain the right to move with the next available voucher after one year). The Housing Act requires agencies to choose locations that are “consistent with the goal of deconcentrating poverty and expanding housing and economic opportunities.”[7] In 2005 HUD adopted a final rule that leaves this judgment up to housing agencies, but requires them to consider seven specified factors as part of their annual plan if they choose to use the project-based option.[8] The regulatory policy is a balanced approach to a difficult issue. The problem is its implementation: it appears that there is no HUD oversight of whether an agency has in fact considered the required factors in identifying neighborhoods where project-basing is permitted or encouraged. It may be appropriate for the Office of Fair Housing to review project-basing plans, at least on a sample basis, and to determine whether further policy changes are required.

SEVRA would allow housing agencies to increase the share of vouchers that may be project-based to 25 percent, and up to 30 percent to provide supportive housing to formerly homeless people. The across-the-board increase to 25 percent, however, may not garner the necessary bipartisan support in the Senate Banking Committee. It may make sense to condition such an increase on use of the authority in objectively-defined opportunity areas. Families would be much more likely to move to an area where vouchers are harder to use if they could bypass the difficult search process by moving to a project that is voucher-assisted.

It is also important to ensure, if exceptions are allowed to the 20 (or 25) percent limitation on the share of vouchers that can be project-based, that the exception is limited and that at least half of an agency’s vouchers are still fully mobile. HUD has allowed agencies in the Moving to Work (MTW) demonstration to project-base an unlimited number of vouchers, and to administer the subsidies without a mobility option, like old-fashioned project-based assistance programs.[9] The House version of SEVRA would cap the share of vouchers that may be project-based under MTW or its successor at 50 percent, and constrain the limitations on the mobility option. Given the size and the degree of racial concentration in some of the cities where agencies have MTW status (e.g., Chicago, Baltimore, Atlanta), this policy change is important.

Create the Right Incentives for Agencies

Many of the policy changes discussed above also would facilitate and encourage agencies to make program design choices that would better promote mobility. From the perspective of the administering agencies, however, there are two key sets of policy changes that will help make it possible and practical for them to make these program choices.

  1. Voucher renewal funding policies must not force agencies to make long-term trade-offs between the adequacy of assistance to help some families live in areas with greater opportunities and the number of families served. The quasi-block grant funding policies adopted by Congress and HUD in 2005 and 2006 prompted many agencies to prohibit portability moves and to reduce payment standards, in order to avoid being caught short and forced to terminate vouchers due to unexpected increases in rental costs. Beginning in 2007 Congress changed these policies and instituted a more reasonable renewal funding policy, but the continuing uncertainty about whether the next year’s appropriations act will change the rules again discourages agencies from adopting mobility-promoting policies that can have the effect of increasing program costs. Permanent renewal funding policies that agencies know they can count on, and which support agencies’ efforts at deconcentration (for example, by assuring agencies that they will have the funding they need to support portability moves across jurisdictional lines), must be incorporated in the Section 8 statute. SEVRA would make this essential change.
  2. Performance evaluation policies and other incentives must be designed to encourage actions and policies that promote mobility. Agencies respond to produce the types of results HUD pays attention to. The experience with voucher utilization is an excellent example. In the late 90s, Congress and HUD were very concerned that hundreds of thousands of vouchers had been authorized and funded by Congress that agencies were not issuing to families. HUD designed a number of policies to encourage utilization, including emphasizing it in the point system in the new Section 8 Management Assessment Program (SEMAP), conditioning eligibility for new vouchers on a minimum 97 percent utilization rate, and promulgating a policy that would reallocate unused vouchers permanently to other agencies. Agencies got the message, and in only a few years increased the share of vouchers in use from less than 90 percent to about 98 percent in 2003-2004. (Unfortunately, the renewal funding policy changes — including the end of the “use it or lose it” policy — and funding shortfalls in the intervening years resulted in a 90 percent utilization rate in 2007.) Here are a few specific policy changes to create the incentives for agencies to adopt policies and practices that support and encourage families to make mobility moves.
  • SEMAP should be redesigned to more effectively measure deconcentration. Now, deconcentration is an optional and minor component of the evaluation system, and the metric is not well-designed. (This is not to criticize those who fought for it and prevailed in getting the current measure included in the face of agency opposition.) SEVRA would require this change, but HUD could adopt it without legislation.
  • Agencies that operate regionally or achieve deconcentration objectives should receive preference for new vouchers. The Senate version of SEVRA includes this policy change. Local agencies that collaborate to make moves between them seamless could qualify for the preference, as well as agencies that administer vouchers over a larger area.
  • Make information about voucher concentration public and require agencies to respond to the data. SEVRA requires HUD to report annually to Congress and to agencies on voucher concentration and rent burdens; agencies must assess whether the data indicate a need to increase voucher payment standards or to address the problem in other ways. (The Senate bill refines this requirement, and specifies that data on voucher concentration must be analyzed separately for racial and ethnic groups.) HUD collects these data now from agencies through the annual rent recertification process, and could, without a legislative directive, analyze the data and disseminate the findings.
  • Simplify the process for agencies to get HUD approval to use higher payment standards to promote deconcentration. SEVRA also bypasses the burdensome procedures HUD rules establish to qualify for an exception payment standard. (Moreover, HUD has virtually ceased to implement these rules for most of the Bush Administration.) The bills would require HUD to approve an increase in the payment standard to 120 percent of FMR if an agency requests the increase to alleviate excess rent burdens or voucher concentration.[10] HUD could revamp its rules along the lines of the SEVRA policies to streamline the procedures for approval of higher payment standards.
  • Provide performance bonuses to agencies that achieve mobility objectives. A portion of agencies’ administrative fees — or a supplement to the formula amount — could be conditioned on performance in achieving mobility and other objectives. The Senate SEVRA bill gives HUD the flexibility to design such a policy; the House-passed bill would preclude it. Not surprisingly, the Senate provision is controversial with agencies, but in my view it is important for HUD to have such flexibility, at least to promote certain policy objectives.

