This is the accessible text file for GAO report number GAO-12-300 
entitled 'Housing Choice Vouchers: Options Exist to Increase Program 
Efficiencies' which was released on April 18, 2012. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

March 2012: 

Housing Choice Vouchers: 

Options Exist to Increase Program Efficiencies: 

GAO-12-300: 

GAO Highlights: 

Highlights of GAO-12-300, a report to congressional requesters. 

Why GAO Did This Study: 

The Department of Housing and Urban Development’s (HUD) Housing Choice 
Voucher (voucher) program subsidizes private-market rents for 
approximately 2 million low-income households. HUD pays a subsidy that 
generally is equal to the difference between the unit’s rent and 30 
percent of the household’s income. HUD also pays an administrative 
fee, based on a formula, to more than 2,400 local housing agencies to 
manage the program. Over time, program expenditures steadily have 
risen, causing some to question how well HUD managed costs and used 
program resources. This report (1) discusses the key drivers of cost 
growth in the voucher program and the actions taken to control this 
growth and (2) analyzes various options to cut costs or create 
efficiencies. For this report, GAO analyzed HUD data; reviewed budget 
documents, program laws and regulations, guidance, academic and 
industry studies, and GAO reports; and interviewed officials from HUD, 
industry groups, and 93 housing agencies. 

What GAO Found: 

Several factors—-including rising rents, declining household incomes, 
and decisions to expand the number of assisted households-—were key 
drivers of the approximately 29 percent increase (before inflation) in 
housing agencies’ expenditures for the voucher program between 2003 
and 2010. Congress and HUD have taken steps to limit cost increases 
while maintaining assistance for existing program participants. For 
example, Congress moved away from providing funding to housing 
agencies based on the number of voucher-assisted households they were 
authorized to subsidize and instead provided funding based on the 
generally lower number of voucher-assisted households housing agencies 
actually subsidized in the prior year. Further, HUD has proposed 
administrative relief and program flexibility for housing agencies, 
including streamlining program requirements and reducing subsidies 
paid. 

GAO identified several additional options that if implemented 
effectively, could substantially reduce the need for new 
appropriations, cut costs (expenditures), or increase the number of 
households assisted. 

* Reduce housing agencies subsidy reserves. Housing agencies have 
accumulated approximately $1.8 billion in subsidy reserves (unspent 
funds). They can hold the funds in reserve or spend them on authorized 
program expenses in future years. Over time, large sums can 
accumulate. Although HUD has sought the authority to offset (reduce) 
its future budget requests by the amount of “ excess” subsidy 
reserves, it has not provided Congress with complete or consistent 
information on how much of these reserve funds housing agencies should 
retain for contingencies. GAO has highlighted the importance of 
providing clear and consistent information on housing agencies’ 
reserves to Congress so it can make informed funding decisions. 

* Implement administrative reform. HUD officials have noted that 
certain requirements for administering the voucher program are 
burdensome and costly and could be streamlined. In addition, the 
formula HUD uses to pay administrative fees to housing agencies is not 
tied to current administrative costs or requirements. HUD has an 
administrative fee study underway, which intends to identify specific 
reforms to ease administrative burden, increase efficiencies, and 
suggest ways to align the administrative fee formula with the 
functions housing agencies perform. Without this information, Congress 
may not have the information necessary to make fully informed policy 
and funding decisions related to the voucher program. 

* Implement rent reform and consolidate voucher administration. Rent 
reform (for example, reducing subsidies by requiring households to pay 
more toward rent) and consolidation of program administration under 
fewer housing agencies could yield substantial cost savings—-
approaching $2 billion—-or allow housing agencies to serve additional 
households, provided annual savings were reinvested in the program. 
However, while these options may have some advantages over the current 
program structure, they would require policymakers to consider some 
potential trade-offs, including increased rent burdens for low-income 
households, increased concentration of assisted households in high 
poverty areas, and more limited local control over voucher programs. 

What GAO Recommends: 

GAO identifies options for increasing efficiencies and recommends that 
HUD (1) determine what level of reserve funding housing agencies 
should maintain and reduce future budget requests by the amount of 
excess reserves and (2) consider proposing options for simplifying 
program administration and changes to the administrative fee formula. 
HUD did not agree or disagree with the recommendations. While it noted 
that the draft provided an accurate assessment, it offered some 
clarifications and responses. 

View [hyperlink, http://www.gao.gov/products/GAO-12-300]. For more 
information, contact Mathew J. Scirè at (202) 512-8678 or 
sciremj@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Several Factors Contributed to Cost Increases in the Voucher Program 
from 2003 through 2010, and Congress and HUD Made Efforts to Limit 
Them: 

Options Exist to Reduce Program Costs or Increase Efficiencies: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: Objectives, Scope, And Methodology: 

Appendix II: Comment From U.S. Department Of Housing And Urban 
Development: 

Appendix III: GAO Contact And Staff Acknowledgment: 

Tables: 

Table 1: Primary Examples of Special-Purpose Vouchers: 

Table 2: Annual Appropriations and HUD Outlays for the Voucher 
Program, from Fiscal Years 2005 through 2011: 

Table 3: Percentage of Expenditure Increases Resulting from Changes in 
Subsidy Costs for Existing Vouchers, Subsidy Costs for New Vouchers, 
and Administrative Fees, 2003 to 2010: 

Table 4: Summary of Fiscal Year 2008 and 2009 Subsidy Reserve 
Rescissions and Offsets: 

Table 5: HUD's Proposed Voucher Program Reforms: 

Table 6: Summary of Subsidy Reserve Rescissions and Offsets and the 
Allocation of Supplemental Funding, Fiscal Years 2008 and 2009: 

Table 7: Administrative Streamlining Efforts Implemented or Proposed 
at MTW Agencies: 

Table 8: Estimated Annual Effect of Selected Rent Reform Options on 
Program Costs and Number of Voucher-Assisted Households Served: 

Table 9: Estimated Number and Percentage of Voucher-Assisted 
Households for Which Rents Would Increase and the Average Monthly 
Increase, by Rent Reform Option: 

Table 10: Comparison of Two-and Three-Bedroom Monthly Rents under a 35 
Percent of Fair Market Rent Structure, for Three Areas: 

Figures: 

Figure 1: Annual Expenditures of Disbursed Funds by Housing Agencies 
for the Voucher Program, from 2003 through 2010: 

Figure 2: Median Gross Rents (2011 Constant Dollars) for Units Leased 
by Voucher-Assisted Households, from 2003 through 2010: 

Figure 3: Median Annual Household Income (2011 Constant Dollars) of 
Voucher-Assisted Households, from 2003 through 2010: 

Figure 4: Estimated Effect of Rent Reform Options on Monthly Rents of 
All Voucher-Assisted Households, by Household Type: 

Figure 5: Estimated Effect of Rent Reform Options on Monthly Rents of 
Affected Voucher-Assisted Households, by Household Type: 

Abbreviations: 

GPRA: Government Performance and Results Act of 1993: 

GPRAMA: Government Performance and Results Modernization Act of 2010: 

HUD: Department of Housing and Urban Development: 

HUDCAPS: HUD Central Accounting and Program System: 

MTW: Moving to Work: 

NRA: Net Restricted Assets: 

PIC: Public and Indian Housing Information System: 

VASH: Veteran Affairs Supportive Housing: 

VMS: Voucher Management System: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548: 

March 19, 2012: 

Congressional Requesters: 

Annually from fiscal years 2003 through 2010, the Department of 
Housing and Urban Development's (HUD) Housing Choice Voucher (voucher) 
program helped provide affordable rental housing to approximately 2 
million households with very or extremely low incomes.[Footnote 1] In 
the same period, program expenditures grew at an average annual rate 
of about 4 percent from $11.7 billion to $15.1 billion. The voucher 
program is unique among HUD's rental assistance programs in paying 
subsidies to landlords to help households rent units (apartments or 
houses) on the private market. The amount of subsidy HUD pays 
generally is equal to the difference between the unit's rent and 30 
percent of the household's income. While Congress and HUD have 
supported and enacted changes to the voucher program designed to limit 
growth in appropriations without reducing the number of households 
that received assistance, they also added new vouchers targeted at 
homeless veterans, nonelderly disabled households, and others. In the 
current constrained economic and budget environment, policymakers have 
questioned the consistent growth in voucher program costs and whether 
HUD and public housing agencies that administer the program adequately 
managed costs. 

In response to your request, this report discusses the factors that 
have affected costs in the voucher program from 2003 through 2011 and 
the actions Congress and HUD took to manage these costs. The report 
also identifies additional steps that could be taken to limit cost 
growth in the voucher program or more efficiently provide decent, 
safe, and affordable housing. 

To determine the factors that have affected costs in the voucher 
program from 2003 to 2011 and the actions Congress and HUD took to 
manage these costs, we reviewed and analyzed appropriations 
legislation, budget documents, and HUD's annual guidance on the 
allocation of the program's appropriation to housing agencies. The 
start year for our analysis reflects the year when Congress began 
changing the voucher program's funding formula.[Footnote 2] We 
analyzed program data that HUD prepared using information derived from 
multiple HUD systems including the Central Accounting and Program 
System (HUDCAPS) and Voucher Management System (VMS) to determine how 
much housing agencies' expenditures changed from 2003 through 2010. 
[Footnote 3] We also reviewed prior work by GAO and others to describe 
what is known about the cost-effectiveness and characteristics of 
vouchers relative to other forms of rental housing assistance. To 
identify additional steps that could be taken to limit cost growth in 
the voucher program and more effectively provide decent, safe, and 
affordable housing, we identified and reviewed relevant legislation, 
draft legislation, and studies. We analyzed HUD's VMS data on the 
balances of Net Restricted Assets (NRA or subsidy reserves) for 
housing agencies as of September 30, 2011, to determine the extent of 
such reserves for housing agencies. In addition, using data from HUD's 
Public and Indian Housing Information System (PIC) on household 
characteristics, income, and rents, we evaluated the cost and policy 
implications of three types of programmatic reform options for the 
voucher program: increasing minimum rents, changing the percent of 
income tenants pay toward rent, and requiring tenants to pay a 
percentage of fair market rent. In identifying and assessing these 
programmatic reform options, we reviewed proposals included in draft 
legislation and in HUD, Congressional Budget Office, and housing 
industry group reports. We also considered reforms certain agencies 
have implemented. In conducting our work, we assessed the reliability 
of datasets that HUD provided, including data files derived from 
HUDCAPS, VMS, and PIC and determined that the data were sufficiently 
reliable for the purposes of this report. Finally, for all of our 
objectives, we interviewed HUD officials and consulted with an 
academic and officials from various housing groups. We also contacted 
or visited 62 housing agencies that administer the voucher program and 
31 of the 35 housing agencies participating in the Moving to Work 
(MTW) demonstration program. See appendix I for a more detailed 
discussion of our scope and methodology. 

We conducted this performance audit from February 2011 through March 
2012 in accordance with generally accepted government auditing 
standards. Those standards require that we plan and perform the audit 
to obtain sufficient, appropriate evidence to provide a reasonable 
basis for our findings and conclusions based on our audit objectives. 
We believe that the evidence obtained provides a reasonable basis for 
our findings and conclusions based on our audit objectives. 

Background: 

The voucher program is not an entitlement program. As a result, the 
amount of budget authority HUD requests and Congress provides through 
the annual appropriations process limits the number of households that 
the program can assist.[Footnote 4] Historically, appropriations for 
the voucher program (or for other federal housing programs) have not 
been sufficient to assist all households that HUD has identified as 
having worst-case housing needs--that is, unassisted households with 
very low incomes that pay more than 50 percent of their incomes in 
rent, live in substandard housing, or both.[Footnote 5] In 2009, 41 
percent of the more than 17 million very low-income renters had worst-
case housing needs, according to HUD. The primary problem affecting 
these renters was rent burden--approximately 97 percent paid more than 
50 percent of their incomes in rent.[Footnote 6] 

To be eligible for assistance, in general, households must have very-
low incomes--not exceeding 50 percent of the area median income, as 
determined by HUD. Under the Quality Housing and Work Responsibility 
Act of 1998 (P.L. 105-276), at least 75 percent of new voucher program 
participants must have extremely low incomes--not exceeding 30 percent 
of the area median income.[Footnote 7] 

Under the voucher program, an assisted household pays 30 percent of 
its monthly adjusted income in rent; the remainder of the rent is paid 
through a HUD-subsidized "voucher," which generally is equal to the 
difference between (1) the lesser of the unit's gross rent (generally, 
rent plus utilities) or a local "payment standard" and (2) the 
household's payment. The payment standard is based on the HUD-
determined fair market rent for the locality, which generally equals 
the 40th percentile of market rents (including utilities) recent 
movers paid for standard-quality units. HUD annually estimates fair 
market rents for metropolitan and nonmetropolitan areas.[Footnote 8] 
Housing agencies--the state and local agencies that administer the 
voucher program on HUD's behalf--can set payment standards (that is, 
pay subsidies) between 90 percent and 110 percent of the fair market 
rent for their areas. By determining fair market rents and setting 
payment standards at a rate sufficient to provide acceptable choices 
for voucher program participants, HUD and housing agencies essentially 
set the upper and lower bounds on the cost of typical, standard-
quality units that voucher holders rent. Participants in the voucher 
program can choose to live in units with gross rents that are higher 
than the payment standard, but they must pay the full difference 
between the unit's gross rent and the payment standard, plus 30 
percent of their income.[Footnote 9] 

In 2011, more than 2,400 housing agencies administered more than 2.2 
million vouchers--their programs ranged in size from more than 96,000 
vouchers to fewer than 5. Housing agencies are responsible for 
inspecting units, ensuring that rents are reasonable, determining 
households' eligibility, calculating and periodically re-determining 
households' incomes and rental payments, and making subsidy payments 
to landlords.[Footnote 10] In addition, housing agencies perform basic 
program functions, such as establishing and maintaining a waiting 
list, processing tenant moves, conducting landlord and tenant 
outreach, and reporting to HUD. HUD disburses appropriated funds to 
housing agencies for subsidy payments to landlords and administrative 
expenses. 

Each year, Congress appropriates funding for subsidies for renewal 
(existing) and incremental (new) vouchers and administrative expenses. 
As part of the appropriations process, Congress outlines a formula 
that determines the amount of renewal funding for which housing 
agencies are eligible ("eligible amount"). However, the amount 
Congress appropriates to the voucher program may not equal the total 
amount for which housing agencies are eligible under the formula. HUD 
is responsible for allocating program funding ("appropriated amount") 
among housing agencies based on their eligible amounts. To the extent 
that the appropriated amount does not fully fund housing agencies at 
their eligible amounts, HUD reduces the funding housing agencies 
receive to fit within the appropriated amount. Housing agencies are 
expected to use all the subsidy funding HUD allocates for authorized 
program expenses (including subsidy and utility payments). However, if 
housing agencies' allocated amounts exceed the total cost of their 
program expenses in a given year, they must maintain their unused 
subsidy funds in NRA (reserve) accounts. Housing agencies may use 
their NRA balances (subsidy reserves) to pay for authorized program 
activities in subsequent years.[Footnote 11] 

Incremental vouchers include various special-purpose vouchers. 
Congress appropriates funding for these vouchers in separate line 
items in the budget, which distinguish them from renewal vouchers. 
Housing agencies must apply to HUD to receive allocations of and 
funding for the special-purpose vouchers, which, as described in table 
1, include Enhanced, Tenant Protection, Family Unification Program, 
Mainstream, Nonelderly Disabled, and Veteran Affairs Supportive 
Housing vouchers. These vouchers may have different or additional 
eligibility and operational requirements than renewal vouchers. After 
the first year, special-purpose vouchers generally become renewal 
vouchers for purposes of determining funding eligibility in the next 
year, but HUD may require that housing agencies separately track them 
as special-purpose vouchers. 

