Assisted Housing: Occupancy Restrictions On Persons With Disabilities
(Letter Report, 11/12/1998, GAO/RCED-99-9).

The Housing and Community Development Act of 1992 allows the owners of
federally assisted housing projects to establish occupancy policies that
favor elderly tenants over nonelderly tenants with disabilities. These
owners are not required to obtain approval from the Department of
Housing and Urban Development (HUD) before imposing such a restriction,
nor to notify HUD once the restrictions occurs.  As a result, little
information is available on the law's effect.  However, concerns have
been raised that the law may make it harder for nonelderly persons with
disabilities to obtain affordable housing. Since fiscal year 1997,
Congress has appropriated funds for new Section 8 rental housing
certificates and vouchers for the exclusive use of nonelderly persons
with disabilities.  This report discusses (1) the extent to which the
occupancy policies of eligible projects restrict occupancy to the
elderly and the portion of units in eligible projects actually occupied
by nonelderly persons with disabilities and (2) the use of Section 8
certificates and vouchers to help nonelderly persons with disabilities
affected by the act.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-99-9
     TITLE:  Assisted Housing: Occupancy Restrictions On Persons With
	     Disabilities
      DATE:  11/12/1998
   SUBJECT:  Persons with disabilities
	     Housing for the elderly
	     Housing for the disabled
	     Public housing
	     Housing programs
	     Rent subsidies
	     State-administered programs
	     Rent policies
	     Unobligated budget balances
IDENTIFIER:  HUD Section 8 Voucher Program
	     HUD Section 8 Certificate Program

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************

Cover
================================================================ COVER

Report to Congressional Requesters

November 1998

ASSISTED HOUSING - OCCUPANCY
RESTRICTIONS ON PERSONS WITH
DISABILITIES

GAO/RCED-99-9

Assisted Housing for Persons With Disabilities

(385724)

Abbreviations
=============================================================== ABBREV

  HUD - Department of Housing and Urban Development
  VA - Veterans Administration

Letter
=============================================================== LETTER

B-280681

November 12, 1998

The Honorable Jerry Lewis
Chairman
The Honorable Louis Stokes
Ranking Minority Member
Subcommittee on VA, HUD, and
 Independent Agencies
Committee on Appropriations
House of Representatives

The Honorable Rodney P.  Frelinghuysen
House of Representatives

Two provisions of the Housing and Community Development Act of 1992
allow the owners of certain federally assisted housing projects to
establish occupancy policies that favor elderly tenants over
nonelderly tenants with disabilities.  Specifically, Sections 651 and
658 of Title VI of the 1992 act allow the owners of projects built
primarily for the elderly (defined as those aged 62 or older) to
restrict the occupancy of nonelderly persons with disabilities.\1

These owners are not required to receive approval from the Department
of Housing and Urban Development (HUD) before imposing such a
restriction, nor to notify HUD once the restriction occurs.\2
Therefore, little information on Title VI's effect is available. 
However, concerns have been raised that Title VI may make it more
difficult for nonelderly persons with disabilities to obtain
affordable housing.  Since fiscal year 1997, the Congress has
appropriated funds for incremental (new) Section 8 rental housing
certificates and vouchers for the exclusive use of nonelderly persons
with disabilities.\3

In its July 11, 1997, report (H.  Rep.  105-175) accompanying its
fiscal year 1998 appropriations bill, the Subcommittee on VA, HUD,
and Independent Agencies, House Committee on Appropriations, directed
GAO to assess the effects of Sections 651 and 658 on nonelderly
persons with disabilities.  In response, this report provides
information on (1) the extent to which the occupancy policies of
eligible projects restrict occupancy to the elderly and the portion
of units in eligible projects actually occupied by nonelderly persons
with disabilities and (2) the use of Section 8 certificates and
vouchers to assist nonelderly persons with disabilities affected by
the act. 

To obtain information on the effects of Sections 651 and 658, we
surveyed a random sample of 1,223 of the 7,696 privately owned,
HUD-assisted housing projects that are potentially eligible to
restrict the number of nonelderly tenants with disabilities.\4 Our
survey results are generalizable to approximately 6,080 of the
potentially eligible projects.\5 This report discusses the responses
of those projects whose managers told us that the projects were
originally designed primarily for the elderly.  We estimate that
there are 4,157 such projects.  To obtain information on HUD's use of
Section 8 certificates and vouchers to assist persons with
disabilities affected by the act, we interviewed appropriate HUD
officials and obtained documentation where available.  We also
interviewed officials of associations representing (1) persons with
disabilities, (2) the elderly, and (3) owners of privately owned
assisted housing.  Appendix I describes our survey methodology in
more detail.  Appendix II presents our survey and the aggregated
responses to our questions.  Appendix III provides the sampling
errors for the responses to our survey. 

--------------------
\1 Occupancy by elderly tenants with disabilities is not restricted
because they meet the age qualification. 

\2 The owners and managers of private housing that receive federal
assistance who decide to restrict the occupancy of nonelderly persons
with disabilities pursuant to the 1992 act may not force such persons
to move out of these units. 

\3 Persons using Section 8 certificates and vouchers, which are
administered by local housing authorities, pay a portion of their
income--usually 30 percent--toward renting privately owned apartments
that meet HUD's rent requirements and housing quality standards.  HUD
generally pays the difference between the tenants' portion and the
total rent charged for a unit or, in the case of vouchers, a rent
standard established by HUD. 