Increase Participation by Owner

The recommendations above will only be effective if there are sufficient owners of properties in opportunity areas that are willing to rent to voucher holders. Experience indicates that agencies can increase participation by owners in the voucher program, including in tight markets, by operating efficient programs and reaching out to owners in various ways to encourage participation. Such landlord outreach is likely to cost far less than direct support for families in their housing search. Many agencies have undertaken such initiatives without special administrative fees or other funds. The key is to create the incentives discussed above for agencies to initiate such efforts. It also would be helpful to streamline the inspection process to reduce burdens on owners, as SEVRA would do, and for HUD to disseminate best practices.

But encouraging participation, operating an effective program and paying a fair rent may not be sufficient, particularly in areas where owners can easily rent their units. Except in the relatively few states and cities that have laws prohibiting discrimination against voucher holders, owners are allowed to refuse to accept voucher payments so long as their refusal is not a pretext for discrimination against protected classes. The only exception is owners that receive certain types of federal subsidies, such as Low Income Housing Tax Credits and HOME block grant funds. Even these owners, who are already prohibited from discriminating against voucher holders, often use policies that have a discriminatory effect. It is important for HUD and Treasury to clarify the types of policies prohibited by current law, such as minimum income tests that ignore the value of the voucher subsidy. In addition, the Senate SEVRA bill would direct the U.S. Government Accountability Office to study the obstacles to use of vouchers in LIHTC and HOME-assisted properties. At the least, such a study should be conducted expeditiously, and Congress and the responsible agencies should act on the results.

Some opportunity areas have so little rental housing that none of these recommendations will be effective. In such areas, new development is needed (though the obstacles posed by regulatory barriers and other means of local resistance should not be underestimated). Project-basing vouchers in a portion of the units in such new developments can ensure that these substantial investments create opportunities for voucher holders. State (and in some cases, local) agencies that allocate Low Income Housing Tax Credits and other production subsidies can design their selection criteria to reward developers that commit to accept project-based voucher contracts. To encourage states and other agencies to adopt such policies, Congress could give a preference in the award of new vouchers to agencies that will partner with state allocating agencies or can directly implement such initiatives.

One lesson we should learn from the Clinton Administration, however, is that efforts to encourage owner participation by limiting protections for families should be approached with caution and closely scrutinized to ensure that the benefits exceed the potential harm. In the early ‘90s, owner groups urged HUD and Congress to make a set of changes in the voucher program that they claimed would remove barriers to increased participation by “good” owners. Among these changes was the elimination of the requirement that owners have good cause to not renew a lease (pejoratively labeled the “endless lease” requirement). Because of that change, owners in the lower-poverty areas where families moved under the Moving-to-Opportunity for Fair Housing (MTO) demonstration in the ’90s were able to decide not to renew families’ leases as markets tightened, forcing families to have to relocate. It may well be time to reconsider that policy change, or at least to investigate whether it in fact has expanded housing opportunity sufficiently to outweigh its evident harm.

End Notes

[1] For background, see “Introduction to the Housing Voucher Program,” Center on Budget and Policy Priorities, revised July 6, 2007.

[2] Deborah Devine et al., Housing Choice Voucher Location Patterns: Implications For Neighborhood Welfare, U.S. Department of Housing and Urban Development, January 2003.

[3] H.R. 1851 passed the House in 2007 by a wide bipartisan margin. The Housing Subcommittee of the Senate Banking Committee held a hearing in the spring of 2008 on the companion bill, S. 2684, but the Senate is not expected to act on the legislation before the end of the 110th Congress. The Center’s website, at, includes our analysis of the Senate and House bills, a side-by-side comparison of the bills with current law, and a powerpoint that provides a brief overview of the bills.

[4] It is not clear how necessary it would be to use 50th percentile FMRs if the first recommendation concerning smaller FMR areas is implemented. This is a statistical question that should be investigated.

[5] Section 8(r), which governs portability, is written sufficiently broadly that HUD could eliminate the administratively complex and burdensome system it created and issue new regulations that streamline portability procedures. Voucher renewal funding policies in annual appropriations acts could, however, hinder HUD’s flexibility considerably. The SEVRA renewal funding policies, discussed below, would give HUD the flexibility it would need to reform portability administratively. Such changes to the portability and renewal funding policies were endorsed by a large number of national and local groups. See

[6] Judith D. Feins, et al., “State and Metropolitan Administration of Section 8: Current Models and Potential Resources, Final Report,” July 1996.

[7] Section 8(o)(13)(c)(ii), 42 U.S.C. §1437f(o)(13)(c)(ii).

[8] 24 C.F.R. §983.57.

[9] Under the project-based voucher program, a family can move from a unit with project-based voucher assistance, after 12 months, using the agency’s next available tenant-based voucher. This policy innovation, which took effect in 2001, is vital to allow economically-constrained families to choose where to live as their needs — or neighborhoods — change. Maintaining this option is central to a housing policy grounded on choice.

[10] The bills differ with regard to the triggers for agency and HUD action, and the Senate bill would apply some additional requirements on agencies to qualify for the increase. These differences are described in the Center’s side-by-side on the bills at Both bills would give agencies the authority to increase the voucher payment standard to 120 percent of FMR without seeking HUD approval as a reasonable accommodation for individuals with disabilities.