Skip to main content

Stewart B. McKinney Homeless Programs


Posted Date: December 12, 1995

Publication Icon

Supportive Housing Demonstration Program

The evaluation of SHDP performed by Westat, Inc., Washington, Maryland, examined projects funded from 1987 through 1990. Data were gathered through a 1992 survey of 732 active projects, to which 623 projects (85 percent) responded; from case studies of 45 projects; and from participants in several focus groups that met in 1993. Publication is pending.

SHDP was created to develop and support innovative approaches to combining housing and supportive services for homeless individuals and families, with an emphasis on those with special needs. The program was authorized by the Stewart B. McKinney Homeless Assistance Act of 1987 (P.L. 100-7). The program was renamed the Supportive Housing Program, and became permanent upon enactment of the Housing and Community Development Act of 1992.

SHDP consisted of two separate programs, the Transitional Housing Program and the Permanent Housing for Handicapped Homeless Program. Each was designed to serve distinct homeless populations, and each had different objectives.

Transitional Housing aimed to help homeless individuals and families make the transition from homelessness to more independent living, and its residents were limited to a maximum stay of 24 months. By contrast, the PH program sought to provide a long-term stable housing option and supportive services to homeless individuals with disabilities or to homeless families that included a disabled adult member. It imposed no time limit on residency. Funds for both programs were awarded through a competitive grant program.

During the evaluation period, 1987 through 1990, the SHDP program provided $372 million for both programs, with $340 million going to the TH component for 496 projects, while $32 million paid for 236 projects under the PH component. These programs are described individually in the following sections.


Funding was available to State and local governments, Indian tribes, private nonprofit organizations, and community health organizations. The majority of grantees were nonprofit sponsors (84 percent) whose primary mission was to provide housing for homeless persons (71 percent). More than half of the grantees had been providing such service for more than 5 years. The 1987 and 1990 grantees were responsible for 496 active TH projects by 1992, containing 8,600 housing units.

The SHDP TH program's transitional housing options varied. Dwelling types were somewhat associated with the type of homeless persons being served. In many instances, projects serving substance abusers relied on SRO units; the use of apartments was predominant among projects serving primarily families and victims of domestic violence. Sponsors that served a variety of client types often employed a mix of building types within the same TH project. In some instances these included nonresidential buildings converted for residential use. Of all transitional housing projects reviewed, 38 percent were apartment buildings with four or more units, 24 percent were SRO dwellings, 19 percent used mobile homes or motels, 17 percent were single-family houses, 15 percent were clusters of two- and three-unit townhouses, and 7 percent were in rowhouse or townhouse configurations. The majority of project sponsors (65 percent) owned their project sites; the balance entered into lease arrangements. The average length of stay in a TH project was 9 months, with a maximum of 24 months allowed.


During the period studied, HUD awarded 535 TH grants totaling $340 million. The average grant was approximately $636,000. During the 1987 to 1990 operating year, SHDP furnished 15 percent of rehabilitation costs for TH projects and 28 percent of total operations and services costs.

Additional sources of funding typically came from the same three funding sources, which in descending order of magnitude, were State funds, local funds, and sponsor contributions. TH project sponsors also relied on an array of other funding sources, including Community Development Block Grants, 4 to 5 percent; bank loans, 3 to 15 percent, depending on the activity receiving the loan; and higher figures, generally supporting acquisition. Individual contributions amounted to 3 to 5 percent, and foundations contributed 4 to 5 percent.

TH project sponsors were able to leverage significant amounts of local resources:

  • The $37 million in SHDP grants in 1987 to 1990 that TH sponsors used for facility acquisition or rehabilitation leveraged about $202 million.
  • The $35 million in SHDP funds in 1991 to 1992 that TH sponsors used for services and other operating costs leveraged about $89 million in funds from other sources.

The ability to create leveraging ratios of nearly 6:1 and 3:1 satisfied a major SHDP program objective: leveraging substantial local resources.

Sponsors were also able to provide housing and deliver services at a reasonable cost. Estimates that include the cost of supportive services, operations, and the amortized cost of acquisition and rehabilitation result in a cost per person of $30/day and per household of $53/day. Because the average stay in a TH project was 9 months, the average total cost for a resident served by the program was $8,100 and, for a household, $14,310.

Most TH projects were successfully implemented in a timely manner. By fall of 1992, 94 percent of all projects supported by TH grants were operational. Figures indicate how effectively the vast majority of TH projects achieved important milestones. From the signing of their agreement with HUD, sponsors took an average of 4 to 5 months to acquire a suitable property, 9 to 11 months to complete a project, 8 months to achieve initial occupancy, and 10 months to achieve full occupancy.

To facilitate the transition into independent living, project sponsors generally provided a full range of supportive services. Services provided by TH sponsors were comprehensive, flexible, and tailored to meet the special needs of differing homeless populations -- a key program objective.

Case management was offered by virtually every TH project (98 percent). The basic elements of case management include needs assessment upon entry, periodic reassessment and progress monitoring, group meetings, and resident enrollment in community-based service programs.