Table 1: Primary Examples of Special-Purpose Vouchers: 

Type of special-purpose voucher: Enhanced; 
Description: Enhanced vouchers are available to tenants facing a 
housing conversion action, including owner opt-outs of Section 8 
project-based contracts, mortgage prepayments, or voluntary 
terminations of the mortgage insurance associated with the 
preservation-eligible property. By statute, HUD must offer enhanced 
vouchers to families affected by housing conversions. 

Type of special-purpose voucher: Tenant Protection; 
Description: Tenant protection vouchers subsidize rents for tenants 
facing a housing conversion action or HUD enforcement actions against 
owners not covered by enhanced vouchers, including terminations or non-
renewals of Section 8 project-based contracts, sales or foreclosures 
of HUD-subsidized mortgages, or demolitions/dispositions of public 
housing. 

Type of special-purpose voucher: Family Unification Program; 
Description: These vouchers are available to families for whom the 
lack of adequate housing is a primary factor in the separation, or 
threat of imminent separation of their children, or in the delay of 
the discharge of children to the family from out-of-home care. The 
vouchers also are available to youths aged 18 to 21 who left foster 
care at age 16 or older and who lack adequate housing. 

Type of special-purpose voucher: Mainstream; 
Description: Mainstream vouchers assist families that include a person 
with disabilities who faces difficulties in locating suitable and 
accessible housing in the private market. 

Type of special-purpose voucher: Nonelderly Disabled; 
Description: These vouchers are designed to provide choices to 
nonelderly disabled residents in their transition out of elderly-
designated public housing or care-giving institutions or developments 
with preferences for elderly tenants. 

Type of special-purpose voucher: Veteran Affairs Supportive Housing 
(VASH); 
Description: The VASH program combines HUD voucher rental assistance 
for homeless veterans with case management and clinical services from 
the Department of Veterans Affairs. 

Source: HUD. 

[End of table] 

Congress appropriates funding for administrative fees, and the formula 
used to calculate the administrative fee generally is based on fair 
market rents, adjusted annually to reflect changes in wage rates. HUD 
pays fees to housing agencies based on the number of units leased 
(vouchers used) as of the first of each month. HUD pays one (higher) 
rate for the first 600 units under lease and a second (lower) rate for 
the remaining units. As with subsidy funding, if the appropriated 
amount does not fully cover housing agencies' fees as determined by 
the formula, HUD will reduce the amount of funding each housing agency 
receives to fit within the appropriated amount. Since fiscal year 
2006, administrative fees have accounted for less than 10 percent of 
total voucher program funding. 

Some housing agencies that administer vouchers can participate in and 
receive funding under MTW, a demonstration program authorized by 
Congress in 1996 and implemented by HUD in 1999.[Footnote 12] MTW 
allows participating housing agencies to test locally designed housing 
and self-sufficiency initiatives in the voucher and other federal 
housing programs. Housing agencies may waive certain statutes and HUD 
regulations to achieve three objectives: (1) reduce cost and achieve 
cost-effectiveness in federal expenditures; (2) give incentives to 
families with children whose heads of household are working, seeking 
work, or in job training, educational or other programs that assist in 
obtaining employment and becoming economically self-sufficient; and 
(3) increase housing choices for low-income families. MTW agencies 
also have "funding flexibility"--they may use their program-related 
funding (including voucher subsidy funding) and administrative fees 
for any purpose (programmatic or administrative).[Footnote 13] 
Currently, 35 housing agencies participate in MTW--according to HUD, 
they administer about 15 percent of all vouchers and account for 
approximately 16 percent of all subsidy and administrative fee funding 
on an annual basis. Congress and HUD fund MTW agencies pursuant to 
their MTW agreements; however, the agencies could have subsidies and 
administrative fees reduced if the amounts Congress appropriated were 
less than the housing agencies' eligible amounts under the formulas. 

Several Factors Contributed to Cost Increases in the Voucher Program 
from 2003 through 2010, and Congress and Hudmade Efforts to Limit Them: 

Several factors affected voucher program costs (as measured through 
congressional appropriations, HUD outlays, and housing agencies' 
expenditures) and in some cases contributed to cost increases from 
2003 through 2010, including: (1) increases in subsidy costs for 
existing vouchers, (2) subsidy costs for new vouchers, and (3) 
administrative fees paid to housing agencies. In addition to these 
factors, the design and goals of the voucher program, such as 
requirements to target assistance to certain households, contributed 
to overall program costs. Despite increases in the cost of the program 
from 2003 through 2010, our work and other published studies have 
found that vouchers generally have been more cost-effective in 
providing housing assistance than federal housing production programs 
designed to add to or rehabilitate the low-income housing stock. In 
addition, Congress and HUD have taken several steps to manage cost 
increases over the period. 

Subsidies for Existing and New Vouchers and Increases in 
Administrative Fees Contributed to Cost Increases in the Voucher 
Program from 2003 through 2010: 

Several factors affected increases in congressional appropriations, 
HUD outlays, and housing agencies' expenditures in the voucher program 
from 2003 through 2010. As shown in table 2, from fiscal years 2005 
through 2011, voucher program appropriations increased from 
approximately $14.8 billion to $18.4 billion (approximately 4 percent 
annually). Over the same period, outlays--funding HUD disburses to 
housing agencies for program expenses--increased from $10 billion to 
$18.6 billion (approximately 11 percent annually). Information on 
appropriations and outlays for the voucher program were not available 
for 2003 and 2004 because HUD did not report this information 
separately from other rental assistance programs. 

Table 2: Annual Appropriations and HUD Outlays for the Voucher 
Program, from Fiscal Years 2005 through 2011: 

Fiscal year[A]: 2005; 
Appropriation[B]: $14.8 billion; 
HUD Outlay: $10.0 billion. 

Fiscal year[A]: 2006; 
Appropriation[B]: $15.4 billion; 
HUD Outlay: $13.0 billion. 

Fiscal year[A]: 2007; 
Appropriation[B]: $15.9 billion; 
HUD Outlay: $16.0 billion. 

Fiscal year[A]: 2008; 
Appropriation[B]: $16.4 billion; 
HUD Outlay: $15.7 billion. 

Fiscal year[A]: 2009; 
Appropriation[B]: $16.8 billion; 
HUD Outlay: $16.0 billion. 

Fiscal year[A]: 2010; 
Appropriation[B]: $18.2 billion; 
HUD Outlay: $18.0 billion. 

Fiscal year[A]: 2011; 
Appropriation[B]: $18.4 billion; 
HUD Outlay: $18.6 billion. 

Source: GAO analysis of budget data. 

[A] In 2005 Congress created a specific budget account (Tenant-based 
Rental Assistance) for the voucher program. Prior to this, Congress 
provided a combined appropriation for both the voucher and Project-
based Section 8 programs through the Housing Certificate Fund. As a 
result, it is not possible to distinguish the voucher program's 
appropriations and outlays from those of the Project-based Section 8 
programs prior to 2005. 

[B] Appropriations include supplemental funding, carryovers, and 
rescissions. 

[End of table] 

Once disbursed, housing agencies expend program funds on activities 
such as making subsidy payments to landlords and for administrative 
expenses. As shown in figure 1, from 2003 through 2010, housing 
agencies' expenditures increased from approximately $11.7 billion to 
$15.1 billion (about 4 percent annually). Expenditure data for 2011 
were not available at the time we were conducting our review. HUD's 
outlays and housing agencies' expenditures can differ somewhat in any 
given year because of differences in the timing of payments and 
fluctuations in the rate of funding utilization--that is, some housing 
agencies may not use all of their apportioned funds and may build 
reserves. Later in this report we discuss the extent to which housing 
agencies have accumulated subsidy reserves and steps Congress and HUD 
could take to reduce future budget requests or reallocate the reserve 
funds. 

Figure 1: Annual Expenditures of Disbursed Funds by Housing Agencies 
for the Voucher Program, from 2003 through 2010: 

[Refer to PDF for image: line graph] 

Year: 2003; 
Expenditures: $10.57 billion. 

Year: 2004; 
Expenditures: $11.25 billion. 

Year: 2005; 
Expenditures: $11.00 billion. 

Year: 2006; 
Expenditures: $10.67 billion. 

Year: 2007; 
Expenditures: $11.18 billion. 

Year: 2008; 
Expenditures: $12.19 billion. 

Year: 2009; 
Expenditures: $13.09 billion. 

Year: 2010; 
Expenditures: $13.47 billion. 

Source: GAO analysis of HUD data. 

Note: Expenditure data were not available for 2011. 

[End of figure] 

As shown in table 3, housing agencies' expenditures increased by a 
total of about 28.9 percent in nominal dollars from 2003 through 2010. 
Once adjusted for inflation, housing agencies' expenditures increased 
by a smaller rate, approximately 8.8 percent. (We evaluated 
expenditure after adjusting for the general effects of inflation using 
a broad base index of price changes for all goods and services. We 
expressed expenditures in 2011 constant dollars, the latest year for 
which complete data on price changes are available.) In the sections 
below, we discuss how (1) increases in subsidy costs for existing 
vouchers, (2) subsidy costs for new vouchers, and (3) administrative 
fees paid each contributed to the nominal and constant dollar 
increases in voucher program costs from 2003 through 2010. 

Table 3: Percentage of Expenditure Increases Resulting from Changes in 
Subsidy Costs for Existing Vouchers, Subsidy Costs for New Vouchers, 
and Administrative Fees, 2003 to 2010: 

Percentage change in expenditures due to: Total; 
Nominal dollar change: 28.9%; 
Constant dollar change: 8.8%. 

Percentage change in expenditures due to: Increases in subsidy costs 
for existing vouchers; 
Nominal dollar change: 19.5%; 
Constant dollar change: 2.4%. 

Percentage change in expenditures due to: Subsidy costs for new 
vouchers; 
Nominal dollar change: 5.3%; 
Constant dollar change: 4.4%. 

Percentage change in expenditures due to: Administrative fees paid; 
Nominal dollar change: 4.1%; 
Constant dollar change: 2.0%. 

Source: GAO analysis of HUD data. 

Note: Nominal 2003 and 2010 dollars were converted to 2011 constant 
dollars using the Bureau of Labor Statistics' Consumer Price Index--
All Urban Consumers series. 

[End of table] 

Increases in Subsidy Costs for Existing Vouchers: 

As shown in table 3 above, in nominal terms, subsidy costs for 
existing vouchers grew by of 19.5 percent, accounting for a majority 
of the increase in housing agencies' expenditures from 2003 through 
2010. After adjusting for inflation, subsidy costs for existing 
vouchers grew by a small amount (2.4 percent) and were a smaller 
contributor to the total increase in expenditures. Two factors 
generally explain the remaining increase in expenditures for existing 
vouchers after adjusting for inflation--changes in market rents and 
household incomes. As previously discussed, the subsidies HUD and 
housing agencies pay to landlords on behalf of assisted households are 
based on market rents and household incomes. As a result, changes in 
market rents and household incomes affect subsidy cost. As shown in 
figure 2, in 2011 constant dollars, median gross rents for voucher-
assisted households increased from about $850 to $880 (or 4 percent) 
over the period. Growth in rents outpaced the rate of general 
inflation. As rents increase, HUD and housing agencies must pay larger 
subsidies to cover the increases, assuming no changes to household 
incomes or contributions to rent. 

Figure 2: Median Gross Rents (2011 Constant Dollars) for Units Leased 
by Voucher-Assisted Households, from 2003 through 2010: 

[Refer to PDF for image: line graph] 

Year: 2003; 
Median gross rent: $848. 

Year: 2004; 
Median gross rent: $855. 

Year: 2005; 
Median gross rent: $847. 

Year: 2006; 
Median gross rent: $845. 

Year: 2007; 
Median gross rent: $853. 

Year: 2008; 
Median gross rent: $853. 

Year: 2009; 
Median gross rent: $886. 

Year: 2010; 
Median gross rent: $879. 

Source: GAO analysis of HUD data. 

Note: Nominal dollars were converted to 2011 constant dollars using 
the Bureau of Labor Statistics' Consumer Price Index--All Urban 
Consumers series. 

[End of figure] 

Housing agencies we contacted reported that this increase in rental 
prices can be explained, in part, by increased demand and competition 
for affordable housing--for example, some noted that the number of 
renters has increased as a result of an increase in the number of 
foreclosures in recent years. National vacancy rates--an indicator of 
the relative tightness of the rental market--decreased from 2009 to 
2010. 

Further, as figure 3 shows, in 2011 constant dollars, the median 
annual income of voucher-assisted households contracted from about 
$11,000 to $10,700 (a decrease of about 3 percent) from 2003 through 
2010. Over the period, incomes of assisted households did not keep 
pace with the rate of general inflation. As incomes decline, voucher-
assisted households are paying less towards rent, requiring larger 
subsidies to cover the difference between the rents and tenant 
payments. 

Figure 3: Median Annual Household Income (2011 Constant Dollars) of 
Voucher-Assisted Households, from 2003 through 2010: 

[Refer to PDF for image: line graph] 

Year: 2003; 
Median annual household income: $11,006. 

Year: 2004; 
Median annual household income: $11,033. 

Year: 2005; 
Median annual household income: $11,101. 

Year: 2006; 
Median annual household income: $11,150. 

Year: 2007; 
Median annual household income: $11,156. 

Year: 2008; 
Median annual household income: $10,947. 

Year: 2009; 
Median annual household income: $11,111. 

Year: 2010; 
Median annual household income: $10,702. 

Source: GAO analysis of HUD data. 

Note: Nominal dollars were converted to 2011 constant dollars using 
the Bureau of Labor Statistics' Consumer Price Index--All Urban 
Consumers series. 

[End of figure] 

More than half of the housing agencies we contacted reported that job 
loss and wage reductions contributed to in their subsidy costs over 
the period of our analysis. One housing agency in California we 
contacted also reported that state cuts to social welfare programs, 
including those that provide direct cash assistance, lowered incomes 
for some households and therefore have increased subsidy costs. HUD 
estimated that reductions in federal welfare and disability income 
payments have resulted in monthly subsidy payment increases of $17 and 
$5, respectively, for households that receive those forms of 
assistance. 

Subsidy Costs for New Vouchers: 

The increase in the number of households assisted with vouchers (that 
is, subsidy costs for new vouchers) from 2003 through 2010 was another 
important contributor to the program's rising costs. As table 3 shows, 
in nominal dollars, subsidy costs for new vouchers grew by 5.3 percent 
over the period. After adjusting for inflation, the addition of new 
vouchers grew by 4.4 percent, accounting for half of the overall 
increase in housing agencies' expenditures over the period. Congress 
increased the size of the program through the addition of special-
purpose vouchers such as Enhanced, Tenant Protection, Family 
Unification Program, Mainstream, Nonelderly Disabled, and Veteran 
Affairs Supportive Housing (see table 1 for a description of each of 
these types of vouchers). HUD was unable to provide the data necessary 
to determine the extent to which each type of voucher contributed to 
the growth in program expenditures during this period. 

Changes in Administrative Fees Paid to Housing Agencies: 

Finally, growth in the fees paid to housing agencies to administer the 
voucher program grew about 4.1 percent in nominal dollars from 2003 
through 2010 (see table 3). In constant dollar terms, administrative 
fees grew by 2 percent over the period. The formula HUD uses to pay 
administrative fees to housing agencies is not directly tied to the 
cost of performing the administrative tasks the program requires. 
Moreover, the fees HUD has paid housing agencies in recent years have 
been less than the amount for which they were eligible under the 
formula because of reductions in appropriations. Housing agencies we 
contacted noted that the cost of doing business increased over the 
period of our analysis. For example, several noted that inspection 
costs have increased with the growing cost of gasoline, especially for 
housing agencies that administer vouchers over large geographic areas. 
Others noted that policies such as portability--the ability of voucher 
holders to use their vouchers outside of the jurisdiction of the 
housing agency that issued the voucher--increased staff costs because 
they have been increasingly complex and difficult to implement and 
monitor. Representatives of housing agencies with whom we spoke said 
that they have frozen salaries and hiring and increased staff hours, 
among other things, to cope with reductions in administrative fees. 

