March 16, 2004
PASSING THE BUCK: THE ADMINISTRATION’S “FLEXIBLE
VOUCHER PROGRAM” WOULD COMPEL HOUSING AGENCIES TO IMPOSE
DEEP CUTS IN 2005 AND SUBSEQUENT YEARS
By Barbara Sard and Will Fischer
PDF of report
If you cannot access the files through the links, right-click on the underlined text, click "Save Link As," download to your directory, and open the document in Adobe Acrobat Reader.
The Administration’s budget contains a proposal both to replace the existing
HousingChoice Voucher Program with a new “Flexible Voucher Program” that is essentially a block grant and to cut voucher funding substantially. In 2005, the budget would cut funding by more than 12 percent below the level needed to maintain current services. The proposal also would cut the administrative funding provided to public housing agencies to operate the voucher program. Under the proposal, PHAs would be forced to institute substantial reductions both in the assistance they provide to low-income families and in their own administrative infrastructures.
The cuts that PHAs would have to impose would likely grow more severe after 2005. Under the proposed block grant, funding levels would no longer be tied to the cost of maintaining the current number of vouchers; funding could instead be frozen from year to year or rise much more slowly than rental costs. Adding to this problem, block grants have historically proven easier to freeze or cut than other programs, in part because of the difficulty of identifying the specific consequences of freezing or reducing federal funding for programs that operate as block grants.
The conversion of the Section 8 voucher program to a funding structure that is more vulnerable to cuts would be particularly dangerous at this time, since Congress faces sizeable budget deficits in the years ahead and the Administration is pushing for the imposition of very tight caps on the overall levels of funding that Congress may provide for non-entitlement programs in each of the next five years. Under the proposed caps, the total amount appropriated for non-entitlement programs would have to be cut by increasing amounts each year.
Moreover, the Administration has already indicated its intention to seek much deeper cuts in the Section 8 program in coming years. The part of the budget that details the Administration’s funding plans for years after 2005 shows voucher expenditures being cut in 2009 to a level $4.6 billion — or 30 percent — below the amount the Congressional Budget Office estimates will be needed to maintain today’s levels of voucher assistance.
In promoting its Flexible Voucher proposal, the Administration has tried to portray the FVP as a painless way to reduce federal spending. Administration officials have soothingly argued that the expanded flexibility PHAs would be accorded under the proposal would enable agencies to achieve the required cost reductions through increased “efficiency.”
But careful examination of the proposal shows that this is not the case. The proposed funding cuts are much too large to be achieved without substantially affecting low-income families — including current voucher-holders — and without embroiling many housing authorities in what could become bitter controversies in their communities about how to institute these steep cuts.
The principal effect of the increased flexibility that the proposal would accord PHAs in areas such as income targeting and setting payment standards would likely be to make local housing agencies — rather than federal policymakers — responsible for the excruciating choices regarding how to cut assistance enough to compensate for the large funding shortfalls the federal funding cuts would create. PHAs would have to decide whether to eliminate large numbers of vouchers and substantially reduce the number of families they serve, to raise rent burdens sharply, or to shift vouchers away from poor families. PHAs would be forced to defend these decisions to voucher holders, advocates, landlords, political leaders and others in the local community, some of whom might direct their ire over these cuts at the PHAs, rather than at
. Furthermore, PHAs would likely have to make these difficult choices over and over again, because under the Administration’s budget the cuts in voucher funding would grow deeper in later years. Washington
The Administration’s proposal does include some policy changes that could reduce PHA administrative costs modestly, such as a reduction in the required frequency of unit inspections. The savings from these changes, however, may not be sufficient even to compensate for the cuts in administrative funding that the Administration is proposing (which would come on top of large administrative funding cuts that Congress has already enacted). And any savings that might remain from such changes, after PHAs had addressed their administrative funding shortfalls, would be too small to offset more than a tiny fraction of the hefty reductions in voucher-subsidy funding. It also bears noting that administrative changes such as a reduction in the frequency of inspections could be achieved without converting the program to a block grant and shrinking program funding.
PHAs Would Be Forced to Make Deep Cuts in Assistance to Low-Income Families in 2005
As explained below, the Administration’s proposal would lead to substantial program cutbacks that would adversely affect low-income families.