\4 We did not assess whether the projects' policies were in
compliance with HUD's regulations. 

\5 Almost 80 percent of the projects we surveyed by mail responded. 
We also conducted a telephone survey of a sample of the projects that
did not respond to our written survey to determine whether their
answers differed from those who responded by mail.  We found some
significant differences between the two groups and have noted them,
where appropriate, in this report.  Sampling errors define the upper
and lower bounds of the estimates made for our survey results. 
Sampling errors were calculated at the 95-percent confidence level. 
This means that 19 out of 20 times, the sampling procedures used
would produce a range that includes the true value.  Wherever
sampling errors were less than 5 percent, we do not present them in
this letter.  Sampling errors greater than 5 percent follow their
corresponding estimates. 

   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

The majority of housing projects designed for the elderly have not
used the 1992 act to restrict the occupancy of nonelderly persons
with disabilities.  Almost three-quarters of the officials for the
projects designed for the elderly reported that they had adopted
their current policies before 1993, when the law went in effect. 
That is, the policies for these projects have not changed as a result
of the act.  A similar number of officials reported that their
projects' occupancy composition has remained about the same over the
last 5 years.  Regardless of when the occupancy policies were
adopted, most projects allow persons with disabilities to reside in
them.  About 42 percent allow the elderly and persons with
disabilities to apply for all units, while about 24 percent reserve
some units for nonelderly persons with disabilities.  Elderly tenants
currently occupy about 90 percent of the available units, while
nonelderly persons with disabilities occupy 8 percent.  Advocates for
the elderly and persons with disabilities and HUD officials generally
were not surprised by our survey results.  Advocates for the elderly
believe that our findings indicate that few projects have used the
act to restrict the occupancy of persons with disabilities.  In
contrast, disability rights advocates believe that many projects were
discriminating against persons with disabilities prior to the 1992
act and that the act merely legitimized those practices. 

Local housing authorities have made limited use of the funds for
Section 8 certificates and vouchers that HUD can award to assist
nonelderly persons with disabilities who are not currently receiving
housing assistance because owners restricted occupancy, as the 1992
act allows.  Of the $45 million appropriated for this purpose in
fiscal years 1997 and 1998, only $8.4 million had been awarded to 13
housing authorities as of September 30, 1998.  HUD officials told us
that these funds were not in greater demand for several possible
reasons.  First, the application process for these funds is more
complicated than the process for funds in the mainstream housing
opportunities program, which is another mechanism available for
housing authorities to apply for certificates and vouchers for
persons with disabilities.  Furthermore, our survey found that very
few owners of private projects were working with staff from their
local housing authority to assist persons with disabilities affected
by Sections 651 and 658.  Finally, housing authorities prefer to
apply for funds from the mainstream program for persons with
disabilities.  In September 1998, HUD transferred $78.3 million of
unobligated Section 8 funds for nonelderly persons with disabilities,
of which $36.6 million was from the $45 million set-aside, into the
mainstream program.  These transferred funds were fully obligated by
the end of fiscal year 1998, according to HUD. 

   BACKGROUND
------------------------------------------------------------ Letter :2

Assisted housing for the elderly (known as elderly housing) has been
developed under several HUD programs, including the Section 202
program, which remains the primary assisted housing program for the
elderly.  All of these programs allow, to some degree, occupancy by
some nonelderly persons with disabilities; however, the eligibility
criteria for persons with disabilities depend on the housing program
and, in some instances, on the year the project was developed and the
definition of disability in effect for that year.  Therefore, some
elderly projects that accept a person with a physical disability may
have the legal discretion to not accept a person with either a
developmental or mental disability.  Many projects allowed persons
with mobility impairments to occupy 10 percent of their units and
elderly persons to occupy the remaining 90 percent.  Other projects,
such as certain Section 202 projects built before 1964, when persons
with physical handicaps became eligible, were not required to accept
any persons with disabilities. 

In response to complaints that the lifestyles of young adults with
disabilities residing in elderly housing conflicted with those of
elderly tenants, the Congress, in Title VI of the Housing and
Community Development Act of 1992, provided that owners of elderly
housing may restrict the number of nonelderly persons with
disabilities who may reside there, and defined the types of housing
programs covered and the types of restrictions permitted.\6

Specifically, Section 651 of Title VI states that eligible elderly
projects with Section 8 project-based assistance may give preference
to elderly families.\7 If preference is given, the project must
reserve some units--10 percent or fewer of the total units--for
nonelderly persons with disabilities under the age of 50.  If there
is an insufficient number of elderly tenants, then the owner may give
preference to persons with disabilities who are near-elderly (ages 50
through 61).  Projects eligible under Section 658 have different
restrictions.  In essence, Section 658 permits elderly projects to
return to their original occupancy policies, as agreed upon with HUD,
if they had over the years allowed tenants other than those
originally eligible to occupy units. 