Program Design and Goals Also Influence Program Costs: 

The design and goals of the voucher program contribute to the overall 
expense of the voucher program, although it is difficult to quantify 
how much of the cost increase from 2003 through 2010 was due to design 
issues. Specifically, the voucher program has various features that 
are intended to target or give priority to the poorest households, and 
serving these households requires greater subsidies. Long-standing 
federal policy generally has required household contributions to rent 
to be based on a fixed percentage of household income, which can be 
reduced through income exclusions and deductions for certain expenses, 
such as child care and health services.[Footnote 14] Further, housing 
agencies are required to lease 75 percent of their new vouchers to 
extremely low-income households. In addition, housing agencies also 
may establish local preferences for selecting applicants from their 
waiting lists.[Footnote 15] Like the income standards and targeting 
requirements, these preferences have a direct impact on subsidy costs--
for example, the Boston Housing Authority has established preferences 
designed to assist "hard-to-house" individuals and families, including 
those experiencing homelessness. According to housing agency 
officials, because these individuals and families have little to no 
income, the agency's annual per-unit subsidy costs are higher than 
they would be absent the preferences. While these types of 
requirements help address Congress's and HUD's goal of serving the 
neediest households, HUD officials noted that such requirements make 
the program more expensive than it would otherwise be if housing 
agencies had more flexibility to serve households with a range of 
incomes. 

Similarly, program goals, such as HUD's deconcentration policy also 
can affect program costs. Specifically, this policy encourages 
assisted households to rent units in low-poverty neighborhoods, which 
typically are more expensive. According to HUD officials, the 
deconcentration goal increases subsidy costs for housing agencies and 
overall costs for the department because, as previously discussed, if 
rents increase, but household contributions to rent remain constant, 
HUD and housing agencies must make up for the increased rent burden in 
the form of higher subsidy payments. 

Although Costs Have Risen, the Voucher Program Generally Has Been More 
Cost-Effective Than Other Types of Housing Assistance: 

Despite increases in the cost of the voucher program from 2003 through 
2010, our work and other published studies consistently have found 
that vouchers generally have been more cost-effective in providing 
housing assistance than federal housing production programs designed 
to add to or rehabilitate the low-income housing stock.[Footnote 16] 
Our 2002 report provides the most recent original estimates of the 
cost differences between the voucher program and certain existing 
production programs.[Footnote 17] We estimated that, for units with 
the same number of bedrooms in the same general location, the 
production programs cost more than housing vouchers. In metropolitan 
areas, the average total 30-year costs of the production programs 
ranged from 8 to 19 percent greater for one-bedroom units. For two-
bedroom units, the average total 30-year costs ranged from 6 percent 
to 14 percent greater.[Footnote 18] The cost advantage of the voucher 
over the production programs was likely understated because other 
subsidies--such as property tax abatements--and potential underfunding 
of reserves to cover expected capital improvements over the 30-year 
cost period were not reflected in the cost estimates for the 
production programs. 

Much of the research over the past several decades reached similar 
conclusions. For example, in 2000, HUD found that average ongoing 
costs per occupied unit of public housing were 8 to 19 percent higher 
than voucher subsidy costs. In 1982, the President's Commission on 
Housing found that subsidy costs for new construction were almost 
twice as much as subsidy costs for existing housing.[Footnote 19] The 
commission's finding set the stage for the eventual shift from 
production programs to vouchers as the primary means through which the 
federal government provides rental housing assistance. 

Notwithstanding the cost-effectiveness of vouchers relative to other 
forms of rental housing assistance, many of these studies noted the 
benefits that production programs can and have conferred on low-income 
households and communities such as supportive services for the elderly 
and persons with disabilities. The voucher program typically does not 
confer such benefits. In addition, research has indicated that some 
markets may have structural issues. For example, regulatory 
restrictions that reduce the supply of housing (and thus, 
opportunities for households to use vouchers) make production programs 
more effective tools for providing affordable housing than vouchers in 
those locations.[Footnote 20] And our work found that voucher holders 
sometimes are unsuccessful in using their vouchers, either because 
they cannot find units that meet their needs or because landlords are 
unwilling to accept their vouchers. These households may benefit more 
from production programs, which can better guarantee access to 
affordable housing, than vouchers. 

Congress and HUD Have Taken Steps to Limit Cost Increases in the 
Voucher Program: 

In light of increasing program costs, Congress and HUD have taken 
several steps to limit the extent of increases from fiscal years 2003 
through 2011, while maintaining assistance for existing program 
participants. These steps include legislative changes to the formula 
HUD uses to calculate and distribute subsidy funding to housing 
agencies, as well as continued efforts to reduce improper rental 
assistance payments. 

Congressional Actions: 

Before fiscal year 2003, Congress and HUD funded housing agencies' 
renewal needs based on their average per-voucher costs from the 
previous year, adjusted for inflation, and multiplied by the number of 
authorized vouchers.[Footnote 21] Meaning, housing agencies received 
funding for all of their authorized vouchers, regardless of whether 
they leased all of those vouchers. In addition, prior to 2003, HUD 
provided each housing agency with reserve funding equal to one month 
of its subsidy funding--housing agencies could use their reserves to 
fund new vouchers (a practiced called "maximized leasing").[Footnote 
22] 

Beginning in fiscal year 2003, Congress changed the voucher program's 
funding formula so that it would provide housing agencies with renewal 
funding that was tied to housing agencies' actual costs and leasing 
rates rather than the number of authorized vouchers (whether used or 
unused). 

* Starting in fiscal year 2003, Congress stopped providing funding for 
vouchers that housing agencies issued in excess of their authorized 
levels, thus prohibiting over-(or maximized) leasing.[Footnote 23] 

* Congress generally based voucher program appropriations for fiscal 
year 2003 and thereafter on the number of leased vouchers (not to 
exceed authorized levels) and actual cost data that housing agencies 
reported to HUD. 

* Congress discontinued the practice of providing reserve funding for 
housing agencies and instead started reserving a portion of renewal 
funding to make adjustments to housing agencies' allocations for 
contingencies such as increased leasing rates or certain unforeseen 
costs. 

* In more recent years, Congress has provided HUD appropriations that 
did not fully fund housing agencies at their eligible amounts under 
the funding formula. 

* In every year since 2004, Congress has provided administrative fees 
that were at least 6 percent lower than the 2003 rate. 

* Finally, as shown in table 4, in fiscal years 2008 and 2009, 
Congress rescinded a portion of housing agencies' subsidy reserves and 
directed HUD, in total, to offset almost $1.5 billion from 1,605 
housing agencies).[Footnote 24] 

Table 4: Summary of Fiscal Year 2008 and 2009 Subsidy Reserve 
Rescissions and Offsets: 

Rescission and offset of excess subsidy reserves: 2008; 
Number of affected housing agencies: 1,197; 
Dollar total: $723,167,604. 

Rescission and offset of excess subsidy reserves: 2009; 
Number of affected housing agencies: 975; 
Dollar total: $750,000,000. 

Rescission and offset of excess subsidy reserves: Total; 
Number of affected housing agencies: 1,605[A]; 
Dollar total: $1,473,167,604. 

Source: GAO analysis of HUD data. 

[A] Of these, 567 housing agencies faced rescissions and offsets in 
both 2008 and 2009. 

[End of table] 

HUD Actions on Improper Payments: 

HUD has taken steps to reduce improper payments in the voucher 
program. According to HUD reports, the department has reduced gross 
improper payments (subsidy over-and underpayments) resulting from 
program administrator errors (that is, a housing agency's failure to 
properly apply income exclusions and deductions and correctly 
determine income, rent, and subsidy levels) by almost 60 percent, from 
$1.1 billion in fiscal year 2000 to $440 million in fiscal year 2009. 
[Footnote 25] 

In addition, HUD has provided housing agencies with fraud detection 
tools--such as the Enterprise Income Verification system, which makes 
income and wage data available to housing agencies--and realized 
continued reductions in improper payments as a result of these tools. 
According to HUD, from fiscal year 2006 through 2009, the department 
reduced gross improper payments resulting from errors in reported 
tenant income--including the tenant's failure to properly disclose all 
income sources--by approximately 37 percent, from $193 million to $121 
million. These efforts do not necessarily reduce the cost of assisting 
households, but they help increase the program's efficiency by helping 
ensure that an appropriate level of assistance is provided and 
potentially serving more households with appropriated funds. 

HUD-Proposed Reforms and Additional Actions to Manage Costs: 

HUD has requested the authority to implement program reforms that 
could have had the potential to decrease voucher program subsidy 
costs, administrative costs, or both. For example, as shown in table 
5, in its fiscal year 2012 budget request, HUD proposed implementing a 
rent demonstration to test alternatives to the current rent structure 
that could result in assisted households paying more in rent. As we 
discuss later in this report, changes to the way assisted household 
contributions to rent are calculated could result in cost savings to 
the program. 

Table 5: HUD's Proposed Voucher Program Reforms: 

Reform: Rent demonstration[A]; 
Budget year(s): 2012; 
Description: HUD sought the authority to test alternatives to the 
current rent structure with non-MTW agencies. 

Reform: Increased time between re-certifications of tenant income; 
Budget year(s): 2012; 
Description: HUD proposed extending the time between re-certifications 
of tenants with fixed incomes from 1 to 3 years. 

Reform: Transforming Rental Assistance[B]; 
Budget year(s): 2011; 
Description: HUD proposed streamlining and improving the delivery and 
oversight of rental assistance across all relevant HUD programs 
through the introduction of more efficient forms of administration 
such as consortiums, consolidation, and other locally designed 
structures. For example, a consortium of housing agencies could 
centralize administrative functions for a large area or for a state. 

Reform: Flexible Voucher Program; 
Budget year(s): 2005, 2006, 2007; 
Description: HUD proposed allowing state and local housing agencies to 
administer the voucher program and encouraging housing agencies with 
overlapping jurisdictions and those with small voucher programs to 
consolidate or enter cooperative agreements to promote administrative 
efficiencies and cut costs. 

Reform: Housing Assistance for Needy Families; 
Budget year(s): 2004; 
Description: Under this block grant, HUD proposed making state housing 
agencies responsible for the financial management and administration 
of the voucher program and giving states the option to directly 
administer the program or contract with local housing agencies or 
other public, nonprofit, or private entities to administer voucher 
assistance at the local level. 

Source: GAO analysis of HUD budget documents. 

[A] On December 6, 2011, HUD's Policy Development and Research issued 
a request for proposal for a separate rent demonstration. According to 
the request, the demonstration will test alternatives to the current 
rent structure using a random assignment experimental model and most 
likely would be undertaken at select MTW agencies because these 
agencies already have the authority to request waivers of voucher 
program laws and regulations. HUD is funding the demonstration under 
its Transformation Initiative. 

[B] Subsequent versions of the Transforming Rental Assistance proposal 
did not include the administrative efficiencies discussed in the table. 

[End of table] 

Although Congress did not grant HUD the authority to implement these 
voucher-related initiatives, HUD recently initiated administrative 
changes to its housing agency consortium rule, a first step in the 
effort to encourage housing agencies to consolidate as envisioned by 
the department's 2011 Transforming Rental Assistance proposal. The 
revised rule would treat participating housing agencies in a 
consortium as one entity. HUD's current regulation requires that 
consortium members be treated separately for oversight, reporting--as 
a result, few housing agencies have formed consortiums since 1998. 

Finally, in 2010, HUD began reviewing the administrative fee structure 
for the voucher program.[Footnote 26] The study aims to ascertain how 
much it costs a housing agency to run an efficient voucher program. 
HUD plans to use the results to help develop a new formula for 
allocating administrative fees. Although not enough time has passed to 
determine whether HUD's findings will positively or negatively affect 
costs in the voucher program, this study represents a positive effort 
on HUD's part to more clearly understand administrative costs in the 
voucher program and identify ways to improve efficiency. According to 
HUD officials, HUD intends to use this study as a basis for future 
legislative proposals, which could have implications for the cost of 
administering the program. 

Finally, in 2009, HUD developed a tool designed to help HUD staff and 
housing agencies forecast voucher and budget utilization (that is, the 
percentage of budget allocation and percentage of authorized vouchers 
they are using) for up to 3 years. Department officials credit the 
tool with increasing voucher program efficiency; however, HUD and 
housing agencies' use of the forecasting tool has not reduced overall 
costs in the voucher program. 

Options Exist to Reduce Program Costs or Increase Efficiencies: 

We identified several options that if implemented effectively, could 
reduce voucher program costs (by approximately of $2 billion annually, 
based on our estimates) or allow housing agencies to assist additional 
households if Congress chose to reinvest the costs savings in the 
program. First, improved information on the level of subsidy reserve 
funding housing agencies should maintain could aid budget decisions 
and reduce the need for new appropriations. ; Second, agency officials 
have noted that the voucher program's requirements are complex and 
burdensome and streamlining these requirements could reduce costs. 
Finally, changes to the calculation of voucher-assisted households' 
payments toward rent--known as rent reform--and consolidating voucher 
administration under fewer housing agencies' could also reduce program 
costs Each of these options would require congressional action to 
implement, and we discuss below possible steps that HUD could take to 
facilitate the implementation of some of them. Rent reform and 
administrative consolidation also involve difficult policy decisions 
that will affect some of the most vulnerable members of the population 
and alter long-standing program priorities and practices. 

Improved Information on Subsidy Reserves Could Aid Budget Decisions: 

Housing agencies have accumulated subsidy reserves (unspent funds) 
that Congress could use to (1) reduce program appropriations (through 
a rescission and offset) and potentially meet other federal needs or 
(2) direct HUD to assist additional households.[Footnote 27] As 
previously discussed, HUD allocates subsidy funding to housing 
agencies based on the formula Congress establishes in annual 
appropriations legislation. In recent years, the formula has specified 
that HUD allocate funds based on housing agencies' leasing rates and 
subsidy costs from the prior year. In any given year, housing agencies 
may under-lease or receive more funding than they can spend. Unless 
these funds are rescinded and offset, housing agencies can keep their 
unused subsidy funding in reserve accounts and spend these funds on 
authorized program expenses (including rent subsidies and utility 
allowance payments) in future years. Over time, large sums of money 
can accumulate. As of September 30, 2011, 2,200 housing agencies had 
more than $1.5 billion in subsidy reserves, which includes unspent 
subsidy funding from prior years and certain set-aside funding and 
funding for new vouchers where insufficient time has passed for 
expenditure. 

In addition, beginning in 2012, HUD implemented changes to how it 
disburses subsidy funds to housing agencies. As a result of these 
changes, although housing agencies may continue to accumulate subsidy 
reserves, HUD, rather than the housing agencies, holds these reserves. 
This change also will allow HUD to better determine the extent of the 
reserves housing agencies have accumulated.[Footnote 28] 

HUD officials told us that the department believes that it requires 
specific authority from Congress to reduce (offset) future voucher 
program budget requests by all or a portion of housing agencies' 
subsidy reserves.[Footnote 29] Although HUD provides quarterly reports 
to the Congressional Budget Office on the extent of housing agencies' 
reserves and has requested the authority to offset and in some cases, 
redistribute "excess" reserves (that is, reserves beyond what is 
needed to fund contingencies, such as cost increases from rising 
rental rates or falling tenant incomes, as defined by HUD), the 
department has not developed specific or consistent criteria defining 
what constitutes excess reserves or how it would redistribute funding 
among housing agencies. For example, in its fiscal year 2012 voucher 
program budget proposal, HUD requested the authority to offset excess 
reserves. According to the proposal, if given this authority, the 
department first would reallocate the funds to housing agencies to 
make up any difference between the appropriated amount and the total 
funding for which housing agencies were eligible based on the renewal 
formula and then redistribute any remaining funds to housing agencies 
based on "need and performance." However, the proposal did not specify 
how HUD would calculate excess subsidy reserves or a detailed 
methodology for redistributing the funds, and HUD officials 
acknowledged that redistributing excess funds among housing agencies 
will increase the size and the cost of the program over time because 
if housing agencies are able to lease more vouchers with these funds, 
Congress will have to appropriate more funding for renewal vouchers in 
subsequent years. 