- “Efficiencies” cited by Administration could achieve only a small portion of the required cuts. HUD has acknowledged that the funding level the Administration has proposed for fiscal year 2005 falls more than $1 billion below the funding level provided for 2004 — and is more than $1.6 billion short of the amount needed to maintain the program at its current level. Similarly, the Chairman of the House Appropriations Committee, Rep. Bill Young, recently stated that the Administration’s funding level is about $1.7 billion short of what is needed to maintain current operations.
Some HUD officials have said PHAs could absorb the proposed cut through increased efficiency, but HUD has provided no convincing explanation of how agencies could achieve more than $1.6 billion in “efficiencies.” (As noted above, for many local agencies, savings from administrative changes such as reducing the frequency of inspections might not be sufficient even to cover cuts in administrative funding.)
The only means of reducing rental subsidy costs — as opposed to administrative costs — that HUD has cited that would not adversely affect low-income families is to reduce errors in calculating rents. Only a fraction of any reduction in errors would translate into cost savings, however, in part because a significant proportion of errors result in underpayment of rent. Furthermore, according to HUD documents, errors and improper payments already were reduced 27 percent in 2003 and will fall further in 2004. Opportunities for further error reduction in 2005 and beyond are likely to be more limited.
- The bulk of the proposed cuts thus would have to be implemented in ways that harm low-income families. The Administration’s proposal would cut voucher renewal funding in 2005 by more than 12 percent below the amount needed to support all authorized vouchers expected to be in use. (This refers just to the reduction in housing assistance payment (HAP) funding, not to the additional reduction in administrative fees the Administration has proposed.) The bulk of these savings would have to be secured in one or more of the following ways:
1. Reducing the number of families assisted. If the cuts were implemented by eliminating vouchers, the number of families served would have to be reduced by about 250,000 in 2005.
2. Shifting rent burdens to low-income families. This could be done either by charging more than 30 percent of adjusted income for rent or lowering payment standards substantially. Covering the proposed shortfall entirely by raising rent burdens would require an average rent hike of about $850 per voucher family in 2005.
3. Shifting vouchers to higher-income families. PHAs could reduce the average cost per voucher by shifting vouchers that become available through turnover away from “extremely low-income” households (i.e., households with incomes below 30 percent of the area median income) to households that have higher incomes and thus need less assistance. Such a step would mean helping fewer of the families with the greatest need, regardless of whether serving homeless families or other extremely-low-income households has been established as a priority for the local community.
Shifting vouchers in this manner could make up, however, for only a modest portion of the proposed funding cut. HUD budget documents include a table showing the savings that could be achieved if vouchers were shifted away from the neediest families at various rates. Even the most aggressive such redirection of vouchers considered by HUD would generate less than one-fifth of the savings needed to compensate for the funding cut the Administration has proposed for 2005.
PHAs would face these unappealing options not only in 2005, but also in the years that follow. The Administration’s budget plans call for funding cuts that grow larger after 2005. The choices that PHAs would be compelled to make in implementing the funding cuts would become even more excruciating over time.
- Housing agencies would need to cut assistance for current participants. The proposed cuts are too large to be achieved through attrition alone, unless about one-quarter of current participants leave a PHA’s program in 2005. Most housing agencies would need to cut assistance to families currently relying on vouchers or to raise tenant contributions to such a degree that substantial numbers of tenants with vouchers might no longer be able to cover their housing costs. Most agencies, no matter how committed they are to the well-being of the families they serve, would have little choice but to make decisions that would likely force some families to relocate to substandard or overcrowded housing — and in some cases, even to shelters or onto the streets. In addition to harming participants, terminating subsidy payments could do considerable damage to the relationships that PHAs have cultivated with landlords over the years.
- Housing agencies would be forced to freeze waiting lists. Only agencies that have an unusually high rate of attrition — or that impose sharp rent increases or payment-standard reductions — would have any funding available to reissue vouchers to families on their waiting lists in 2005, as other families leave the program. As a consequence, agencies would be forced to close their waiting lists or watch the lists grow steadily longer, with no possibility of providing vouchers to the rising number of families waiting for aid. In addition to creating further hardship for families waiting for assistance, a freeze on the issuance of new vouchers could have serious consequences for social service programs in the community that depend on the availability of some housing vouchers each year to assist families that have recently entered homeless or domestic violence shelters or to help people with disabilities leave institutions. Communities also would lose the ability to provide vouchers to families who require rental assistance to avoid slipping into homelessness.
Have PHAs Behaved Irresponsibly?