To offset the potential loss of housing for nonelderly persons with
disabilities and to provide them with greater housing choices, local
housing authorities may apply for HUD's Section 8 rental housing
certificates and vouchers from one of three separate funding
set-asides.  Of the $278.9 million that has been made available for
certificates and vouchers for persons with disabilities since the
1992 act, $136.9 million was set aside for housing authorities that
have HUD-approved allocation plans,\8 $45 million was set aside for
housing authorities to assist those nonelderly persons with
disabilities who have been affected by private projects'
establishment of occupancy policies restricting their number, and $97
million was set aside for the mainstream housing opportunities
program.\9 Each set-aside has different application requirements and
restrictions.  Once housing authorities are awarded funds for a
specific number of certificates or vouchers, they are responsible for
providing the certificates and vouchers to eligible applicants. 

--------------------
\6 Elderly housing generally consists of efficiency apartments and
one-bedroom units, while family housing consists of apartments with
two or more bedrooms.  Some family projects also have efficiency or
one-bedroom apartments for families with one or two members and do
allow such families to reside there.  However, elderly projects
rarely include apartments with more than one bedroom, and their
primary eligibility criteria are age and disability rather than
family size. 

\7 Section 8 project-based assistance, unlike Section 8 certificates
and vouchers, is tied to specific property units.  Rent is paid for
eligible tenants when they occupy a specific housing development or
unit. 

\8 A local housing authority wanting to restrict occupancy must
submit to HUD an allocation plan that, among other things, describes
the local supply and demand of affordable housing for nonelderly
persons with disabilities and the steps the housing authority will
take to secure alternative housing options. 

\9 The mainstream housing opportunities program provides Section 8
certificates and vouchers to persons with disabilities to find
affordable housing of their choice in their communities (hence, the
term "mainstream").  The mainstream program does not have an age
criterion for eligibility. 

   THE 1992 ACT HAS HAD LITTLE
   EFFECT ON OCCUPANCY
   RESTRICTIONS FOR NONELDERLY
   PERSONS WITH DISABILITIES
------------------------------------------------------------ Letter :3

Most elderly projects have not used the 1992 act to restrict the
occupancy of nonelderly persons with disabilities.  Specifically,
occupancy policies and the relative percentages of elderly tenants
and nonelderly tenants with disabilities have not changed much since
the enactment of the 1992 act, according to survey responses. 
Advocates for the elderly and persons with disabilities and HUD
officials differ on whether our survey results show that nonelderly
persons with disabilities have been adversely affected by the act. 

      MOST OCCUPANCY POLICIES
      PRECEDE THE 1992 ACT AND
      ALLOW OCCUPANCY BY SOME
      PERSONS WITH DISABILITIES
---------------------------------------------------------- Letter :3.1

Of an estimated 4,157 projects originally designed for the elderly,
about 71 percent--or 2,968 ( 207) projects with a total of 241,184
units ( 24,017)--had current occupancy policies that preceded the
enactment of the 1992 act.  In other words, these projects did not
use the 1992 act to change their policies on the occupancy of
nonelderly persons with disabilities.  Officials for most of these
projects reported that their current occupancy policies were adopted
when the projects were developed.  The remaining projects included
those that adopted their current policies in 1993 or later but were
built before 1993, those that adopted their current policies in 1993
or later but were built in 1993 or later, and those that did not
provide complete data. 

   Figure 1:  Most Occupancy
   Policies Precede the 1992 Act

   (See figure in printed
   edition.)

\a Respondents did not complete questions on occupancy policy, date
of policy, and/or date the project was built or substantially
rehabilitated. 

Of the projects that were built before 1993 but have adopted policies
since then, 59.9 ( 10.3) percent adopted policies that restrict the
number of nonelderly persons with disabilities, either by limiting
their number or by renting only to elderly tenants.  Another 24.3 (
8.9) percent allowed only the elderly and persons with disabilities
for all units.  The remaining projects selected other policies or did
not respond to our survey. 

Regardless of when a project's occupancy policy was adopted, at least
70 percent of the 4,157 projects built primarily for the elderly
currently accept some persons with disabilities (see fig.  2).  Most
projects had one of three types of policies.  One type of policy
allows persons with disabilities and elderly persons to apply for all
units--42 percent, or 1,738
( 181) projects, had this policy.\10 Under a second type of policy,
projects reserve a percentage or number of units (most commonly 10
percent or fewer of the total units) for nonelderly persons with
disabilities and allow the elderly and/or near-elderly to occupy the
remaining units--24 percent had this policy.\11 A third type of
policy restricts occupancy to the elderly--21 percent had this
policy.  Projects under Section 658 were twice as likely as those
under Section 651 to restrict occupancy to elderly persons. 

   Figure 2:  Most Elderly
   Projects Allow Occupancy by
   Some Persons With Disabilities

   (See figure in printed
   edition.)

\a Respondents did not complete question on occupancy policy. 

\b In this category, 21 percent allow only elderly persons. 

Projects that reserved some units for nonelderly persons with
disabilities generally limited these units to those with physical
disabilities.  That is, officials for almost 70.5 ( 6.8) percent of
the projects that reserved some units reported that only persons with
disabilities requiring physically accessible units were eligible for
occupancy in those reserved units, while officials for about 26 (
6.5) percent of the projects reported that they would consider
persons with disabilities who do not require accessible units.  In
contrast, projects that considered the elderly and persons with
disabilities for all units more often had policies to admit persons
with disabilities regardless of their need for accessible units.\12

--------------------
\10 Section 651 projects that responded to our telephone survey
reported significantly more often than respondents to the mail survey
that they accept both the elderly and persons with disabilities. 
There were no significant differences between Section 658 telephone
and mail survey respondents. 