Furthermore, HUD's definition for what constitutes excess reserves has 
varied. For example, HUD officials told us that housing agencies 
should retain one month (approximately 8.5 percent) of their annual 
funding allocations in reserves. However, in its fiscal year 2010 and 
2011 budget proposals, HUD defined excess reserves as those in excess 
of 4 and 6 percent, respectively, of housing agencies' allocated 
amounts.[Footnote 30] Further, HUD generally has excluded housing 
agencies with 250 and fewer vouchers from its proposed offsets. HUD 
officials told us that they have been considering lowering this 
threshold or developing a sliding scale methodology (generally based 
on size) to determine the amount of reserves housing agencies should 
maintain and the amount of excess reserves that HUD would propose 
offsetting and redistributing. 

In past work, we highlighted the importance of HUD's clearly 
identifying the existence and amount of unexpended subsidy funds 
(reserves) so that Congress could have confidence in the department's 
capacity to effectively manage the funding appropriated for the 
voucher program. We concluded that HUD should take steps to ensure 
that reserves did not reach unreasonable levels--that is, in excess of 
what is prudently needed to address contingencies.[Footnote 31] More 
recently, we stated that agency reporting about key areas such as 
financial management or program reforms should competently inform 
congressional decision making, and agencies should engage Congress 
about how to present this information.[Footnote 32] 

While a reserve for contingencies is prudent, without clear and 
consistent criteria for determining what constitutes an appropriate 
level for housing agency reserves, it is difficult to judge how well 
HUD managed the funding Congress has provided for the voucher program. 
For example, as previously discussed, in fiscal years 2008 and 2009 
Congress rescinded and directed HUD to offset excess subsidy reserves. 
However, as shown in table 6, the 2009 rescission and offset were too 
large for 288 (about 18 percent) of the 1,605 housing agencies that 
were subject to the 2008 and 2009 rescissions and offsets to absorb. 
Congress had to provide these 288 and an additional 152 housing 
agencies with supplemental funding to prevent the termination of 
voucher assistance. 

Table 6: Summary of Subsidy Reserve Rescissions and Offsets and the 
Allocation of Supplemental Funding, Fiscal Years 2008 and 2009: 

Housing agencies that had reserves rescinded and offset in 2008, 2009, 
or both years: 
Number of affected housing agencies: 1,605; 
Dollar total: $1,473,167,604. 

Housing agencies that received supplemental funding in 2009: 
Number of affected housing agencies: 440; 
Dollar total: $122,280,516. 

Housing agencies that had reserves rescinded and offset in 2008, 2009, 
or both years: 
Number of affected housing agencies: 288; 
Dollar total: $96,684,883. 

Housing agencies that did not have reserves rescinded and offset in 
either 2008 or 2009: 
Number of affected housing agencies: 152; 
Dollar total: $25,595,633. 

Source: GAO analysis of HUD data. 

Note: Of the 1,605 housing agencies that had reserves rescinded and 
offset, 567 housing agencies were had reserves rescinded and offset in 
both 2008 and 2009. 

[End of table] 

Similarly, in the fiscal year 2012 budget, Congress rescinded and 
directed HUD to offset housing agencies' subsidy reserves by $650 
million. Based on our analysis, as of September 30, 2011, housing 
agencies had approximately $606 million in excess reserves, 
approximately $44 million short of the $650 million rescission amount. 
Our analysis assumed that housing agencies retained in reserves the 
equivalent of one month or about 8.5 percent of their annual funding 
allocations--HUD's current thinking on the appropriate level of 
reserves--and also excluded certain set-aside funding and funding for 
new vouchers.[Footnote 33] As a result, to meet the $650 million 
rescission goal, HUD would have to offset more funds from housing 
agencies' reserves than would be required under a one-month reserve 
criterion, potentially resulting in some housing agencies holding less 
than a one month reserve for contingencies.[Footnote 34] 

Administrative Reforms Could Streamline Burdensome Requirements and 
Reduce Costs: 

HUD officials have noted that certain requirements for administering 
the voucher program have grown burdensome and costly and could be 
streamlined. In May 2010, the Secretary of HUD testified that the 
department's rental assistance programs "desperately need 
simplification." He further stated that HUD must streamline and 
simplify its programs so that they are easier for families to access, 
less costly to operate, and easier to administer at the local level. 
[Footnote 35] For example, under current rules, housing agencies must 
re-examine household income and composition at least annually and 
adopt policies describing when interim re-examinations will be 
conducted.[Footnote 36] HUD has expressed support for extending the 
time between re-examination of income for households on fixed incomes 
from 1 to 3 years and the time between unit inspections from 1 to 2 
years [Footnote 37]--according to one program administrator that 
manages voucher programs for five housing agencies, annual re-
examinations and inspections account for more than 50 percent of 
administrative costs in the voucher programs the agency administers. 

However, overall data are not available on the actual costs of 
specific administrative activities, such as annual income re-
examinations and inspections, and how they vary across housing 
agencies. To help address this lack of information, HUD has initiated 
a study to determine (1) what constitutes an efficient voucher 
program, (2) what a realistic expectation would be for what a housing 
agency should be doing to run an efficient program, (3) how much it 
costs to run an efficient program, and (4) what an appropriate formula 
would be for allocating administrative fees to housing agencies 
operating voucher programs.[Footnote 38] According HUD, the study will 
allow the department to analyze all aspects of voucher program 
administration to reduce and simplify housing agencies' administrative 
responsibilities.[Footnote 39] Such information will be important as 
congressional decision makers consider potential reforms of 
administrative requirements. 

Although some of the changes needed to simplify and streamline the 
voucher program would require Congress to amend existing statutes, 
HUD's administrative fee study and the experiences of housing agencies 
participating in MTW may provide insight into specific reforms to ease 
housing agencies' reported administrative burden, as well as any 
potential cost savings resulting from these reforms. For example, 
according to a HUD report, while conclusive effects of many MTW 
reforms, particularly as they relate to assisted households, are not 
known, some of the demonstration's most compelling results to date are 
those related to housing agency operations.[Footnote 40] As shown in 
table 7, many of the housing agencies that participate in the 
demonstration have implemented reforms that Congress has been 
considering through draft legislation, HUD has proposed in its fiscal 
year 2012 budget request, or both.[Footnote 41] According to the MTW 
agencies, many of these initiatives have resulted in both time and 
cost savings in their programs. 

Table 7: Administrative Streamlining Efforts Implemented or Proposed 
at MTW Agencies: 

Administrative efficiency: Allow housing agencies to make subsidy 
payments to landlords as they correct non-life-threatening conditions 
identified as part of an inspection; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
5[A]. 

Administrative efficiency: Conduct biennial inspections; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
18[B]. 

Administrative efficiency: Allow inspections from alternate sources, 
including those conducted for other federal, state, and local housing 
assistance programs; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
2. 

Administrative efficiency: Conduct triennial re-examinations for 
households with fixed incomes; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
26[C]. 

Administrative efficiency: Allow fixed-income households' incomes to 
be adjusted by applying an inflationary factor established by the 
Secretary in any year the housing agency does not conduct a review of 
income; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
5. 

Administrative efficiency: Allow housing agencies to conduct interim 
re-examinations only when a household's income or deductions change by 
an amount that is estimated to result in an increase of 10 percent or 
more in annual adjusted income; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
6[D]. 

Administrative efficiency: Allow housing agencies to use households' 
prior-year income in determining income for annual reviews; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
0. 

Administrative efficiency: Allow housing agencies to rely on income 
determinations made for other federal, means-tested, public assistance 
programs; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
2. 

Administrative efficiency: Exclude from incomes imputed returns on net 
assets that do not exceed $50,000; 
Number of MTW agencies that implemented the (or a similar) efficiency: 
20[E]. 

Source: GAO analysis of draft legislation and interviews with MTW 
agencies. 

[A] All housing agencies allow either the tenant or owner, or both to 
self-certify correction of non-life-threatening conditions. 

[B] Most of these housing agencies set criteria that need to be meet 
before changing the inspection schedule of a unit. For example, some 
agencies specify that units must have passed one inspection before 
placing them on a biennial schedule. 

[C] The majority of these housing agencies conduct biennial re-
examinations of fixed-income households rather than triennial. In 
addition, 12 have an alternate re-examination schedule for other types 
of households. 

[D] These housing agencies have made changes to their interim re-
examination policies. For example, one of these agencies limits 
interim re-examinations based on a criteria that includes a decrease 
in adjusted income of 10 percent or more (elderly/disabled families 
are exempt). 

[E] The amount of asset income excluded from income varies. For 
example, some set this amount at $5,000 and others exclude asset 
income. 

[End of table] 

In addition, and as previously discussed, the existing administrative 
fee formula generally is linked to fair market rents that are adjusted 
annually to reflect changes in wage rates, and HUD pays fees to 
housing agencies based on the number of units leased (vouchers used) 
as of the first of each month. This formula is not tied to the 
program's current administrative costs or requirements. Further, 
housing agencies we contacted reported that the cost of administering 
the voucher program has been on the rise, with contributing factors 
including higher postage, fuel, and employee health care costs, as 
well as increased reporting and other requirements. 

Without more specific information about potential reform options, 
policymakers will not be able to make an informed decision about how 
to reform the administrative fee formula and the activities required 
to administer an efficient voucher program. These efforts--using the 
administrative fee study to identify specific reforms and leveraging 
the experiences of MTW agencies--are in line with the goals of the 
Government Performance and Results Act of 1993 (GPRA), which Congress 
enacted, in part, to inform its decision making by helping to ensure 
that agencies provide objective information on the relative 
effectiveness and efficiency of their programs and spending.[Footnote 
42] Whether HUD's study will yield findings that eventually will 
result in measureable cost or time savings is not clear. While 
reforming administrative requirements for the voucher programs could 
lead to increased efficiencies and cost savings, the administrative 
fee paid to housing agencies is a relatively modest share of the 
program's overall annual appropriations--approximately 9 percent in 
recent years. Nevertheless, such efforts will provide Congress with 
timely and meaningful information, which will enhance its ability to 
make decisions about funding for and requirements of the voucher 
program. 

Rent Reform and Consolidation Could Result in Reduced Costs or More 
Households Served, but Both Involve Trade-offs: 

If implemented, rent reform (that is, changes to the calculation of 
households' payment toward rent) and the consolidation of voucher 
administration under fewer housing agencies could yield substantial 
cost savings, allow housing agencies to serve additional households if 
Congress were to reinvest annual cost savings in the voucher program, 
or both.[Footnote 43] Further, these reform options are not mutually 
exclusive; that is, cost savings or additional households served could 
be greater if both options were implemented. Further, implementation 
of these options may involve some trade-offs, including increased rent 
burdens for assisted households. 

Rent Reform: 

As previously discussed, under current program rules, an assisted 
household generally must contribute the greater of 30 percent of its 
monthly adjusted income or the housing-agency established minimum 
rent--up to $50--toward its monthly rent. HUD's subsidy is the 
difference between (1) the lesser of the unit's gross rent or the 
applicable payment standard and (2) the household's rental payment. 
Therefore, as an assisted household's income increases, HUD's subsidy 
payment decreases, and vice versa. Under existing program rules, a 
household could pay no rent--if the household has no monthly income 
after adjustments, the housing agency from which the household 
receives assistance does not have a minimum rent, or the household 
obtained a hardship exemption.[Footnote 44] However, such households 
make up a small share of all voucher-assisted households, with more 
than 99 percent making some dollar contributions to their rent. 
[Footnote 45] 

Because about 90 percent of voucher program funds are used to pay 
subsidies, decreasing the level of subsidy for which households are 
eligible (or, alternatively stated, increasing the amount households 
must contribute toward rent) necessarily will yield the greatest costs 
savings for the program. We estimated the effect, both in terms of 
cost savings and additional households that could be served with those 
savings if Congress chose to reinvest the savings in the program, of 
several options including requiring assisted households to pay: 

* higher minimum rents; 

* 35 percent of their adjusted income in rent; 

* 30 percent of their gross income in rent (with no adjustments); 
[Footnote 46] or; 

* a percentage of the applicable fair market rent.[Footnote 47] 

Using HUD data, we determined that each of these options could reduce 
the federal cost burden--in some cases, quite considerably--or if 
Congress chose to reinvest cost savings in the program, allow housing 
agencies to serve more households without additional funding. For 
example, as shown in table 8, increasing minimum rents to $300 would 
yield the greatest cost savings on an annual basis--an estimated $1.8 
billion--or allow housing agencies to serve the greatest number of 
additional households--an estimated 287,000. Requiring assisted 
households to pay 30 percent of their gross income in rent would yield 
the least savings for the voucher program and serve the fewest 
additional households. Further, HUD operates a number of other rental 
assistance programs where household subsidies are based on the same 
calculations as those for the voucher program. Implementation of these 
rent reform options in its other rental assistance programs has the 
potential to create additional cost savings. 

Table 8: Estimated Annual Effect of Selected Rent Reform Options on 
Program Costs and Number of Voucher-Assisted Households Served: 

Reform option: Increase minimum rents to:[B,C] 

Reform option: $50[D]; 
Estimated annual cost savings[A]: $11 million; 
Estimated additional households served: 1,400. 

Reform option: 75; 
Estimated annual cost savings[A]: $67 million; 
Estimated additional households served: 8,600. 

Reform option: 100; 
Estimated annual cost savings[A]: $124 million; 
Estimated additional households served: 16,000. 

Reform option: 150; 
Estimated annual cost savings[A]: $318 million; 
Estimated additional households served: 43,000. 

Reform option: 200; 
Estimated annual cost savings[A]: $602 million; 
Estimated additional households served: 85,000. 

Reform option: 250; 
Estimated annual cost savings[A]: $1.1 billion; 
Estimated additional households served: 167,000. 

Reform option: 300; 
Estimated annual cost savings[A]: $1.8 billion; 
Estimated additional households served: 287,000. 

Reform option: Require households to pay:[E] 

Reform option: 35 percent of adjusted income in rent; 
Estimated annual cost savings[A]: $1.1 billion; 
Estimated additional households served: 164,000. 

Reform option: 30 percent of gross income in rent; 
Estimated annual cost savings[A]: $513 million; 
Estimated additional households served: 76,000. 

Reform option: 35 percent of the fair market rent[F]; 
Estimated annual cost savings[A]: $927 million; 
Estimated additional households served: 136,000. 

Source: GAO analysis of HUD data. 

[A] To estimate the effect of these options on program costs and 
households assisted, we analyzed household characteristic and rent 
data as of December 2010. These estimates illustrate the relative 
effects of the options if fully implemented in one year. Actual 
implementation of such options likely would be done gradually and not 
all of the savings or efficiencies would be realized in the first year. 

[B] We assumed that all households paid the greater of the minimum 
rent or 30 percent of adjusted income. Our minimum rent calculations 
did not take into account any payment households received for utility 
assistance. 

[C] On January 31, 2012, the House Subcommittee on Insurance, Housing, 
and Community Opportunity released a revised draft of the Section 8 
Savings Act entitled the Affordable Housing and Self-Sufficiency 
Improvement Act of 2012. The draft bill proposes implementing a 
minimum rent of at least $69.45 (adjusted annually). We estimated that 
this increase would save approximately $56 million annually or could 
be used to serve an additional 7,100 households if Congress chose to 
reinvest the savings. 

[D] We assumed all applicable households paid $50 in rent. As 
previously discussed, although housing agencies are permitted to set a 
minimum rent of up to $50, not all do and many offer hardship 
exemptions from the requirement. 

[E] For the adjusted and gross income options, we did not impose a 
minimum rent requirement. 