Administration budget documents seek to justify these proposed funding cuts in part by suggesting that irresponsible behavior on the part of PHAs in setting payment standards has driven up voucher costs. This charge is wide of the mark. There is no reason to believe that PHAs have abused their flexibility to set payment standards or that it would be possible to reduce voucher payment standards substantially in most areas of the country without compromising the housing choice that underlies the voucher program’s success.
No evidence indicates that PHAs on average have set payment standards far above the Fair Market Rents. Any effort to cut costs markedly by reducing the amount of rent that a voucher can cover thus would entail reducing payment standards significantly below the Fair Market Rent. Such a reduction would almost certainly push down “success rates,” force many voucher families to make up for the reduction by taking on higher rent burdens, and leave large numbers of families with little choice but to live in areas with higher concentrations of poverty, poorer schools, higher crime rates, and fewer job opportunities.
That actions to lower payment standards would make it more difficult for housing agencies to avoid concentrating large numbers of voucher holders in a limited number of neighborhoods is a matter of particular concern. Such concentrations tend to be disadvantageous for the families involved and can be controversial at the local level.
Conversion to a Block Grant Would
Pave Wayfor Further Cuts in Future Years
The “Flexible Voucher Program” is a block grant and would likely result in voucher funding eroding over time, as has occurred with numerous other block grants. Although the Administration has taken pains to avoid referring to the FVP as a block grant, it clearly is a block grant. Instead of providing funding based on the expected costs of vouchers that will be in use, Congress would pick a funding level each year. And instead of changing the program’s parameters to comport with the funding reduction, the proposal would simply eliminate key federal rules and toss to local agencies the decisions as to how to alter the program to fit within the sharply diminished resources.
Of particular note, the FVP proposal contains no mechanism to adjust funding from year to year based on changes in housing costs or any other factor. With no link between funding levels and the actual cost of vouchers, Congress could freeze or even reduce funding for the program without being clearly accountable for the cuts that local agencies would then have to impose.
For this and other reasons, it is very likely that funding for the FVP would fail to keep pace with rental costs and that PHAs would have to cut assistance more deeply as time passed. Other block grants that provide core services to low-income families have not fared well in maintaining funding. For example, the welfare-reform block grant, which provides funds for basic cash assistance to poor families with children as well as for welfare-to-work and child care programs, is on track to decline 22 percent in purchasing power from 1998 (the block grant’s first full year) to 2009. Funding for the State Children’s Health Insurance Program, which insures low-income children, will be nearly 20 percent lower in 2005 than when the block grant was established in 1997. And funding for the Social Services Block Grant, which funds an array of social services for low-income children, seniors, and people with disabilities, has lost 84 percent of its purchasing power since the block grant’s inception in 1973.
The HOME block grant has done better than a number of these other low-income block grants. Yet it, too, has lost ground. From fiscal year 1992 (the first year the program was funded) to fiscal year 2004, the funding level of the HOME block grant (excluding funding set aside for special purposes) has fallen 7 percent, after adjusting for inflation. Since rental housing costs have risen more rapidly than general inflation during this period, the level of housing production and other activities that the HOME grant can support has fallen by a greater percentage in many areas.
HousingOperating Fund is not a block grant, but it underscores the point that it is easier for Congress to freeze or cut funding when the consequences of the reduction are not readily apparent. Congress frequently provides less money for the Operating Fund than the fund’s formula calls for, even though the formula is designed to set funding at a level that is adequate (in combination with rental income) to cover the day-to-day costs of running public housing. In underfunding the Operating Fund, Congress leaves many PHAs no choice but to reduce costs by cutting back on activities such as maintenance or security. When a roof leaks or the crime rate rises, it is the PHA — rather than Congress — that appears to be responsible. The Administration’s voucher proposal would shift apparent responsibility for voucher cuts to PHAs in a similar manner.
Administration Budget Calls for Exceptionally Deep Cuts in Later Years
Adding to these concerns, the Administration’s budget shows cuts of stunning depth in the voucher program in years after 2005. In its recent analysis of the Administration’s budget, the Congressional
BudgetOffice reported that under the budget, Section 8 expenditures would be $4.6 billion lower in 2009 than the level that would be needed to support the voucher program at its current level of operation.
If this $4.6 billion reduction were implemented by reducing the number of families served, state and local housing agencies would have to eliminate about 600,000 vouchers by 2009. This amounts to 30 percent of the number of vouchers now in use.