\11 Sections 651 and 658 projects that responded to our telephone
survey reported significantly less often than respondents to our mail
survey that they reserve units for nonelderly persons with
disabilities, limiting the remaining units to the elderly. 

\12 Section 651 projects that responded to our telephone survey were
significantly more likely to report accepting both disabled persons
who require physically accessible units and those who do not require
such units than projects that responded by mail.  There were no
differences in these policies for Section 658 telephone and mail
survey respondents. 

      PERCENTAGES OF ELDERLY AND
      NONELDERLY WITH DISABILITIES
      AT ELDERLY PROJECTS HAVE
      REMAINED RELATIVELY STABLE
      OVER THE LAST 5 YEARS
---------------------------------------------------------- Letter :3.2

According to officials for most of the projects, elderly persons
currently occupy 90 percent of units, while nonelderly persons with
disabilities occupy 8 percent overall.\13 That is, of the 321,152 (
25,979) units of federally supported housing located in elderly
housing projects, about 287,629 ( 24,151) units are occupied by the
elderly and 25,796 ( 4,169) units are occupied by nonelderly persons
with disabilities.  Nonelderly, nondisabled tenants reside in fewer
than 2 percent of the units.\14

The percentages of elderly residents and residents with disabilities
have remained the same over the last 5 years, according to a majority
of those responding.  About three-quarters (77 percent) of the
officials of elderly projects reported that the percentage of elderly
tenants has remained the same; and 70 percent reported that the
percentage of tenants with disabilities (regardless of age) has
remained the same.  Increases in the percentage of elderly tenants
were experienced by 11 percent of the projects, while officials for
21 percent of the projects reported that they had increases in the
percentage of tenants with disabilities. 

--------------------
\13 Another 4.6 percent of elderly projects did not provide
sufficient data on occupancy for this analysis. 

\14 Officials for some projects reported vacancies, although our
survey did not ask about vacant units. 

      ADVOCATES AND HUD PROVIDE
      DIFFERENT REASONS FOR
      POTENTIALLY LOW IMPACT OF
      1992 ACT
---------------------------------------------------------- Letter :3.3

We provided the preliminary results of our survey to national
representatives of, and advocates for, persons with disabilities, the
elderly, and owners of projects that are eligible under Sections 651
and 658 to restrict the occupancy of persons with disabilities, as
well as to HUD officials.  These experts offered different
interpretations of our findings--that most projects adopted their
current occupancy policies before the 1992 act's enactment, that
occupancy patterns have been stable, and that occupants are
predominantly elderly.  Specifically, they disagreed over the meaning
of stability in policies and occupancy.  Some believed that these
results showed that nonelderly persons with disabilities were being
adequately served, while others believed that they showed that
project owners were continuing past discriminatory practices. 

Officials representing project owners and the elderly believed that
the survey data confirm their experience that projects have not
generally used the 1992 act to restrict occupancy by persons with
disabilities.  They believed that it is a misperception that project
owners and managers favor elderly tenants over younger tenants with
disabilities.  Instead, they explained that many younger persons with
disabilities (1) do not want to live with elderly persons and prefer
other types of housing, including housing projects and group homes
just for persons with disabilities or (2) want to choose their own
housing using Section 8 certificates and vouchers.  Therefore,
younger persons with disabilities are not applying in higher numbers
for admission to elderly housing.  The officials believed that our
survey results confirm these views because our survey found that
nonelderly persons with disabilities represented 8 percent of
applicants on elderly projects' waiting lists.\15 As a result,
elderly housing is primarily composed of elderly tenants. 

In contrast, advocates for persons with disabilities maintained that
few projects used the 1992 act to restrict the occupancy of such
persons because many projects were already imposing such
restrictions--which could have been illegal--before the 1992 act was
passed.  In other words, the 1992 act codified existing occupancy
practices, regardless of whether those practices were legal.  The
advocates maintained that such practices occurred because of
discrimination against people with disabilities and ignorance of the
law and HUD's occupancy regulations.  In their view, the project
owners were and are able to maintain such potentially illegal
practices because HUD is not required to approve changes to the
projects' occupancy policies and because HUD is not enforcing
compliance with its occupancy regulations.  The advocates also
believed that project owners use "redlining," telling potential
applicants that a waiting list is closed or a project is accepting
persons with only certain types of disabilities, such as physical
disabilities.  Finally, the advocates said that most low-income
people will not file complaints against a project out of ignorance or
fear of retaliation or intimidation.  If they do file a complaint,
they are more likely to do so with a local organization than with
HUD.  As a result, HUD does not know the extent to which persons with
disabilities experience discrimination. 

Finally, HUD officials believed that our survey results confirm that,
overall, persons with disabilities have not to date been harmed by
the 1992 act.  HUD program officials disagreed with the assertion of
the disability rights advocates that the 1992 act codified
potentially illegal practices.  While the officials acknowledged that
some projects maintained illegal occupancy practices before the act's
passage, they did not believe that such practices were widespread. 
Otherwise, they believed, significantly more complaints would have
been brought to HUD's attention.  However, HUD officials were
surprised by some of our findings.  For instance, they did not
understand why 11 percent of the projects eligible under Section 651
reported that they accept only elderly tenants.  According to HUD's
program officials, Section 651 projects that accept only the elderly
should be rare.  Furthermore, they were surprised that the percentage
of nonelderly persons with disabilities on waiting lists was so
small. 