[F] For this option, we evaluated the effect of requiring households 
to pay 12, 15, 20, 30, and 35 percent of the fair market rent and no 
minimum rent. Only the 35 percent option resulted in cost savings or 
additional households served on an annual basis--all other percentages 
resulted in cost increases and fewer households served. 

[End of table] 

These reform options could be implemented individually and some could 
be implemented together, depending on the objective policymakers were 
trying to achieve--such as maximizing cost savings, minimizing the 
impact on assisted households, or promoting work and self sufficiency 
among families with children (that is, nonelderly, nondisabled 
households).[Footnote 48] To illustrate, one housing agency in the MTW 
program put in place a rent structure that gradually increases 
household rents--from 27 percent of gross income in years 1 and 2, to 
the greater of $100 or 29 percent of gross income in years 3 and 4, 
and to the greater or $200 or 31 percent of gross income in all 
subsequent years--to promote self-sufficiency among all assisted 
households. Under this approach, our analysis showed that households 
receive more subsidy in the first 2 years, but pay more rent over time 
than under the current rent structure.[Footnote 49] 

In addition to estimating the cost savings that could result from each 
of these rent reform options, we evaluated each option in terms of its 
effect on (1) changes in the rent paid by assisted households, (2) 
household attrition rates, (3) HUD's goals of encouraging households 
to move to the neighborhoods of their choice (mobility) and 
discouraging households from choosing communities that have higher 
levels of poverty (deconcentration), (4) incentives to seek work, (5) 
program administration, and (6) housing agency and industry support. 
While each of these options has advantages over the current rent 
structure--they could reduce costs or create administrative 
efficiencies--each also involves trade-offs. 

Under each rent reform option, some households would have to pay more 
in rent than they currently pay. For example, as shown in table 9, if 
all households were required to pay at least $50 in rent per month, an 
estimated 36,000 households (2 percent) would experience an average 
increase of $31 in their monthly rent. HUD's fiscal year 2013 budget 
request proposes increasing the minimum rent to $75 per month for all 
assisted household. Under this option, 207,000 households (11 percent) 
would experience an average increase of $27. Table 9 also shows 
options that change the formula for calculating the households' 
payment toward rent. For example, setting the households rental 
payment to 30 percent of gross income (that is, without any 
deductions) would affect about 1,662,000 households (86 percent) and 
increase mean household rent by $27. 

Table 9: Estimated Number and Percentage of Voucher-Assisted 
Households for Which Rents Would Increase and the Average Monthly 
Increase, by Rent Reform Option: 

Minimum rent[B]: $50; 
Number of households experiencing an increase in their monthly 
payment[A]: 36,000; 
Percentage of households experiencing an increase in their monthly 
payment: 2%; 
Mean change in monthly payment of affected households: $31. 

Minimum rent[B]: $75[C]; 
Number of households experiencing an increase in their monthly 
payment[A]: 207,000; 
Percentage of households experiencing an increase in their monthly 
payment: 11%; 
Mean change in monthly payment of affected households: 27. 

Minimum rent[B]: $100; 
Number of households experiencing an increase in their monthly 
payment[A]: 256,000; 
Percentage of households experiencing an increase in their monthly 
payment: 13%; 
Mean change in monthly payment of affected households: 45. 

Minimum rent[B]: $150; 
Number of households experiencing an increase in their monthly 
payment[A]: 358,000; 
Percentage of households experiencing an increase in their monthly 
payment: 19%; 
Mean change in monthly payment of affected households: 75. 

Minimum rent[B]: $200; 
Number of households experiencing an increase in their monthly 
payment[A]: 698,000; 
Percentage of households experiencing an increase in their monthly 
payment: 36%; 
Mean change in monthly payment of affected households: 71. 

Minimum rent[B]: $250; 
Number of households experiencing an increase in their monthly 
payment[A]: 1,012,000; 
Percentage of households experiencing an increase in their monthly 
payment: 52%; 
Mean change in monthly payment of affected households: 92. 

Minimum rent[B]: $300; 
Number of households experiencing an increase in their monthly 
payment[A]: 1,225,000; 
Percentage of households experiencing an increase in their monthly 
payment: 63%; 
Mean change in monthly payment of affected households: 122. 

Household Rent Formula: 35% of adjusted income; 
Number of households experiencing an increase in their monthly 
payment[A]: 1,751,000; 
Percentage of households experiencing an increase in their monthly 
payment: 92%; 
Mean change in monthly payment of affected households: $50. 

Household Rent Formula: 30% of gross income; 
Number of households experiencing an increase in their monthly 
payment[A]: 1,662,000; 
Percentage of households experiencing an increase in their monthly 
payment: 86%; 
Mean change in monthly payment of affected households: 27. 

Household Rent Formula: 35% of fair market rent[D]; 
Number of households experiencing an increase in their monthly 
payment[A]: 1,172,000; 
Percentage of households experiencing an increase in their monthly 
payment: 61%; 
Mean change in monthly payment of affected households: 155. 

Source: GAO analysis of HUD data. 

[A] To estimate the effect of each rent reform options on the number 
of households affected and their monthly payments, we analyzed 
household characteristic and rent data as of December 2010. 

[B] The Affordable Housing and Self-Sufficiency Improvement Act of 
2012proposes implementing a minimum rent of at least $69.45 (adjusted 
annually). We estimated that, if implemented, approximately 198,000 
households would experience an average increase of $23 in their 
monthly payment. 

[C] HUD's fiscal year 2013 proposed budget proposes increasing the 
minimum rent to $75 per month for all assisted households. 

[D] Under this option, approximately 755,000 households would 
experience an average decrease of $139 in their monthly payment. 

[End of table] 

Increasing minimum rents primarily would affect families with children 
that tend to report little or no income. Conversely, assisted elderly 
and disabled households almost always report income (most likely 
because they are on fixed incomes, like Social Security) and a large 
percentage of them already pay close to $200 in rent.[Footnote 50] On 
a programwide level, imposing minimum rents of $200 or less does not 
change the amount these households pay in rent, when considering all 
assisted households. Figure 4 shows the mean change in all households' 
monthly rent resulting from each of these rent reform options. 
Increases in monthly rental payments for elderly and disabled 
households begin to increase more significantly with a $200 minimum 
rent and under each of the rent formula changes. As a result, higher 
minimum rents or increases to the percentage of their incomes paid in 
rent will yield the greatest cost savings. For the rent formula change 
to 35 percent of adjusted income, the mean change in monthly rent 
generally would be similar across each household type. 

Figure 4: Estimated Effect of Rent Reform Options on Monthly Rents of 
All Voucher-Assisted Households, by Household Type: 

[Refer to PDF for image: illustrated horizontal bar graph] 

Rent reform option: Minimum rent: $50; 
Mean change in monthly tenant payment by household type: 
Families with children: $1; 
Disabled: $0; 
Elderly: $0; 
Elderly, disabled: $0. 

Rent reform option: Minimum rent: $75; 
Mean change in monthly tenant payment by household type: 
Families with children: $6; 
Disabled: $0; 
Elderly: $0; 
Elderly, disabled: $0. 

Rent reform option: Minimum rent: $100; 
Mean change in monthly tenant payment by household type: 
Families with children: $11; 
Disabled: $1; 
Elderly: $1; 
Elderly, disabled: $0. 

Rent reform option: Minimum rent: $150; 
Mean change in monthly tenant payment by household type: 
Families with children: $27; 
Disabled: $2; 
Elderly: $2; 
Elderly, disabled: $0. 

Rent reform option: Minimum rent: $200; 
Mean change in monthly tenant payment by household type: 
Families with children: $47; 
Disabled: $8; 
Elderly: $7; 
Elderly, disabled: $3. 

Rent reform option: Minimum rent: $250; 
Mean change in monthly tenant payment by household type: 
Families with children: $71; 
Disabled: $30; 
Elderly: $23; 
Elderly, disabled: $22. 

Rent reform option: Minimum rent: $300; 
Mean change in monthly tenant payment by household type: 
Families with children: $99; 
Disabled: $61; 
Elderly: $49; 
Elderly, disabled: $52. 

Rent reform option: Household rent formula: 35% adjusted income; 
Mean change in monthly tenant payment by household type: 
Families with children: $44; 
Disabled: $47; 
Elderly: $49; 
Elderly, disabled: $48. 

Rent reform option: Household rent formula: 30% gross income; 
Mean change in monthly tenant payment by household type: 
Families with children: $23; 
Disabled: $21; 
Elderly: $27; 
Elderly, disabled: $19. 

Rent reform option: Household rent formula: 35% fair market rent; 
Mean change in monthly tenant payment by household type: 
Families with children: $61; 
Disabled: $23; 
Elderly: $8; 
Elderly, disabled: $22. 

Source: GAO analysis of HUD data. 

Note: To estimate the effect each rent reform option on monthly 
household payments, we analyzed household characteristic and rent data 
as of December 2010. The Affordable Housing and Self-Sufficiency 
Improvement Act of 2012 proposes implementing a minimum rent of at 
least $69.45 (adjusted annually). Under this change, only families 
with children would experience a change in rent ($5 on average). 

[End of figure] 

Figure 5 shows the mean change in monthly rent only for those 
households whose payments toward rent have changed as a result of each 
reform option. Among these affected households, changes in rental 
payments would be similar across household types for some of the rent 
structure options. For example, if households were required to pay a 
$75 minimum rent, mean rental payments would increase by $30 for 
disabled households (on the high end) and $24 for elderly, disabled 
households (on the low end). However, if households were required to 
pay a $200 or higher minimum rent, families with children again would 
experience higher mean changes in rent than disabled and elderly 
households. 

Figure 5: Estimated Effect of Rent Reform Options on Monthly Rents of 
Affected Voucher-Assisted Households, by Household Type: 

[Refer to PDF for image: illustrated horizontal bar graph] 

Rent reform option: Minimum rent: $50; 
Mean change in monthly tenant payment by household type: 
Families with children: $31; 
Disabled: $33; 
Elderly: $33; 
Elderly, disabled: $29. 

Rent reform option: Minimum rent: $75; 
Mean change in monthly tenant payment by household type: 
Families with children: $27; 
Disabled: $30; 
Elderly: $25; 
Elderly, disabled: $24. 

Rent reform option: Minimum rent: $100; 
Mean change in monthly tenant payment by household type: 
Families with children: $45; 
Disabled: $46; 
Elderly: $38; 
Elderly, disabled: $33. 

Rent reform option: Minimum rent: $150; 
Mean change in monthly tenant payment by household type: 
Families with children: $77; 
Disabled: $63; 
Elderly: $51; 
Elderly, disabled: $39. 

Rent reform option: Minimum rent: $200; 
Mean change in monthly tenant payment by household type: 
Families with children: $106; 
Disabled: $25; 
Elderly: $32; 
Elderly, disabled: $13. 

Rent reform option: Minimum rent: $250; 
Mean change in monthly tenant payment by household type: 
Families with children: $135; 
Disabled: $55; 
Elderly: $52; 
Elderly, disabled: $44. 

Rent reform option: Minimum rent: $300; 
Mean change in monthly tenant payment by household type: 
Families with children: $166; 
Disabled: $90; 
Elderly: $81; 
Elderly, disabled: $78. 

Rent reform option: Household rent formula: Household payment 
increases: 35% adjusted income; 
Mean change in monthly tenant payment by household type: 
Families with children: $53; 
Disabled: $47; 
Elderly: $50; 
Elderly, disabled: $48. 

Rent reform option: Household rent formula: Household payment 
increases: 30% gross income; 
Mean change in monthly tenant payment by household type: 
Families with children: $24; 
Disabled: $22; 
Elderly: $28; 
Elderly, disabled: $20. 

Rent reform option: Household rent formula: Household payment 
increases: 35% fair market rent; 
Mean change in monthly tenant payment by household type: 
Families with children: $196; 
Disabled: $113; 
Elderly: $125; 
Elderly, disabled: $110. 

Rent reform option: Household rent formula: Household payment 
decreases: 35% adjusted income; 
Mean change in monthly tenant payment by household type: 
Families with children: N/A; 
Disabled: N/A; 
Elderly: N/A; 
Elderly, disabled: N/A. 

Rent reform option: Household rent formula: Household payment 
decreases: 30% gross income; 
Mean change in monthly tenant payment by household type: 
Families with children: N/A; 
Disabled: N/A; 
Elderly: N/A; 
Elderly, disabled: N/A. 

Rent reform option: Household rent formula: Household payment 
decreases: 35% fair market rent; 
Mean change in monthly tenant payment by household type: 
Families with children: -$183; 
Disabled: -$106; 
Elderly: -$115; 
Elderly, disabled: -$96. 

Source: GAO analysis of HUD data. 

Note: To estimate the effect each rent reform option on monthly 
household payments, we analyzed household characteristic and rent data 
as of December 2010. The Affordable Housing and Self-Sufficiency 
Improvement Act of 2012proposes implementing a minimum rent of at 
least $69.45 (adjusted annually). We estimated that, if implemented, 
mean rental payments would increase by $23 for families with children; 
$26 for disabled households; and $21 for both elderly and elderly, 
disabled households. 

[End of figure] 

Also as shown in figure 5, under the option where the rental payments 
are based on 35 percent of the fair market rent, some households will 
have to pay more in monthly rent, while others will pay less. Further, 
a higher proportion of affected households will see an increase in 
their rental payments. Specifically, of the approximately 1.9 million 
total households whose monthly rental payments would change under this 
option, about 61 percent (approximately 1.2 million households) would 
experience an increase in their monthly payments and about 39 percent 
(755,000 households) would experience a decrease. 

Requiring households' rental payments to be based on a percentage of 
the applicable fair market rent rather than 30 percent of adjusted 
income primarily would affect households living in high-cost (mostly 
urban) areas and large families, as well as those at the lower end of 
the income scale. HUD's fair market rents reflect market prices and 
unit sizes--thus, household rent shares will increase if they live in 
a more expensive fair market area or rent larger units in the same 
fair market rent area under a rent option based on percentage of fair 
market rents. Table 10 illustrates how fair market rents and household 
payments based on a percentage of the fair market rent can vary by 
location and unit size. 

Table 10: Comparison of Two-and Three-Bedroom Monthly Rents under a 35 
Percent of Fair Market Rent Structure, for Three Areas: 

2 bedroom unit: 
Boston, Massachusetts: 
Fair market rent: $1,357; 
Monthly tenant payment: $475; 
Concord, New Hampshire: 
Fair market rent: $993; 
Monthly tenant payment: $348; 
Farmington, Maine: 
Fair market rent: $687; 
Monthly tenant payment: $240. 

3 bedroom unit: 
Boston, Massachusetts: 
Fair market rent: $1,623; 
Monthly tenant payment: $568; 
Concord, New Hampshire: 
Fair market rent: $1,226; 
Monthly tenant payment: $429; 
Farmington, Maine: 
Fair market rent: $820; 
Monthly tenant payment: $287. 

Source: GAO analysis of HUD data. 

[End of table] 

In addition, under an option where households' rental payments are 
based on a percentage of the fair market rent, lower-income households 
would pay a larger percentage of their income toward rent than higher-
income households. And while many of the lowest-income households 
would experience rent increases ($116 per month, on average for 
families with children), many of the highest-income households would 
experience rent decreases ($97 per month).[Footnote 51] 

Under each of these rent reform options, a small number of households 
might lose their subsidies--that is, their subsidy payments would be 
reduced to zero because their new, higher rental payments would fully 
cover the gross rent. For example, under the option where households 
pay 35 percent of their adjusted income in rent, we estimated that 
approximately 1.8 percent of households would lose their subsidies. 
[Footnote 52] Further, other affected households might leave the 
program because they would have to pay more in rent and no longer 
choose to participate in the program. However, because the demand for 
rental assistance by low-income households generally exceeds the 
number of available vouchers, eligible household likely would replace 
the one that left because similar unassisted households have much 
higher rent burdens than assisted households. Consequently, these rent 
reform options likely would not result in a sharp decline in program 
participation rates. 