If the cut were carried out by raising the rents of voucher tenants, the average tenant family would face an annual rent increase of about $2,000 per year by 2009.
The levels of funding that the Administration’s budget shows for the Section 8 program in years after 2005 should be taken very seriously. The Administration is proposing that Congress enact legislation imposing overall annual caps on the total level of appropriations that can be provided for non-entitlement programs in each of the next five years. Under the proposed caps, if Congress wanted to set funding for the Section 8 program above the shrunken funding levels shown in the Administration’s budget, Congress would have to set funding for other programs below the levels the budget includes for them. This would not be easy to do. The Administration’s budget already contains cuts by 2009 in virtually every domestic non-entitlement program area in the budget, except for space exploration.
Large Voucher Cuts Less Likely if Block Grant is Rejected
Even so, the $4.6 billion reduction by 2009 is not inevitable under the proposed caps. The 30-percent cut in voucher-program funding in 2009 that is shown in the Administration’s budget is considerably larger than the funding reductions that the budget shows for most other programs. The average cut proposed for domestic non-entitlement programs outside of homeland security is 10.4 percent in 2009. If Congress had to take responsibility for the severe hardship that would result from deep cuts in voucher funding, as it must do under the current funding structure, it is likely that Congress would feel compelled to make the difficult choices necessary to provide substantially higher levels of funding for the voucher program. Indeed, in fiscal year 2004, despite a tight fiscal environment, Congress increased funding for vouchers by more than $1 billion above the level the Administration requested. Congress took this action because the increase was needed to ensure that the number of low-income families assisted under the program would not be reduced.
If the current funding structure is replaced by a block grant, the consequences of a large reduction in federal funding for the voucher program would be much less apparent. Congress would consequently be less likely to make the tough choices needed to avoid substantial cuts in voucher funding. PHAs and housing advocates almost certainly would find it more difficult to protect the voucher program and the families it serves against the type of deep cuts that the Administration’s budget contains
Administrative Funding Would Be Too Low For Most Agencies
The Administration’s budget, including the FVP proposal, would also lead to substantial reductions in administrative funding for PHAs. The budget would cut total administrative funding for PHAs in fiscal year 2005 by $59 million below the 2004 level. This represents a reduction of about 5 percent.
Furthermore, under the FVP proposal, the “base” amount of administrative funding that all housing agencies would receive would be limited to 7 percent of the funding that an agency receives for voucher renewals. This represents a substantial reduction from 2004, when the amount of administrative funding provided by Congress equaled 9.5 percent of voucher renewal funding, and the 2004 level itself was a reduction from the administrative funding level in previous years (and may not be adequate for some housing agencies). In 2002 and earlier years, agencies received administrative fees equivalent, on average, to at least 10.4 percent of the amount of voucher subsidy (HAP) funding they received. Because of the proposed 7-percent limitation, many agencies would experience cuts in administrative funding significantly larger than the 5-percent overall reduction.
Moreover, the 7-percent limit would be even more stringent, since it would equal 7 percent of subsidy funding levels that had been substantially reduced by the Administration’s cuts in Section 8 funding. An agency that earned $95,000 in fees and received funding for $1 million in HAP payments in 2004 could see its HAP (or rental subsidy) funding allocation reduced to $880,000 in 2005, as a result of the Administration’s proposed voucher funding cut. Seven percent of this lower HAP level would equal $61,600, or 35 percent less than the agency’s level of administrative funding in 2004.
If the still-deeper cuts in the voucher program that the Administration’s budget shows for years after 2005 are imposed, agencies could experience ever more drastic cuts in administrative funding in those years. The reductions could be especially severe for those housing agencies that serve low-rent areas, since “base administrative” fees would be determined as a percentage of each PHA’s subsidy funding, rather than through a formula that also takes account of wages and other administrative costs.
Under the Administration’s FVP proposal, the 7-percent “base” administrative funds would be supplemented by “performance-based” incentive bonuses. Approximately $336 million of the total of $1.176 billion that the Administration has requested for administrative fees in fiscal year 2005 would be reserved for such bonuses. For agencies that receive them, these bonuses could make up part or all of the shortfall in administrative funding. The overall cut in administrative funding would be so large, however, that most agencies would experience reductions even after the bonuses are taken into account.