--------------------
\15 Waiting lists are an imperfect measure of demand because they
reflect demand only at the time the individuals are listed or lists
are updated.  They do not reflect the demand of those persons who
need housing but, for various reasons, are not listed. 

   USE OF SECTION 8 CERTIFICATES
   AND VOUCHERS TO OFFSET THE
   IMPACT OF PRIVATE PROJECTS'
   OCCUPANCY RESTRICTIONS IS
   LIMITED
------------------------------------------------------------ Letter :4

Of the $278.9 million that has been made available for incremental
(new) Section 8 certificates and vouchers for persons with
disabilities since the passage of the 1992 act, $45 million ($25
million in fiscal year 1997 and $20 million in fiscal year 1998) was
specifically for offsetting the effects of occupancy restrictions by
private projects that meet the criteria of Sections 651 and 658.\16
Housing authorities can apply for certificates and vouchers funded
from this program if they can identify nonelderly persons with
disabilities in their communities who are not receiving housing
assistance because of private projects' occupancy restrictions. 

As of September 30, 1998, only about 19 percent of the $45 million
had been used to assist persons with disabilities affected by private
projects' occupancy restrictions.  HUD had awarded about 1,610
certificates and vouchers worth about $8.4 million to 13 housing
authorities for this purpose, leaving about $36.6 million
unobligated.  In addition, HUD had $41.7 million unobligated from the
funding set-aside targeted to local housing authorities with approved
allocation plans.  As a result, in September 1998, HUD transferred a
total of $78.3 million in unobligated funds from these two funding
set-asides into the mainstream program for Section 8 certificates and
vouchers, raising the amount of funding available in that program
from $48.5 million to $126.8 million.\17 All told, 276 housing
authorities have submitted applications for approval, requesting
22,000 certificates and vouchers worth almost $633 million. 
According to HUD, by the end of September 1998, it had obligated all
$126.8 million of the mainstream Section 8 certificates and vouchers. 

We were not able to account for housing authorities' greater interest
in applying for mainstream certificates and vouchers for persons with
disabilities than in applying for certificates and vouchers to assist
nonelderly persons with disabilities affected by private projects'
occupancy restrictions.  According to HUD officials, it is possible
that the mainstream program is more popular in part because housing
authority officials are more familiar with its application process. 
Furthermore, housing authority officials may find the process for
applying for mainstream certificates and vouchers simpler than the
process for applying for certificates and vouchers set aside for
housing authorities that have approved allocation plans or that can
identify persons affected by occupancy restrictions in private
housing.  Because of legislative requirements, the mainstream program
has fewer application requirements than the other two set-asides
have.  For instance, a housing authority desiring certificates and
vouchers in connection with an allocation plan must first submit the
plan to designate housing.  Information for this plan could require
participation from current tenants, the local government, and
advocacy groups. 

Similarly, a housing authority that applies for certificates and
vouchers to assist persons with disabilities affected by private
projects' occupancy restrictions have additional requirements.  The
housing authority must first identify those private projects that
have restricted occupancy and then submit statements signed by the
owners of the private projects certifying that the projects have
established occupancy policies favoring the elderly and providing the
number of nonelderly disabled families on their waiting lists.  In
addition, the housing authority must identify the number of other
nonelderly disabled families residing in the community who would
qualify for these certificates and vouchers.  According to HUD
Section 8 officials, even though HUD has encouraged the owners of
private projects to cooperate with the housing authorities to
determine the need for certificates and vouchers, many housing
authorities have found that project owners are often reluctant to
cooperate.  Only 1.8 percent of the managers of elderly housing
projects reported in our survey that they are working with a housing
authority to help persons with disabilities find housing. 

Communication problems among HUD's offices and among HUD, housing
authorities, and private projects may have contributed to the low
level of interest in the Section 8 certificates and vouchers
set-aside.  According to HUD's former Section 8 director, he had
asked the Office of Housing's Multifamily Housing Division to (1)
notify project owners that Section 8 certificates and vouchers were
available for qualified persons with disabilities who are affected by
the owners' changes in occupancy policies and (2) request that the
project owners notify their local housing authority to assist them in
identifying eligible candidates for certificates and vouchers. 
However, because of changes in HUD's management and organization, the
division has not acted on his request.  Likewise, the Section 8
office has not formally provided housing authorities with information
about Sections 651 and 658, including the allowable occupancy
policies and how these policies affect low-income nonelderly persons
with disabilities looking for affordable housing.  Moreover, other
than issuing an April 30, 1998, notice of funding availability for
the fiscal year 1998 funds, HUD has not made any formal efforts to
educate housing authorities on the type of funding available for
nonelderly persons with disabilities. 

Recognizing that housing authorities do not always apply for some of
these funds, the Subcommittee on VA, HUD, and Independent Agencies,
House Committee on Appropriations, in HUD's fiscal year 1999
appropriations, provided the Secretary of Housing with waiver
authority to allow nonprofit organizations to apply for these
certificates and vouchers. 