Rent structures that decrease the amount of subsidy households receive 
may discourage HUD's deconcentration efforts, as well as household 
mobility. With less subsidy, households (especially those with lower 
incomes) may not have the means to move from neighborhoods of 
concentrated poverty to those with a diversity of people and 
opportunities. But HUD's deconcentration goal presents its own trade-
offs--chief among them that fewer households ultimately would be 
served, albeit with more generous subsidies. Among the rent reform 
structures we evaluated, all but one would decrease household 
subsidies. A rent structure under which households would pay 30 
percent or less of the applicable fair market rent would increase 
subsidies for almost all households and thus could further HUD's 
deconcentration and mobility goals.[Footnote 53] 

Two of the rent structures we evaluated--higher minimum rents and 
rents based on a percentage of the fair market rent--could create work 
incentives for households with little to no income. Under the current 
rent structure, and as previously discussed, a household with no 
income generally does not pay rent--HUD's subsidy covers the gross 
rent. Consequently, some have argued that these households have little 
incentive to seek employment because, for every $1 they earn, their 
subsidies are reduced by 30 cents (for every $100 they earn on a 
monthly basis, they will pay $30 in rent). Rent structures that do not 
take into account household income may do more to encourage assisted 
households to find and retain employment.[Footnote 54] Housing 
agencies in the MTW program that have implemented these types of rent 
structures simultaneously have offered self-sufficiency training and 
services to assisted households. Additionally, rent structures that 
eliminate household income from the rent equation may allow Congress 
and HUD to more accurately forecast funding needs. As we previously 
discussed, market rents and tenant incomes are two of the primary 
drivers of program costs, and predicting changes in market rents and 
incomes when developing budget proposals for future years is 
difficult. These types of rent structures also would encourage 
assisted households to make choices about housing consumption similar 
to unassisted households. For example, households would not have an 
incentive to over-consume housing because their share of the rent 
would increases with the size of the unit they rented. 

Moving toward a rent structure either based on fair market rents or 
gross income would introduce significant administrative efficiencies 
into the program and could allow housing agencies to further reduce 
improper payments from administrator (housing agency) error or 
tenants' underreporting of income. Some housing agencies we contacted 
noted the complexity of the current income and rent determination 
process and their frustrations with tracking the existence of and 
changes to tenant incomes. HUD noted that requiring assisted 
households to pay higher minimum rents or 35 percent of their adjust 
income in rent would not create administrative efficiencies in the 
voucher program. Our 2005 report on improper subsidy payments in HUD's 
rental assistance programs made similar observations, finding that the 
complexity of HUD's income and rent determination policies were of 
major concern to HUD field offices, program administrators, and 
industry groups.[Footnote 55] HUD officials noted at the time that the 
department was considering various approaches for statutory, 
regulatory, and administrative streamlining and simplification of its 
policies for determining subsidies. 

Finally, nearly all of the housing agencies we contacted said that 
they supported some type of rent reform--among the most popular 
options were increasing minimum rents and increasing tenant rental 
payments to 35 percent of adjusted income. Some housing agencies have 
suggested that they have been successful in implementing rent reform 
under the MTW program with community support.[Footnote 56] Despite 
this, some industry groups have voiced concern about rent reform. For 
example, in commenting on a provision included in the draft Section 8 
Savings Act of 2011 that would permit HUD to pursue a rent 
demonstration, the National Low Income Housing Coalition stated that 
the demonstration would put HUD-assisted households at risk of having 
significant rent burdens.[Footnote 57] The Coalition also said that 
any demonstration should include parameters that require HUD to 
monitor these burdens and stop or change the demonstration if it were 
found to harm assisted households.[Footnote 58] 

Consolidation of Voucher Administration: 

Based on our literature review and interviews with HUD and housing 
industry officials, consolidation of voucher program administration 
under fewer housing agencies (administrative consolidation) could 
yield a more efficient oversight and administrative structure for the 
voucher program and cost savings for HUD and housing agencies; 
however, current information on the magnitude of these savings was not 
available. 

HUD spends considerable resources in overseeing housing agencies. More 
than 2,400 local housing agencies administer the voucher program on 
HUD's behalf. According to a 2008 HUD study, the department dedicated 
from more than half to two-thirds of its level of oversight to 10 
percent of its units (generally those housing agencies that administer 
400 or fewer vouchers), and an even lower level of risk in relation to 
the amount of subsidy funds they administered (about 5 percent of 
total program funds).[Footnote 59] According to agency officials, 
consolidating the administration of vouchers under fewer agencies 
would decrease HUD's oversight responsibilities. 

According officials from HUD and some housing agencies with whom we 
spoke, administering the voucher program through small local housing 
agencies may be less cost effective, in part because of the 
differences in the economies of scale. For example, larger housing 
agencies can realize cost efficiencies in conducting large numbers of 
voucher unit inspections that smaller agencies cannot. Also, larger 
housing authorities collect sufficient fees to support fraud detection 
units to ensure that households report all of their income sources. 
Although there are no current data on the comparative costs of 
administering the voucher program though small and large housing 
agencies, the current administrative fee structure recognizes that 
economies of scale exist in larger housing agencies. As previously 
discussed, HUD pays housing agencies a higher rate for the first 600 
vouchers a housing agency has under lease and a lower rate for the 
remaining units under lease. Congress passed this two-tiered fee 
structure based in part on a 1994 HUD study that found that flat fee 
rates were leading to administrative fee deficits in small housing 
agencies and large administrative fee reserves at larger housing 
agencies.[Footnote 60] 

HUD has acknowledged that oversight and administrative efficiencies 
could be realized. As previously discussed, in recent years, the 
department has advanced several proposals aimed at streamlining and 
simplifying administration of the voucher program. Several of these 
proposals have advocated administrative consolidation as a means of 
creating administrative efficiencies. For example, HUD's 2011 version 
of the Transforming Rental Assistance initiative was intended to 
streamline and improve the delivery and oversight of rental assistance 
across all of the department's rental assistance programs by means 
such as promoting consortiums, consolidation, and other locally 
designed structures for administrative functions. 

In addition, HUD recently initiated changes to its housing agency 
consortium rule. The revised rule would treat all housing agencies in 
a consortium as one entity--HUD's current regulation requires that 
consortium members be treated separately for oversight, reporting, and 
other purposes. Some have argued that the current rule does not allow 
HUD or housing agencies to realize the full benefits of consolidation--
less oversight (one versus multiple agencies) and shared and thus 
reduced administrative responsibilities--and therefore discourages the 
formation of consortiums. Since 1998, nine housing agencies that 
administer vouchers have formed four consortiums. 

We evaluated the administrative consolidation in terms of its effect 
on assisted households and selected voucher program goals. More 
specifically, we looked at implications for, or likelihood of 
achieving (1) HUD's mobility and deconcentration goals, (2) program 
administration, and (3) housing agency and industry support. Like the 
rent reform options we evaluated using similar criteria, consolidation 
has advantages over the current administrative structure, but also 
involves some trade-offs. 

Consolidation might help HUD more readily achieve deconcentration 
goals. Although vouchers theoretically allow recipients to use them 
anywhere in the United States, the current system of program 
administration creates numerous hurdles for households to move out of 
high-poverty, central city jurisdictions in which they typically live. 
Most housing agencies originally were established to construct and 
manage public housing developments.[Footnote 61] As a result, program 
administration does not always align with housing markets. In urban 
areas within the same market, several housing agencies may operate 
voucher programs with different admissions criteria and subsidy 
levels. A paper by researchers at the Brookings Institution argued 
that this "fragmentation of local program administration undermines 
the potential of the [voucher] program as a mechanism for 
deconcentrating urban poverty."[Footnote 62] Extending the 
jurisdiction of housing agencies (through consolidation, for example) 
likely would give assisted households access to more housing options, 
particularly in surrounding suburbs. On the other hand, regionalized 
administration of the voucher program may make it harder for 
households to make or maintain contact with program administrators 
when necessary--for example, assisted households may not have access 
to transportation or may have to travel long distances to meet with 
housing agency officials. 

Several states offer examples of regional or statewide administration. 
Thirty-one states have programs in which one housing agency 
administers a voucher program throughout a state. These housing 
agencies administer from less than one percent to all of their 
respective state's total voucher allocation. In addition, as part of 
our work, we visited a number of housing agencies in the Boston, 
Massachusetts, metropolitan area. As a result of litigation in the mid-
1990s, local housing agencies in the state are permitted to lease 
vouchers throughout the state (that is, outside their original 
jurisdictions, which typically align with city limits).[Footnote 63] 
Although all of the housing agencies with which we spoke suggested 
that it was important that housing agencies maintain local control of 
their programs, each leased at least one voucher outside their 
original jurisdiction. In Brookline--a city with relatively high 
housing costs compared with the surrounding area and the nation--more 
than half of voucher holders rent apartments outside the city limits. 

Although consolidation will not alleviate housing agencies' current 
administrative burden, it may begin to address some of the issues 
housing agencies and industry groups have raised about a particular 
policy--portability. Although portability is one of the hallmark 
objectives of the voucher program, almost all the housing agencies we 
contacted said that HUD's portability polices should be revised or 
eliminated, noting that they are complicated and costly to administer. 
Under HUD's portability rules, an assisted household may move to the 
jurisdiction of a different housing agency--the receiving agency 
either may bill the sending agency for assistance for the transferring 
household or absorb the household into its own program. According to 
the 2000 Brookings Institution report, because of the complexity of 
the portability process--for example, receiving agencies may calculate 
subsidy levels differently than sending agencies, or apply more 
rigorous screening criteria--many housing agencies do not fully 
explain portability to households and do not encourage them to 
consider moving.[Footnote 64] 

In addition, consolidated waiting lists and single points of contact 
for housing assistance within a single housing market, region, or 
state may make the process of applying for and obtaining rental 
assistance less confusing and more transparent for households seeking 
assistance. For example, a large number of housing agencies in 
Massachusetts participate in a consolidated waiting list--households 
seeking assistance in the state need only put their name on one list 
and receive communications from one agency. HUD officials said that 
the department has been considering taking steps to maintain the 
waiting lists of each housing agency in a centralized system. 

Finally, housing agencies we contacted were split on the idea of 
consolidation--about one quarter supported it as a way to cut costs 
and introduce administrative efficiencies in the voucher program, 
while almost half were against it. Some housing industry groups and an 
academic with which we spoke argued that consolidation would not save 
money--one noted that the administrative fees that small housing 
agencies receive are relatively insignificant in terms of total 
program dollars--and would sacrifice local discretion and control of 
voucher programs. Others noted that administrative costs savings could 
result from the consolidation and single-source management of waiting 
lists and elimination or substantial reformation of the portability 
process; however, no data currently are available to assess this point. 

Conclusions: 

Over the past decade, Congress has responded to the increasing cost of 
vouchers by changing the way the program is funded. Specifically, 
rather than providing funding based on the number of vouchers housing 
agencies are permitted to lease, Congress currently provides funding 
based on housing agencies' prior-year subsidy expenses. Congress also 
has capped appropriations so that housing agencies do not always 
receive the amount of subsidy or administrative funding for which they 
are eligible based on the funding formulas Congress annually 
establishes. While this approach gives Congress some control over cost 
increases, it does not directly address the market and policy factors 
we identified as contributing to increases in program costs. 

Although policy makers can do little to alter or control market 
changes such as changes in rents and tenant incomes, our analysis 
suggests that savings could continue to be realized (or, in some 
cases, more households could be served without additional program 
funding if Congress chooses to reinvest the funds in the program) if 
HUD provided Congress better information on housing agencies' subsidy 
reserves. Enhanced information would include the extent of housing 
agencies' subsidy reserves, clear and consistent criteria for 
determining how much housing agencies would need to retain to help 
ensure effective program management, and how much could be rescinded 
in future appropriations. Without such information, HUD faces 
difficulties in effectively manage the funding Congress provides for 
the voucher program, including ensuring that funds disbursed to 
housing agencies are used to assist households rather than remaining 
unused in reserve accounts. 

In tandem with providing information about the use of program funds, 
HUD also has an opportunity to advance proposals that would help 
increase the efficiency of program administration. In particular, HUD 
now has or will have richer, relevant experience and data from which 
to draw. In addition to previous reforms HUD has proposed, examples 
from the MTW program and HUD's study on administrative fees can offer 
options to Congress for streamlining and simplifying administrative 
activities and aligning the administrative fee structure with actual 
administrative expenses. For example, information and analyses from 
these sources could help identify all current administrative 
requirements, determine which of those actions are necessary and which 
could be eliminated or streamlined, and determine the cost of 
performing these activities--which could help reduce program costs in 
the future. 

Although Congress and HUD have taken several steps to control rising 
costs in the voucher program, we have identified a range of options 
that offer the additional promise of managing program costs or 
increasing efficiency in the long term. These options would also be 
applicable to HUD's other rental assistance programs and would have 
the potential to generate even greater savings. Implementing rent 
reform and administrative consolidation would require policymakers to 
consider some potential trade-offs--in the balance are issues such as 
the rent burden of assisted households, concentration of poverty, and 
the extent of local control over voucher programs. Nevertheless, these 
options have certain advantages over the current program structure. 
For example, these options could save money or streamline program 
administration--both of which are important objectives in a time of 
fiscal constraint. Currently Congress is considering a variety of 
measures to address some of these issues. 

Recommendations for Executive Action: 

To help reduce voucher program costs or better ensure the efficient 
use of voucher program funds, we recommend that the HUD Secretary 
provide information to Congress on (1) housing agencies' estimated 
amount of excess subsidy reserves and (2) its criteria for how it will 
redistribute excess reserves among housing agencies so that they can 
serve more households. In taking these steps, the Secretary should 
determine a level of subsidy reserves housing agencies should retain 
on an ongoing basis to effectively manage their voucher programs. 

Further, the Secretary should consider proposing to Congress options 
for streamlining and simplifying the administration of the voucher 
program and making corresponding changes to the administrative fee 
formula to reflect any new or revised administrative requirements. 
Such proposals should be informed by results of HUD's ongoing 
administrative fee study and the experience of the MTW program. 

Agency Comments and Our Evaluation: 

We provided a draft of this report to HUD for comment. In its written 
response, reproduced in appendix II, HUD neither agreed nor disagreed 
with our recommendations, but provided technical comments that we have 
incorporated where appropriate. While the response noted that the 
draft report provided an accurate assessment of the program and its 
current outcomes, HUD identified several points for clarification and 
emphasis, including: 

* HUD commented that the stated purpose of our report of identifying 
options for increasing efficiencies and simplifying program 
administration was inconsistent with our recommendations for agency 
action because some of the options do not result in both efficiencies 
and simplification. We clarified, where appropriate, that the focus of 
our report was to identify reform options that could reduce costs or 
create efficiencies. 

* HUD also commented that the draft report's discussion of growth in 
HUD's outlays could be misleading because this growth reflects only a 
change in HUD's disbursement policy and does not relate at all to 
changes in program costs. Specifically, HUD stated that starting in 
2006, the program was required to disburse all eligible funds, instead 
of the department's maintaining those reserves. HUD did not provide 
any support that outlays reflect only a change in HUD's disbursement 
policy and do not relate at all to changes in program costs. While we 
recognize that disbursement policies may affect outlays, changes in 
program size and other factors would also affect outlays. Further, 
although the draft provides information on the trends in actual HUD 
outlays, it focuses on housing agencies' expenditures because they are 
a better measure of what housing agencies are paying in subsidies to 
assisted households with vouchers. Therefore, we made no changes in 
response to this comment. 

* HUD also commented that the draft report did not address HUD's 
ongoing efforts to limit the accumulation of subsidy reserves. We 
added additional language to the report on these efforts, such as the 
assistance HUD provides to housing agencies in ensuring that all 
available voucher funds are utilized. 