Furthermore, the examples the Administration has cited of likely performance criteria for the award of these bonuses suggest that the bonuses could be awarded to the agencies that most faithfully pursue the Administration’s policy priorities, rather than to agencies that administer the program most competently. For example, one criterion would reward use of vouchers for homeownership. Agencies that shift vouchers away from generally lower-income renters to generally higher-income homeowners would earn more points. Agencies that move households off of the voucher program quickly also would be considered high performers. The Administration has not explained how this would affect agencies that serve substantial numbers of elderly and disabled people or of working families that are unable to raise their earnings to a level where they can afford housing on their own, or for that matter, how this criterion would affect agencies that serve substantial numbers of homeless people who face major barriers to work. PHAs that serve these groups apparently would be less likely to be rated high performers.
Elimination of Federal Funding for the Family Self-Sufficiency Program
The Administration also is proposing to eliminate dedicated funding for FSS coordinators. HUD could set aside bonus funds to support FSS, but there is no assurance that HUD would do so, or if it did, that the funds would be sufficient to support a significant number of current FSS coordinators. PHAs could be forced to choose between ending their FSS programs by breaking contracts with current participants and funding FSS on their own (and using scarce administrative funds to do so).
Benefits of Flexibility for PHAs Could Be Achieved Without Risky Transformation to a Block Grant
Changes to simplify administration of the voucher program and allow PHAs to economize on administrative costs, such as simplification of rules regarding inspections, recertifications, reporting, homeownership, project-basing, and rents, could be instituted without converting the program to a block grant and slashing program funding. Some of the members of Congress of both parties introduced legislation in 2002 containing provisions simplifying voucher program administrative requirements and reduce administrative costs. Similar legislation may be introduced this year.
By contrast, under the proposed block grant, the added flexibility that PHAs would be accorded would likely turn out to be flexibility primarily in how to institute substantial reductions in assistance. The Administration’s proposal would grant PHAs new flexibility in areas such as rent rules, payment standards, and income targeting. Because of the magnitude of the cuts in voucher funding that the Administration is proposing, PHAs would be forced to use this flexibility principally to determine how the cuts would be distributed among the needy families they serve.
No federal funding would be available to PHAs to serve new populations or undertake new initiatives, unless a PHA chose to free up resources by imposing even stiffer cuts on current participants. Moreover, if a PHA did so but federal voucher-funding then continued to decline, the PHA might be forced to choose in subsequent years between abandoning its new initiative and imposing still harsher cuts on the remainder of its program in order to continue funding the initiative.
Initial Block Grant Flexibility Can Erode Over Time
Finally, under other block grants, initial flexibility has sometimes given way over time to new federal mandates and constraints. A pattern has developed whereby state and local administering agencies may be given considerable flexibility at the time a block grant is created, only to see that flexibility whittled down in subsequent years as Congress imposes conditions on the use of block-grant funds and approves new mandates and set-asides. For example, Congress gave states broad latitude to administer the Temporary Assistance for Needy Families (TANF) welfare-reform block grant when this block grant replaced the old Aid to Families with Dependent Children program in 1996. But now, over the bipartisan opposition of governors, the Administration is pressing to impose new program requirements and restrictions that would force many states to jettison program reforms they have adopted and to conform instead to Administration priorities. The House of Representatives has passed legislation approving these new federal requirements.
The Administration’s proposals for increased flexibility in the voucher program appear intended in no small part to reduce the intensity of PHA opposition to substantial funding cuts. There is significant risk that if the Administration’s block-grant proposal is approved, PHAs may see their newfound flexibility eroded in subsequent years, while remaining stuck with substantially reduced funding levels.
PHAs that administer the housing voucher program have long sought broader administrative flexibility and simplification of certain program rules. It is possible to provide PHAs with greater flexibility or simplified rules in many areas — such as inspections, recertifications, reporting, homeownership, project-basing, and calculation of rents — while retaining the voucher program’s basic structure. Changes that increase the efficiency of the program without compromising its effectiveness in serving needy families would be in the interest of all parties with a stake in the program, including PHAs, tenants, families on waiting lists, landlords, local communities and federal taxpayers.
The Administration’s proposals would make radical changes in the voucher program that are largely unrelated to expanding administrative flexibility and would run counter to the interests of PHAs and the families and communities they serve. The sharp voucher funding reduction the Administration is seeking in 2005 would force PHAs to make substantial cuts in both the assistance they provide to low-income families and their own administrative infrastructures. In this context, the flexibility that PHAs would receive would largely be flexibility to make — and take responsibility for — controversial decisions about what cuts to institute in the program.