--------------------
\16 Our report entitled Public Housing:  Impact of Designated Public
Housing on Persons With Disabilities (GAO/RCED-98-160, June 9, 1998)
discusses in detail the use and limitations of Section 8 certificates
and vouchers for persons with disabilities.  This report focuses only
on Section 8 certificates and vouchers awarded to offset the effects
of private projects' restrictions of occupancy to the elderly. 

\17 While $48.5 million of this amount can be used to assist any
family or individual with disabilities regardless of age, the $78.3
million must be used for nonelderly persons with disabilities. 

   CONCLUSIONS
------------------------------------------------------------ Letter :5

While our survey data might indicate that, to date, few nonelderly
persons with disabilities have been affected by the occupancy
restrictions enacted by privately owned elderly housing projects, the
impact could be greater as more elderly projects enact occupancy
restrictions in accordance with Sections 651 and 658 of the 1992 act. 
Section 8 certificates and vouchers are also available to assist
nonelderly persons with disabilities who may no longer be eligible
for elderly projects, but much of these Section 8 funds remained
unspent because, among other reasons, housing authorities may find
the application process cumbersome in comparison with that for the
mainstream program.  Regardless, as demonstrated by the
oversubscription of the mainstream program, which has an application
process that is more familiar to housing authorities, there is a
demonstrated demand for certificates and vouchers for persons with
disabilities. 

   AGENCY COMMENTS
------------------------------------------------------------ Letter :6

We provided a draft of this report to HUD for its review and comment. 
HUD agreed with the report and responded that it found the
information in the report to be useful for continuing to implement
its programs for elderly persons and persons with disabilities.  The
Department also provided some technical clarifications, which we
incorporated where appropriate.  HUD's letter appears in appendix IV. 

---------------------------------------------------------- Letter :6.1

We conducted our review from December 1997 through October 1998 in
accordance with generally accepted government auditing standards. 

We are sending copies of this report to the appropriate Senate and
House committees; the Secretary of Housing and Urban Development; and
the Director of the Office of Management and Budget.  We will also
make copies available on request. 

Please call me at (202) 512-7631 if you and your staff have any
questions about this report.  Major contributors to this report are
listed in
appendix V. 

Judy A.  England-Joseph
Director, Housing and Community
 Development Issues

SURVEY METHODOLOGY
=========================================================== Appendix I

To obtain information on the impact of Sections 651 and 658 of Title
VI of the Housing and Community Development Act of 1992, we surveyed
a stratified random sample of 1,223 privately owned, federally
assisted housing projects that were potentially eligible to restrict
the occupancy of nonelderly persons with disabilities.  We identified
potentially eligible projects using a list of projects that receive
federal assistance provided by the Department of Housing and Urban
Development (HUD).  These projects receive assistance through the
following programs: 

  -- Programs eligible under Section 651:  Section 8 New Construction
     Program, Section 8 Substantial Rehabilitation Program, State
     Housing Agencies program (insofar as it involves new
     construction and substantial rehabilitation), the New
     Construction Set-Aside for Section 515 Rural Rental Housing
     Projects Program, and the Section 8 Housing Assistance Program
     for the Disposition of HUD-Owned Projects (insofar as it
     involves substantial rehabilitation).  Projects receiving
     assistance through these programs may also be known by the
     section of the National Housing Act, as amended, authorizing the
     assistance:  Section 221(d)(3) (excluding below-market interest
     rate), Section 221(d)(4), or Section 231. 

  -- Programs eligible under Section 658:  Section 202 of the Housing
     Act of 1959 as it existed before the enactment of the
     Cranston-Gonzalez National Affordable Housing Act; Section
     221(d)(3) of the National Housing Act, using a below-market
     interest rate determined by Section 221(d)(5); and the Section
     236 Interest Reduction Subsidy or other programs providing
     assistance under Section 236 of the National Housing Act. 

The data HUD used to develop the universe of projects affected by the
1992 act came from the Multifamily Data Warehouse, Field Office
Multifamily National System, and Tenant Rental Assistance
Certification System.  None of these systems has undergone detailed
data reliability assessments.  We spoke with HUD officials familiar
with the data to assess the accuracy of the data fields that were
critical to the creation of the project universe and to our use of
the data for the survey mailing.  We determined that the data were
sufficiently reliable for our purposes. 

From the lists we obtained from HUD, we eliminated family projects,
which HUD defines as being not elderly or handicapped.  We thus
obtained a list of 1,938 projects that were potentially eligible to
restrict the occupancy of nonelderly persons with disabilities under
Section 651 and 5,758 projects that were potentially eligible to
restrict the occupancy of such persons under Section 658, for a total
of 7,696 potentially eligible projects.  Those projects that are
actually eligible are the ones that were originally built for the
elderly.  We relied on a survey question to identify which of the
potentially eligible projects were actually eligible. 

We randomly sampled 646, or a sampling fraction of one-third, of the
Section 651 projects and 577, or a sampling fraction of about
one-tenth, of the Section 658 projects, for a total of 1,223
projects.  We mailed a questionnaire addressed to the project manager
at the addresses provided by HUD for these projects.  We sent
follow-up mailings to encourage responses and called some respondents
by phone to clarify unclear responses. 