* HUD noted that it currently provides quarterly reports to the 
Congressional Budget Office on subsidy reserve levels. However, these 
quarterly reports do not include information on the estimated amount 
of housing agencies' subsidy reserves that exceed prudent levels, as 
we are recommending. By providing the estimated amount of excess 
subsidy reserves, Congress will be better positioned to make informed 
funding decisions, as we illustrated in our draft report. 

As agreed with your offices, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. At that time, we will send copies to the 
Secretary of Housing and Urban Development and other interested 
committees. In addition, the report will be available at no charge on 
the GAO Web site at [hyperlink, http://www.gao.gov]. 

If you or your staff have any questions concerning this report, please 
contact me at (202) 512-8678 or sciremj@gao.gov. Contact points for 
our Office of Congressional Relations and Public Affairs may be found 
on the last page of this report. Key contributors to this report are 
listed in appendix III. 

Signed by: 

Mathew J. Scirè: 
Director, Financial Markets and Community Investment: 

List of Congressional Requesters: 

The Honorable Charles E. Grassley: 
Ranking Member: 
Committee on the Judiciary: 
United States Senate: 

The Honorable Shelley Moore Capito: 
Chairwoman: 
Subcommittee on Financial Institutions and Consumer Credit 
Committee on Financial Services: 
United States House of Representatives: 

The Honorable Judy Biggert: 
Chairman: 
Subcommittee on Insurance, Housing and Community Opportunity 
Committee on Financial Services: 
United States House of Representatives: 

[End of section] 

Appendix I: Objectives, Scope, and Methodology: 

The objectives of our review were to (1) determine the factors that 
have affected costs in the Housing Choice Voucher (voucher) program 
from 2003 through 2011 and the actions Congress and the Housing and 
Urban Department (HUD) took to manage these costs and (2) identify 
additional steps HUD, housing agencies, or policy makers can take to 
limit cost growth in the voucher program and more effectively provide 
decent, safe, and affordable housing. 

To determine the factors that have affected costs in the voucher 
program from 2003 through 2011 and the actions Congress and HUD took 
to manage these costs, we reviewed and analyzed appropriations 
legislation, budget documents--including HUD budget proposals, 
Congressional Research Service reports, monthly statements from the 
Department of the Treasury, and the Office of Management and Budget SF-
133 reports on budget execution and budget resources. We also reviewed 
HUD's annual guidance on the allocation of the program's appropriation 
to housing agencies. We used these sources to determine the annual 
appropriations and outlays over the period. The starting year for our 
analysis reflects the year when Congress began changing the voucher 
program's funding formula.[Footnote 65] 

We analyzed program data that HUD prepared using information derived 
from multiple HUD systems including the Central Accounting and Program 
System (HUDCAPS) and Voucher Management System (VMS) to determine how 
much housing agencies' expenditures changed from 2003 through 2010. 
Specifically, we assessed the extent to which certain factors, such as 
subsidy paid to a landlord, program size (that is, the number of 
assisted households), and administrative expenses, contributed to the 
change in program expenditures over this period. We identified these 
factors by reviewing GAO, HUD, and stakeholder studies. We also 
reviewed prior work by GAO and others to describe what is known about 
the cost-effectiveness and characteristics of vouchers relative to 
other forms of rental housing assistance. 

To identify additional steps HUD, housing agencies, or policy makers 
can take to limit cost growth in the voucher program and more 
effectively provide decent, safe, and affordable housing, we 
identified and reviewed relevant legislation, draft legislation, and 
studies. We analyzed HUD's VMS data on the Net Restricted Assets (NRA) 
balances (or subsidy reserves) of housing agencies as of September 30, 
2011, to determine the extent of housing agencies' "excess" subsidy 
reserves. To derive our estimates of the potential "excess" balances, 
we used HUD's 8.5 percent (about a month) threshold to estimate the 
excess NRA balance. Also, we analyzed HUD data to determine the number 
of housing agencies and amount of funding that Congress offset in 
fiscal years 2008 and 2009 and the additional funding Congress 
appropriated for and HUD provided to certain housing agencies in 2009. 
Further, we visited nine housing agencies in Massachusetts. We 
selected these housing agencies based on Massachusetts' use of both 
local and regional housing agencies to provide voucher assistance and 
the housing agencies' proximity to one another. In addition, we 
interviewed 31 of the 35 housing agencies participating in the Moving 
to Work (MTW) demonstration program to identify the activities the 
agencies had implemented in their voucher programs to reduce program 
costs and introduce efficiencies in the program.[Footnote 66] For 
example, as part of these interviews, we identified alternate rent 
structures these agencies had implemented or proposed. 

We also evaluated the cost and policy implications of three types of 
programmatic reforms to the voucher program: increasing minimum rents, 
changing the percent of income tenants pay toward rent, and requiring 
tenants to pay a percentage of fair market rent. In identifying and 
assessing these programmatic reforms, we reviewed proposals included 
in draft legislation and HUD, Congressional Budget Office, and housing 
industry group reports. We also considered reforms certain agencies 
have implemented. To estimate the effects of these alternative 
approaches to calculating tenant payments on the subsidy levels that 
result, we analyzed a December 2010 extract of tenant records from 
HUD's Public and Indian Housing Information Center (PIC). These 
records contain information about participating households, as of 
December 2010, including information on gross and adjusted income 
levels, housing unit size and rent, tenant contributions and housing 
assistance payments, as well as information on age, sex, and 
disability status of each household member. To focus on the core of 
the assisted household population, we examined only those households 
with five or fewer members, and living in units with one, two or three 
bedrooms. We determined the elderly and disability status of each 
household. Specifically, we defined a household as an elderly 
household if either of the first two household members (the head of 
household and possibly a spouse or co-head) were age 62 or over, and 
we placed a household in disability status if any of the five members 
were identified as having a disability. For the identified subsidy 
alternatives, we calculated an alternative tenant contribution using 
information on income and applicable fair market rent in the PIC file 
as appropriate, and calculated the resulting assistance payment. (The 
assistance payment is the difference between the lesser of the payment 
standard and gross rent, and the tenant payment, subject to any 
existing minimum tenant payments.) We did not consider the possible 
effects of any change in household behavior, either in terms of 
continued participation in the voucher program or in choice of housing 
unit or rent level that could be induced by changes in tenant 
contributions. 

In conducting our work, we assessed the reliability of datasets 
provided by HUD, including data files derived from HUDCAPS, VMS, and 
PIC. Specifically, we performed basic electronic testing of relevant 
data elements, such as housing assistance payment amounts, total 
tenant payment, and unit months leased. We reviewed HUD's data 
dictionaries, instructions, and other relevant documentations. We also 
interviewed HUD officials knowledgeable about the data to obtain 
clarifications about key variables and calculation rules. Where 
possible, we compared our results with other sources to ensure the 
reasonableness of the information. We determined that the data were 
sufficiently reliable for the purpose of this report. 

Finally, for all of our objectives, we interviewed HUD officials and 
consulted with one academic and officials from various housing groups 
including the Center on Budget and Policy Priorities, Council of Large 
Public Housing Authorities, National Low-Income Housing Coalition, 
National Association of Housing Redevelopment Officials, Public 
Housing Authorities Directors Association, Quadel Consulting, and the 
Urban Institute. Further, we contacted 53 housing agencies that 
administer the voucher program. In selecting these housing agencies, 
we considered the number of authorized vouchers, location (that is, 
HUD-defined regions), and leasing and spending rates for the voucher 
program as of March 2011. 

We conducted this performance audit from February 2011 through March 
2012 in accordance with generally accepted government auditing 
standards. Those standards require that we plan and perform the audit 
to obtain sufficient, appropriate evidence to provide a reasonable 
basis for our findings and conclusions based on our audit objectives. 
We believe that the evidence obtained provides a reasonable basis for 
our findings and conclusions based on our audit objectives. 

[End of section] 

Appendix II: Comment from U.S. Department of Housing and Urban 
Development: 

U.S. Department of Housing and Urban Development: 
Office of Public And Indian Housing: 
Washington, DC 20410-5000: 

March 14, 2012: 

Mr. Mathew J. Scirè: 
Director: 
Financial Markets and Community Investment: 
U.S. Government Accountability Office: 
441 G Street, N.W., Room 2440: 
Washington, D.C. 20548:  

Dear Mr. Scirè: 
 
Thank you for the opportunity to comment on GAO's draft report on 
Options to Increase Housing Choice Program Efficiencies, GAO-12-300. 
GAO's review of the Housing Choice Voucher program provides an 
accurate assessment of the program and its current outcomes. There are 
a few items, however, to which Housing and Urban Development (HUD) 
would like to respond.  

The major points HUD would like to clarify or amplify are as follows:  

1. The title should be "Housing Choice Vouchers", not "Housing 
Vouchers".  

2. GAO's stated purpose is inconsistent with its recommendations. In 
the "Highlights" portion of the report, GAO says the focus is 
increasing efficiencies and simplifying program administration. 
However, two of the recommendations made for rent reform focus on 
reducing costs but have no impact on streamlining program management. 
Specifically, implementing higher minimum rents and increasing tenant 
rents to 35 percent of their adjusted income. Neither of these options 
would reduce administrative burdens for PHAs nor simplify the rent 
determination process.  

3. The discussion of outlay growth could be misleading. On page 9, GAO 
discusses the fact that measured growth in outlays grew 11 percent 
over the evaluation period. However, this growth in outlays reflects 
only a change in HUD disbursement policy and does not relate at all to 
changes in program costs. Starting in 2005, the program transitioned 
to a budget based approach rather than a unit based method. With a 
congressional requirement (2005 Consolidated Appropriations Act HR 
4818/H Report 108-792) to notify PHAs of their full funding allocation 
at the beginning of the year, HUD determined that disbursements would 
be made on a monthly basis assuming 1/12 of renewal funding 
eligibility, on the basis that PHAs were responsible for managing 
within the funds available. PIH Notice 2006-3 went on to describe that 
PHAs were no longer required to submit year-end settlements, but must 
maintain all excess disbursed funds in an Undesignated Fund Account 
(the pre-cursor of the Net Restricted Assets (NRA) and Unrestricted 
Net Assets (UNA)) to be used only for eligible program expenses and 
HAP costs.  

The important point illustrated in the GAO text is that housing agency 
expenditures grew at essentially the same rate as the growth in 
appropriations. 

4. The impact of changes in state level assistance is understated. On 
page 12, GAO makes reference to "one housing agency in California... 
that reported state cuts to social welfare programs, including those 
that provide direct cash assistance." Of course, cuts in TANF and SSI 
in California affect all of the agencies in California — a state that 
serves twenty percent of the HCV assisted households. HUD estimates 
that, on average, reductions in TANF and SSI resulted in monthly 
increases of $17 and $5, respectively, in housing subsidy for families 
receiving those forms of assistance. 

5. The discussion of reserves could be misleading. A portion of the 
reserve total cited has not been accumulated over time. This funding 
consists of recently issued new increment funding where insufficient 
time has passed for expenditure. Recently awarded Contract Renewal set 
aside funding also comprises a portion of the total reserve figure. 
These vouchers would thus be protected in the upcoming rescission, as 
would funding deemed a necessary cushion against unforeseen 
circumstances. The 9/30/2011 report takes the 9/30 NRA contractor's 
report balance of $1.88 billion and excludes: 1) all 2011 new 
increments; 2) the set-aside which was not awarded until September 
2011; and 3) a reserve allowance of 8.5% of ABA (approximately one 
month). This results in an "excess" reserve balance of approximately 
$606 million. The 2012 Appropriations Act subsequently recaptured 
these reserves ($650 million) through offset to 2012 funding 
eligibility. 

6. Reserve accumulation issues raised did not address several of HUD's 
ongoing efforts to address this issue. No mention was made of HUD's 
optimization efforts. These efforts have already helped PHAs project 
the funding necessary to maintain desired leasing levels. The 
optimization tool takes into account factors such as attrition, PHA 
reserves and new vouchers awarded to the PHA. 

More importantly, HUD is in the process of implementing new cash 
management procedures for the disbursement of HCV funding. The process 
of disbursing only the funds required to meet current HAP costs which 
began in January 2012 (based upon the most recent validated quarter of 
leasing and expense data from VMS) will result in the re-establishment 
of HUD-held reserves, whereby unused HAP funds will remain obligated 
but undisbursed at the HUD level rather than held by PHAs. Existing 
net restricted asset (NRA) balances currently held by PHAs will be 
transitioned back to program reserves. This change will reduce program 
risk and provide for greater transparency and accountability regarding 
the use and availability of program reserves. 

7. Reporting Reserve levels to Congress. HUD currently provides 
quarterly reports to the Congressional Budget Office on reserve levels. 

Thank you for your consideration of these comments. Please do not 
hesitate to contact me at (202) 402-1380 should you have any questions. 

Sincerely, 

Signed by: 

Milan M. Ozdinec: 
Deputy Assistant Secretary: 
Office of Housing and Voucher Programs: 

[End of section] 

Appendix III: GAO Contact and Staff Acknowledgment: 

GAO Contact: 

Mathew J. Scirè, (202) 512-8678, sciremj@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Daniel Garcia-Diaz, Acting 
Director; Stephen Brown, William Chatlos, Karen Jarzynka-Hernandez, 
Cory Marzullo, John McGrail, Josephine Perez, and Barbara Roesmann 
made key contributions to this report. 

[End of section] 

Footnotes: 

[1] The voucher program is authorized under Section 8 of the United 
States Housing Act of 1937, as amended. Very low-income households are 
those with incomes at or below 50 percent of the area median income; 
extremely low-income households are those with incomes at or below 30 
percent of the area median income. 

[2] Funding for 2003 and 2004 was provided through the Housing 
Certificate Fund, which accounted for both the project-based Section 8 
and the voucher programs. As a result, we were unable to determine the 
appropriated and outlay amounts for the programs for these years. 

[3] Expenditure data for 2011 were not available at the time we 
conducted our analysis. 

[4] Budget authority is the authority federal law provides to enter 
into financial obligations that will result in immediate or future 
outlays involving federal government funds. 

[5] HUD has standards for housing quality. For example, the voucher 
program regulations set basic housing quality standards that all units 
must meet (1) before assistance can be paid on behalf of a household, 
and (2) at least annually during the assisted tenancy. HUD housing 
quality standards include requirements for all unit types, including 
those in single-family and multifamily dwellings. See 24 C.F.R. § 
982.401. 

[6] See HUD, Office of Policy Development and Research, Worst Case 
Housing Needs 2009: Report to Congress (Washington, D.C.: February 
2011). In 2009, more than 7 million renters had worst case housing 
needs, up 20 percent from 2007 levels and almost 42 percent from 2001 
levels. 

[7] See 42 U.S.C. 1437n (b)(1). 

[8] Under 42 U.S.C. 1437f(c)(1), HUD annually must publish fair market 
rents for the voucher program. See related regulations at 24 C.F.R. 
Part 888. 

[9] By law, whenever a household moves to a new unit where the gross 
rent exceeds the payment standard, the household may not pay more than 
40 percent of its adjusted monthly income for rent. 

[10] Laws and HUD regulations provide 44 different income exclusions 
and deductions: (1) HUD regulations cite 20 income sources to be 
excluded when determining households' eligibility for assistance and 
calculating tenant rents. See 24 C.F.R. § 5.609. (2) Under various 
statutes, 19 other income sources qualify as exclusions. (3) In 
addition, program administrators (housing agencies) must apply 5 
income deductions, which reduce the amount of income that can be 
considered in calculating tenant rents. See 24 C.F.R. § 5.611. Once 
program administrators have collected information from households on 
income and applicable exclusions and deductions, HUD policy requires 
that they independently verify this information ("third-party" 
verification). After verifying households' income information, program 
administrators must compute the amounts the households will pay in 
rent. See 24 C.F.R. § 5.628. 

[11] Housing agencies report NRA (or subsidy reserves) under Housing 
Assistance Payment (HAP) Equity on their income statements. NRA is the 
amount of HAP Equity (subsidy) for the voucher program through the end 
of the housing agency's fiscal year and equals total HAP revenues 
minus total HAP expenses for eligible unit months (or vouchers) leased 
on a calendar-year basis. 