Moreover, these difficult cuts would not be a one-time event that PHAs could simply try to weather in 2005. The conversion of the voucher program would likely pave the way for further funding reductions of substantial depth in years after 2005. The cuts that would be made in the voucher program by 2009, under the Administration’s budget, would represent the deepest cuts in a low-income program since the early years of the Reagan Administration. PHAs would be forced to take painful actions to shrink their programs over and over again.
Administration budget documents include funding levels for years after 2005 for the Section 8 program as a whole, but not separate levels for the voucher and project-based components of Section 8. The 2009 voucher funding level shown in this chart was determined by subtracting a projected funding level for the project-based Section 8 program in 2009 from the total Section 8 level that the budget documents show for 2009. The funding level for the renewal of project-based units in 2009 was projected by using an estimate of the number of units that will need to be renewed in 2009. (This estimate was developed by adding the number of units requiring renewal in 2005, as reported by HUD, to the number of additional units expiring between 2005 and 2009.) We multiplied the estimate of the number of project-based units requiring renewal in 2009 by an estimate of the average cost per unit for project-based units, which was derived by applying the rate of inflation in project-based costs, as estimated by CBO, to the weighted average of HUD’s estimates of the per-unit cost of various categories of project-based units in 2005.
Total project-based funding for any year also includes funding for contract administrators. We assumed that $275 million would be provided for contract administrators in 2009. This is HUD’s estimate of the level of contract administrator “program activity” in 2005. That level in fact could increase or decrease by 2009.
We subtracted the projected funding for the project-based program in 2009 from the total amount of Section 8 funding the Administration’s budget shows for 2009, and thereby produced the estimated funding level for the voucher program in 2009. In so doing, we did not subtract any funding for tenant-protection vouchers. Any funding needed for such vouchers would further reduce the resources available for voucher renewals. Finally, the estimate of the 2009 voucher funding level assumes no funds appropriated in prior years will remain available in 2009 to support voucher renewals.
 HUD Fiscal Year 2005 Congressional
BudgetJustifications, Table 4. The Administration’s modeling of cost savings assumes that 12.5 percent of vouchers will become available in 2005 to be issued to new households, as current families leave the program. The HUD scenario that assumes the most rapid shift of vouchers from poorer to higher-income households appears to conclude that $609 million would be saved in a single year if all of the vouchers that became available when extremely low-income households leave the program are reissued to families with higher incomes. It appears, however, that the model unrealistically assumes that all of the vouchers that become available through turnover in a year become available at the start of the year. If, as is more likely, households leave the program at a relatively even rate over the course of the year, $305 million would be saved under this scenario. That would close less than one-fifth of the more-than-$1.6 billion shortfall in 2005 that the Administration’s budget request would create.
 PHAs that have reserves remaining in fiscal year 2005 would likely deplete their reserves before terminating assistance to current families. It is unclear, however, if any agencies will have reserve funds remaining in 2005. New funding rules adopted by Congress for fiscal year 2004 require PHAs to make much greater use of their one-month reserves than previously. (See “Omnibus Bill Appropriates Sufficient Funding to Renew
HousingVouchers,” available on the internet at http://www.centeronbudget.org/12-24-03hous.htm.) It appears that the Administration does not intend to replenish reserves used in 2004 unless Congress requires it to do so. Moreover, documents that HUD has submitted to Congress indicate that HUD intends to recapture $300 million of PHA program reserves in fiscal year 2005 as part of the rescission and reappropriation of $1.556 billion included in the proposed funding for the HousingCertificate Fund. If this occurred, the funds in program reserves would be further reduced (and perhaps eliminated entirely).
Housingagencies also would permanently lose the right to obtain additional funding to support vouchers that are not in use at the end of fiscal year 2004 but that agencies might be able to put to use later.
 The gap is even larger if one compares OMB’s estimate of Section 8 expenditures in 2009 under the Administration’s budget proposal to the CBO estimate of the level of expenditures that will be needed in 2009 to maintain today’s levels of assistance.
 Administrative funding in 2004 may make up a lower or higher percentage of actual subsidy costs, since HUD may not use up all subsidy funds or, more likely, state and local housing agencies may use reserve funds appropriated in prior years to increase subsidy funding somewhat above the level appropriated by Congress in 2004.
 The Center will issue shortly an analysis examining the impact of the Administration’s proposal on the Family Self-Sufficiency program.