Managers at 523 of the Section 651 projects (81 percent) and 452 of
the Section 658 projects (78 percent) responded to our survey.  In
our analyses of survey data, we weighted each survey in proportion to
its sampling fraction.  The results of our survey are presented as
estimates for the 6,080 ( 190) potentially eligible projects that we
estimate would have responded to our survey if we had mailed the
survey to all projects. 

   SAMPLING ERROR
--------------------------------------------------------- Appendix I:1

Because we used a sample (called a probability sample) to develop our
estimates, each estimate has a measurable precision, or sampling
error, that may be expressed as a plus/minus figure.  A sampling
error indicates how closely we can reproduce from a sample the
results that we would obtain if we were to take a complete count of
the universe using the same measurement methods.  By adding the
sampling error to and subtracting it from the estimate, we can
develop upper and lower bounds for each estimate.  This range is
called a confidence interval.  Sampling errors and confidence
intervals are stated at a certain confidence level--in this case, 95
percent.  For example, a confidence interval at the 95-percent
confidence level means that in 95 out of 100 instances, the sampling
procedure we used would produce a confidence interval containing the
universe value we are estimating. 

   ANALYSIS OF SURVEY NONRESPONSES
--------------------------------------------------------- Appendix I:2

About 6.5 percent of the addresses for surveyed projects provided by
HUD were identified by the U.S.  Postal Service as inadequate for
mail purposes.  We were able to obtain corrected addresses for some
projects and mailed surveys to the managing agent or owner of the
project listed in HUD's database for others.  We were able to obtain
an additional 31 completed responses in this manner.  This effort
reduced the nonresponse rate for undeliverable surveys to 4 percent
of the total universe, a rate we judged to be satisfactory. 

To test for potential differences between the projects that did and
did not respond to our survey, we performed two analyses.  The first
analysis was a statistical comparison of respondents to our mail
survey with those who did not respond on background information
provided by HUD for all projects.  This information included the type
of funding received by the project, the primary type of occupancy in
the project, and the total number of units in the project.  The only
statistically significant difference between the mail respondents and
nonrespondents for Section 651 projects was the primary type of
occupancy.  The projects of nonrespondents were significantly more
often devoted only partially to the elderly than those of respondents
and less often devoted wholly to the elderly.  For Section 658
projects, the projects of nonrespondents had significantly fewer
units than those of mail respondents (48.2  8.9 versus 64.2  5.7). 
In addition, the projects of significantly more Section 658
nonrespondents than respondents were funded under the Section 202
program for the handicapped while more of the projects of respondents
than nonrespondents were funded under the pre-1974 Section 202
program for the elderly. 

The second analysis was a comparison of those who responded to our
mail survey and those who did not respond to items from our
questionnaire.  To acquire information on these questions from
nonrespondents, we conducted a follow-up telephone survey of a random
sample of those who did not respond to our mail survey--91 of the 123
Section 651 nonrespondents as of May 4, 1998, and 94 of the 126
Section 658 nonrespondents as of June 11, 1998. 

We obtained some information about all 91, or 100 percent, of the
projects in our telephone survey of Section 651 projects and 86, or
91 percent, of the projects in our telephone survey of Section 658
projects.  In most cases, we spoke with either project managers or
representatives of the projects or their managing agents.  From
discussions with project representatives and HUD, we determined that
3 of the 91 Section 651 projects did not belong in our universe
because 2 are no longer receiving HUD subsidies and 1 no longer
exists.  We identified one Section 658 project that no longer
receives HUD subsidies. 

We tested for statistically significant differences between the mail
respondents and the telephone respondents on four survey questions: 
(1) the number of units; (2) the primary purpose for which the
project was built; (3) the project's current policy on occupancy by
persons with disabilities; and (4) the project's willingness to
accept persons with disabilities who require physically accessible
units, persons with disabilities who do not require physically
accessible units, and/or near-elderly persons with no disabilities. 
There were no differences between the mail respondents and telephone
respondents whose projects were built primarily for the elderly in
the number of units in the projects.  There was a significant
difference between telephone and mail respondents for Section 651
projects in whether projects were built primarily for the elderly. 
That is, Section 651 projects surveyed by telephone were
significantly less often built primarily for the elderly than were
the Section 651 projects responding by mail.  We also found
statistically significant differences in the projects' occupancy
policies and the types of disabled persons the projects accepted for
occupancy.  Officials of those Section 651 projects surveyed by
telephone reported significantly more often than mail respondents
that they accept both elderly and disabled persons for occupancy and
reported less often than the mail respondents that they reserve units
for nonelderly disabled persons, limiting the remaining units to the
elderly.  There were no significant differences between Section 658
telephone respondents and mail respondents on the policy of accepting
elderly and disabled persons in all units.  However, Section 658
telephone respondents reported significantly less often than mail
respondents that they reserve units for nonelderly disabled persons. 
Section 651 telephone respondents were also significantly more likely
to report accepting both disabled persons who require physically
accessible units and those who do not require such units, while there
was no difference in these policies for Section 658 telephone and
mail respondents. 