[12] See Omnibus Consolidated Rescissions and Appropriations Act of 
1996 (P.L. 104-134, 110 Stat. 1321), April 26, 1996. In addition, GAO 
is reviewing the effectiveness of MTW, including assessing the steps 
HUD has taken to help ensure that participating agencies address 
statutory purposes and meet statutory requirements, and the potential 
benefits of and concerns about expanding the number of MTW agencies. 

[13] This funding flexibility ("fungibility") only applies to MTW 
agencies' regular vouchers. Like regular housing agencies, MTW 
agencies separately must maintain allocated amounts that exceed the 
cost of their authorized special-purpose voucher program activities in 
reserve accounts and only may use those balances to pay for authorized 
program activities in subsequent years. 

[14] This "30-percent rule" has its origins in the "Brooke Amendment"--
in 1969, then-Senator Edward Brooke of Massachusetts offered, and 
Congress passed, an amendment to the United States Housing Act of 
1937, as amended, that mandated that no family would have to pay more 
than 25 percent of its income toward rent in federally assisted 
housing. In 1981, Congress increased the maximum to 30 percent. 

[15] Housing agencies must establish an application and selection 
process that treats applicants for voucher assistance fairly and 
consistently and provides an effective method for determining 
eligibility. However, voucher program regulations provide flexibility 
for each housing agency to develop an application and selection 
process tailored to its particular circumstances, including the 
ability to establish local preferences for assistance. See 24 C.F.R. § 
982.207. 

[16] For a detailed discussion of the relevant literature, see Edgar 
O. Olsen, "The Cost-Effectiveness of Alternative Methods of Delivering 
Housing Subsidies," presentation at the thirty-first annual APPAM 
research conference (Charlottesville, Virginia: Oct. 30, 2009). 

[17] See GAO, Federal Housing Assistance: Comparing the Costs and 
Characteristics of Housing Programs, [hyperlink, 
http://www.gao.gov/products/GAO-02-76] (Washington, D.C.: Jan. 31. 
2002). We analyzed the characteristics and costs of the housing under 
six federal housing programs that continue to increase the number of 
assisted households: the voucher program, Low-income Housing Tax 
Credits, HOPE VI, Section 202, Section 811, and Section 515. With the 
exception of the voucher program, each of these programs is a 
production program. 

[18] Because of data limitations, we used a different methodology to 
present total costs for the HOPE VI program--a HUD program to 
revitalize distressed public housing. We found that the total 30-year 
cost of a HOPE VI unit with an average size of 2.4 bedrooms was about 
27 percent more expensive than vouchers. 

[19] See HUD, Office of Policy Development and Research, Economic Cost 
Analysis of Different Forms of Assisted Housing (Washington, D.C.: 
December 2000) and The President's Commission on Housing, The Report 
of the President's Commission on Housing (Washington, D.C.: April 
1982). 

[20] See HUD, Office of Policy Development and Research, Targeting 
Housing Production Subsidies: Literature Review (Washington, D.C.: 
December 2003). 

[21] Authorized voucher leasing levels for housing agencies are 
outlined in their Annual Contributions Contracts with HUD. 

[22] HUD provided this reserve funding separately to the funding 
reserves (NRA) housing agencies can accumulate and to which we refer 
throughout this report. 

[23] Beginning in 2003, housing agencies were prohibited from leasing 
more vouchers than were authorized in their contracts. 

[24] Later in this report, we discuss how continued monitoring and 
reduction of housing agencies subsidy reserves could reduce the need 
for new appropriations for or increase the number of households 
assisted under the voucher program. 

[25] See GAO, HUD Rental Assistance: Progress and Challenges in 
Measuring and Reducing Improper Rent Subsidies, [hyperlink, 
http://www.gao.gov/products/GAO-05-244] (Washington, D.C.: Feb. 18, 
2005). 

[26] Later in this report, we discuss how HUD's administrative fee 
study could be used to streamline administrative requirements and 
reduce administrative costs. 

[27] Because HUD has allocated subsidy reserve funds to housing 
agencies, congressional rescissions must be implemented through an 
offset, whereby Congress requires housing agencies to spend down their 
reserve funds in order to make up for reductions in the appropriated 
amount. 

[28] See HUD, Notice PIH 2011-67, Implementation of New Cash 
Management Requirements for the Housing Choice Voucher Program (Dec. 
9, 2011). 

[29] Because housing agencies' reserves are resources that HUD has 
disbursed and expended, HUD effectively recaptures any excess reserves 
by reducing or offsetting the housing agencies' funding allocation in 
another year. 

[30] As previously discussed, Congress rescinded $750 million from 
HUD's 2009 appropriation for the voucher program, and HUD offset this 
amount from housing agencies' reserve accounts. Similarly, in 2008, 
Congress rescinded the amount in the housing agencies' reserves 
accounts that exceeded 7 percent of the amount of the program's 2007 
renewal funding (about $723 million), and HUD offset this amount from 
housing agencies' reserve accounts. In 2005, Congress directed HUD to 
reduce the housing agencies' reserves account to no more than 1 week 
(approximately 2 percent) of subsidy funding. The Affordable Housing 
and Self-Sufficiency Improvement Act of 2012 (January 31, 2012, draft) 
would permit housing agencies to retain subsidy reserves in an amount 
equal to no less than 6 percent of their current year's annual budget 
authority. This amount would be exempt from any offsets imposed by HUD. 

[31] GAO, Section 8 Tenant-Based Housing Assistance: Opportunities to 
Improve HUD's Financial Management, [hyperlink, 
http://www.gao.gov/products/GAO/RCED-98-47] (Washington, D.C.: Feb. 
20, 1998). In this report we highlighted that the Office of Management 
and Budget guidance on budget formation instructs agencies to consider 
available funding on hand before requesting new funding. 

[32] GAO, Government Performance: GPRA Modernization Act Provides 
Opportunities to Help Address Fiscal, Performance, and Management 
Challenges, [hyperlink, http://www.gao.gov/products/GAO-11-466T] 
(Washington D.C., Mar. 16, 2011). 

[33] Based on our analysis, MTW agencies had no excess (nonfungible) 
subsidy reserves. 

[34] According to HUD, to fund the rescission, the department will 
offset reserve balances that exceed approximately one month of housing 
agencies' 2012 eligible amounts. See HUD, Notice PIH 2012-9, 
Implementation of the Federal Fiscal Year 2012 Funding Provisions for 
the Housing Choice Voucher Program( Feb. 8, 2012). For housing 
agencies that administer 50 or fewer vouchers, the offset will equal 
that portion of the reserve balance that exceeds approximately 6 
months of housing agencies' 2012 eligible amounts. 

[35] House Financial Services Committee, Transforming Rental 
Assistance, Testimony of Shaun Donovan Secretary of the Department of 
Housing and Urban Development, 111thCong., 2nd sess., 2010. 

[36] See 24 C.F.R. §982.516. 

[37] Insurance, Housing, and Community Opportunity Subcommittee, House 
Financial Services Committee, Legislative Proposals to Reform the 
Housing Choice Voucher Program, Testimony of Sandra B. Henriquez, 
Assistant Secretary of the Department of Housing and Urban 
Development, 112th Cong., 1st sess., 2011. HUD's fiscal year 2012 
budget request also proposed extending the time between re-
certifications for households on fixed incomes. 

[38] HUD, Statement of Work: Housing Choice Voucher Program 
Administrative Fee Study, Solicitation Number RCHI00992, (Washington, 
D.C.: Feb. 8, 2010). 

[39] Testimony of Sandra B. Henriquez, Assistant Secretary of the 
Department of Housing and Urban Development, 112th Cong., 1st sess., 
2011. 

[40] HUD, Office of Public and Indian Housing and Office of Policy 
Research and Development, Report to Congress Moving to Work: Interim 
Policy Applications and the Future of the Demonstration (Washington, 
D.C.: August 2010). 

[41] See Affordable Housing and Self-Sufficiency Improvement Act of 
2012, discussion draft dated Jan. 31, 2012. 

[42] GPRA, §§ 2(a)(1), 2(b)(5). The GPRA Modernization Act of 2010 
(GPRAMA) updated the federal government's performance measurement 
framework established in GPRA. 

[43] For a detailed discussion of how we selected which rent reform 
options to include in our analysis, see the detailed scope and 
methodology in appendix I. Academics have advanced other rent reform 
options that we did not include in this report. For example, see Amy 
Crews Cutts and Edgar O. Olsen, "Are Section 8 Housing Subsidies Too 
High?" Journal of Housing Economics, vol. 11 (2002): 214-243. 

[44] If a housing agency adopts a minimum rent policy, the housing 
agency must grant exemptions from the requirement to any household 
that the housing agency determines is unable to pay the amount because 
of financial hardship, unless the hardship is temporary. See 24 CFR 
§5.630. 

[45] Our analysis showed that 0.5 percent of all assisted households 
pay no rent and 1.3 percent pay from $0 to $50. 

[46] A rent structure based on gross income would eliminate the 
deductions and exclusions to income that households currently may 
claim. 

[47] In past work we highlighted some of the limitations of HUD's 
process for estimating fair market rents and ways to improve the 
accuracy of these estimates. See GAO, Rental Housing: HUD Can Improve 
Its Process for Estimating Fair Market Rents, [hyperlink, 
http://www.gao.gov/products/GAO-05-342] (Washington, D.C.: Mar. 31, 
2005). 

[48] Our analysis of current household characteristics showed that 81 
percent of nonelderly, nondisabled households are comprised of at 
least one parent and one child. As a result, we refer to these 
households as "families with children" throughout this report. 

[49] Our analysis of current household characteristics and incomes 
showed that this approach would save approximately $691 million 
annually. 

[50] We considered a household disabled if any member of the household 
had a disability. 

[51] We segmented the incomes of assisted household into 10 equal 
groups. We considered assisted households in the lowest tenth as the 
lowest-income households and assisted households in the highest tenth 
as the highest-income households. 

[52] In addition, under all other rent reform scenarios, less than 0.5 
percent of households would lose their subsidies. 

[53] We evaluated the effect (in terms of costs and households served) 
of requiring assisted households to pay 12, 15, 20, 30, and 35 percent 
of the fair market rent. (The October 5, 2011, draft Section 8 Savings 
Act includes a provision that would require certain households to pay 
the greater of $75 or 12 percent of the fair market rent.) Our 
analysis showed that requiring households to pay 12, 15, 20, or 30 
percent of the fair market rent would increase program costs and 
reduce the number of households served. For example, requiring 
households to pay 12 percent of the fair market rent would increase 
program costs by approximately $4.1 billion and reduce the number of 
households served by approximately 451,000; requiring households to 
pay 30 percent of the fair market rent would increase program costs by 
approximately $165 million and reduce the number of households served 
by approximately 27,000. 

[54] In the early stages of the MTW demonstration, several housing 
agencies experimented with time limits on assistance as a means of 
encouraging self sufficiency among assisted households. All of these 
housing agencies largely abandoned time limits. However, some were in 
favor of mandatory minimum rents or subsidies that decreased over 
time, regardless of a household's income. Applied Real Estate 
Analysis, Inc. and The Urban Institute, The Experiences of Public 
Housing Agencies That Established Time Limits Policies under the MTW 
Demonstration (Washington, D.C.: May 2007). 

[55] See GAO, HUD Rental Assistance: Progress and Challenges in 
Measuring and Reducing Improper Rent Subsidies, [hyperlink, 
http://www.gao.gov/products/GAO-05-224] (Washington, D.C.: Feb. 18, 
2005). 

[56] The MTW statute requires that "an application to participate in 
the demonstration shall include a plan that… includes criteria for 
establishing a reasonable rent policy, which shall be designed to 
encourage employment and self-sufficiency by participating families, 
consistent with the purpose of this demonstration, such as by 
excluding some or all of a family's earned income for purposes of 
determining rent." See HUD, Policy Development and Research, Report to 
Congress, Moving to Work: Interim Policy Applications and the Future 
of the Demonstration (Washington, D.C.: August 2010). 

[57] The National Low Income Housing Coalition advocates for the 
affordable housing needs of low-income people. 

[58] Subcommittee on Insurance, Housing, and Community Opportunity, 
House Committee on Financial Services, testimony of Linda Couch, 
Senior Vice President for Policy, National Low Income Housing 
Coalition,112th Cong., 1st sess., 2011. 

[59] See HUD, Office of Policy, Program, and Legislative Initiatives, 
Rebalancing HUD's Oversight and Small PHAs' Regulatory Burdens 
(Washington, D.C.: 2008). About 58 percent of housing agencies 
administer 400 or fewer vouchers. 

[60] See HUD, Office of Policy Development and Research, Section 8 
Administrative Fees: A Report to Congress (Washington, D.C.: June 
1994). 

[61] Approximately 850 housing agencies only administer vouchers. 

[62] See Bruce Katz and Margery Austin Turner, Who Should Run the 
Housing Voucher Program? A Reform Proposal. Working paper prepared for 
the Brookings Institution, Center on Urban and Metropolitan Policy 
(Washington, D.C.: November 2000). 

[63] Williams v. Hanover Housing Authority, 871 F. Supp. 527 (D. Mass. 
1994); see also Williams v. Hanover Housing Authority, 926 F. Supp. 10 
(D. Mass. 1996), rev'd and remanded on other grounds, 113 F.3d 1294 
(1ST. Cir. 1997). 

[64] See Who Should Run the Housing Voucher Program? A Reform Proposal. 

[65] Funding for 2003 and 2004 was provided through the Housing 
Certificate Fund, which accounted for both Section 8 programs--the 
project-based and tenant-based programs. As a result, we were unable 
to determine the appropriated and outlay amounts for the programs for 
these years. In addition, in 2005 the Housing Certificate Fund was 
split into two accounts, one of which was the tenant-based rental 
assistance account. Because of the split, about $4.2 billion from the 
advance appropriation enacted in fiscal year 2004 and available in 
2005 does not appear in this account. Instead, it appears in the 
Housing Certificate Fund where it was appropriated. Total available 
resources for tenant-based rental assistance in fiscal year 2005 were 
$14.8 billion. 

[66] We did not contact the Philadelphia Housing Authority because the 
agency currently is under administrative receivership. We did not 
contact the Boulder Housing Partners, and Lexington-Fayette Urban 
County Housing Authority because they had not yet entered into MTW 
agreements with HUD at the time of our analysis. In addition, the 
Housing Authority of the County of Santa Clara administers its own MTW 
program as well as the Housing Authority of the City of San Jose's MTW 
program. 

[End of section] 

GAO’s Mission: 

The Government Accountability Office, the audit, evaluation, and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the 
performance and accountability of the federal government for the 
American people. GAO examines the use of public funds; evaluates 
federal programs and policies; and provides analyses, recommendations, 
and other assistance to help Congress make informed oversight, policy, 
and funding decisions. GAO’s commitment to good government is 
reflected in its core values of accountability, integrity, and 
reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO’s website [hyperlink, http://www.gao.gov]. Each 
weekday afternoon, GAO posts on its website newly released reports, 
testimony, and correspondence. To have GAO e-mail you a list of newly 
posted products, go to [hyperlink, http://www.gao.gov] and select “E-
mail Updates.” 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of 
production and distribution and depends on the number of pages in the 
publication and whether the publication is printed in color or black 
and white. Pricing and ordering information is posted on GAO’s 
website, [hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or 
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card, 
MasterCard, Visa, check, or money order. Call for additional 
information. 

Connect with GAO: 

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 
Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; 
E-mail: fraudnet@gao.gov; 
Automated answering system: (800) 424-5454 or (202) 512-7470. 

Congressional Relations: 

Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 512-4400
U.S. Government Accountability Office, 441 G Street NW, Room 7125
Washington, DC 20548. 

Public Affairs: 
Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149 
Washington, DC 20548.