   RESULTS OF DATA RELIABILITY
   CHECKS
--------------------------------------------------------- Appendix I:3

To assess the reliability of responses to our survey and HUD's data,
we compared the two on the number of units in each project and the
section of the act under which the project received assistance.  For
the number of units, the two sets of numbers were highly correlated
(Pearson correlation coefficient of .98 for Section 651 projects and
.90 for Section 658 projects).  However, the survey responses
differed significantly from HUD's data on the section of the act
under which the project received assistance.  As many as 26 percent
of the Section 651 projects that identified themselves as receiving
Section 8 subsidies reported differently from HUD on whether they
received new construction or substantial rehabilitation funding, and
23 percent of the respondents who reported the section of the act
under which they received funding provided an answer different from
HUD's records.  Only 6 percent of the Section 658 projects assigning
themselves to classifications under a section of the act used
classifications different from those HUD designated. 

To test the completeness of the universes covered by Sections 651 and
658 of the 1992 act provided by HUD headquarters, we compared the
listings provided by HUD headquarters with similar listings of
projects provided to us by the HUD field offices at five locations
across the country.\1
Because the HUD headquarters lists were not generated at the same
time as the field office lists, we recognized that some minor
differences would exist.  In most cases, the listings provided to us
by HUD agreed with the listings provided by the field staff.  In the
two locations that had differences, these differences were minor.\2
Thus, we concluded that the information provided by HUD headquarters
was sufficiently complete for our survey. 

(See figure in printed edition.)Appendix II

--------------------
\1 These five locations were Anaheim and San Francisco, California;
Fall River, Massachusetts; Corinth, Mississippi; and Woodbury, New
Jersey. 

\2 In one location, the HUD headquarters list omitted one project. 
In another location, the headquarters list had one additional project
and the local list had one additional project.  The total number of
projects in these locations from the headquarters list was 47. 

SURVEY RESULTS OF PRIVATELY OWNED
HOUSING PROJECTS
=========================================================== Appendix I

(See figure in printed edition.)

(See figure in printed edition.)

(See figure in printed edition.)

SAMPLING ERRORS FOR QUESTIONS IN
THE SURVEY OF PRIVATELY OWNED
HOUSING PROJECTS
========================================================= Appendix III

Survey question               Estimate based on response     Sampling error for response
----------------------------  -----------------------------  ----------------------------
1a                            439,170                        27,930

1b                            6,055                          21.9

3a                            3.4                            1.7

3b                            92.9                           2.2

4a                            407,399                        25,582

4b                            6,047                          33

5a                            4,157                          197

5b                            1,780                          194

5c                            92                             46

6a                            287,629                        24,151

6b                            1,034                          544

6c                            3,680                          1,238

6d                            25,796                         4,169

6e                            11,838                         1,863

6f                            9,608                          1,292

6g                            2,046                          268

6h                            321,152                        25,979

7a                            3.5                            1.5

7b                            7.2                            2.1

7c                            76.9                           3.3

7d                            8.3                            2.0

7e                            3.1                            1.4

8a                            52.9                           8.2

8b                            22.0                           6.9

8c                            5.4                            3.3

8d                            11.0                           4.9

8e                            7.0                            4.2

9a                            4.2                            1.6

9b                            17.2                           2.8

9c                            70.1                           3.6

9d                            5.0                            1.7

9e                            1.7                            1.1

10a                           61.4                           7.0

10b                           15.9                           5.3

10c                           6.9                            3.2

10d                           7.0                            3.5

10e                           5.8                            3.6

11a                           291,468                        109,188

11b                           4,070                          199

12a                           240,344                        95,974

12b                           3,761                          1,757

12c                           22,149                         10,402

12d                           267,485                        105,558

13a                           89.7                           2.5

13b                           9.2                            2.4

14a                           19.7                           11.1

14b                           16.8                           9.7

14c                           10.8                           7.5

14d                           17.9                           10.4

14e                           28.9                           12.3

15a                           41.6                           4.0

15b                           54.9                           4.0

16a                           23.8                           3.5

16b                           0.7                            0.4

16c                           1.1                            0.7

16d                           *

16e                           3.0                            1.2

16f                           21.2                           3.5

16g                           *

16h                           41.8                           3.9

16i                           6.3                            1.9

17a                           16.4                           2.9

17b                           73.9                           3.5

18a                           41.9                           4.0

18b                           57.0                           4.0

19a                           78.6                           5.1

19b                           13.4                           4.2

19c                           8.0                            3.4

20a                           68.8                           3.8

20b                           42.8                           3.8

20c                           5.1                            1.4

20d                           21.0                           3.5

21a                           75                             47

21b                           3,960                          201

21c                           122                            56
-----------------------------------------------------------------------------------------

(See figure in printed edition.)Appendix IV
COMMENTS FROM THE DEPARTMENT OF
HOUSING AND URBAN DEVELOPMENT
========================================================= Appendix III

MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V

HOUSING AND COMMUNITY DEVELOPMENT
ISSUES

Stanley J.  Czerwinski, Associate Director
Mathew Scire, Assistant Director
Martha Chow
J.  Davis
Diana Gilman
Stewart Seman

DESIGN, METHODOLOGY, AND TECHNICAL
ASSISTANCE GROUP

Carolyn Boyce

OFFICE OF THE GENERAL COUNSEL

John T.  McGrail

*** End of document. ***