More than 90 percent of 1987 to 1990 TH projects also provided housing location services, money and household management, prevocational training, and vocational counseling. Services offered by fewer than half the projects were prenatal care (48 percent), medication monitoring (48 percent), detoxification (34 percent), English as a second language (30 percent), physical therapy (28 percent), shelter workshops (25 percent), and Parents Anonymous (22 percent). In most projects (93 percent), the cluster of household management services were offered onsite. The most frequently unmet need, according to project sponsors, was for employment-related services.


By fall of 1992, the 1987 to 1990 TH projects were annually serving an estimated 6,960 family units and 12,672 individuals. The most commonly served type of household was that of unaccompanied adults without children (54 percent). Women accounted for the majority (60 percent) of adults served by the program, and children under 18 years of age accounted for the largest single age group (41 percent). Although most households in TH projects were composed of single adults with no children, families with children (43 percent) made up a higher percentage of TH projects than are proportionately represented in the national homeless population (10 percent). African Americans constituted the largest ethnic group (47 percent).

Half of the individuals in TH households had come directly from emergency shelters or from the streets. The rest arrived from relatives' homes, drug or psychiatric treatment facilities, hospitals, or even from their own residences. Unlike many programs that require extensive screening, many TH sponsors accepted clients immediately after an eviction or other major residential disruption.

The TH program achieved its goal of helping residents achieve self-sufficiency and find independent living situations. This conclusion was based on analysis of three measures of success: (1) increased residential stability, (2) improved employment status, and (3) increased income. Substantial numbers of residents who were TH program clients eventually entered stable housing (56 percent) and gained employment (38 percent).

Rates of success in achieving residential stability were higher for those who completed a prescribed transitional program; 70 percent graduated to stable housing arrangements. Among the populations served by TH projects, this trend held with the exception of projects serving battered women, among whom improved employment status remained unchanged even though 61 percent who voluntarily left found stable housing. Families with children were the most successful in securing permanent housing (90 percent of those that completed their transitional program), followed by the severely mentally ill (74 percent), and substance abusers (67 percent). Generally, those who achieved success in independent living were residents who had completed TH programs.


In contrast with the TH program, PH grants were made only to States, and they were required to use nonprofit organizations or public housing authorities as project sponsors. From 1987 to 1990, 248 PH grants were awarded. Only 16 percent of all PH sponsors were public housing authorities. The majority (84 percent) were nonprofit organizations, most (67 percent) with more than 5 years' experience in serving and sheltering homeless persons. Many (48 percent) of these community-based sponsors described their primary mission as providing housing for the homeless, while a sizeable number (43 percent) indicated that their primary mission was providing mental health services. Nearly 90 percent of all project sponsors owned their own buildings and land.

The majority of PH residents (54 percent) came directly from either a transitional housing facility, a relative's home, or a psychiatric facility. Unlike the TH program, PH projects imposed no time limits on residency for their clients. Innovative approaches were encouraged in provision of permanent housing and supportive services for the long-term needs of disabled homeless persons.


From 1987 to 1990, HUD invested $32 million in the SHDP PH program, accounting for approximately 38 percent of all funding for PH projects. During that period, PH projects required $29 million for acquisition and $28 million for rehabilitation, a total of $57 million. In addition to PH grants, sources of funding included State and local funds (34 percent), bank loans (10.7 percent), sponsor contributions (5.3 percent), and Community Development Block Grants (3.0), among others.

Of the total SHDP funding available, $11 million was used to acquire property and facilities, and $6 million went for rehabilitation and program expansion. In a mail survey of 181 PH projects that were asked for figures on their most recently completed fiscal year, 1-year operating and services costs totaled more than $22 million, with SHDP grants providing 18 percent, or $4 million, in total support. The PH program was able to achieve good leveraging ratios of 3:1 for acquisition funding and 4:1 for housing rehabilitation and expansion.

By 1992, 94 percent of the 1987 to 1990 PH projects were providing housing and services to homeless clients. Most (58 percent) PH projects were in single-family home configurations of one, two, or three bedrooms but other dwelling types were also common. A typical project offered 9 to 10 permanent housing units. Fully 30 percent of all permanent housing units were SRO facilities. The majority (89 percent) of PH project sponsors owned their sites. By the fall of 1992, the 1987 to 1990 projects operated 1,616 dwelling units.

The average cost for acquisition of a PH project was $139,000. Average per-unit acquisition cost was $76,000 for single-family units, $22,000 for apartment units, and $20,000 for SROs. Most properties (95 percent) required some rehabilitation or physical expansion. Costs averaged $123,000 per project but varied by type of dwelling unit: $84,000 for single-family-only projects, $25,000 for apartment-only projects, and $21,000 for SRO-only projects.

The average daily cost to provide permanent housing and services to a disabled homeless client through the PH program was $45 per person and $51 per household, and the average yearly cost was $16,537 per individual and $18,475 per household. These figures compare favorably with an analysis completed in the Seattle metropolitan area, where it was found that the average cost was $292 per day for a skilled nursing residential treatment program for an individual with AIDS, or $1,000 per day for hospital treatment.

PH project sponsors provided an array of services to meet the special needs of their disabled homeless residents. In 90 percent of all projects, supportive services that included household management, money management, and medication monitoring were provided onsite. More than 60 percent of mental health services were provided onsite whereas most physical health, alcohol, and substance abuse treatment services were provided offsite by outside organizations. Services needed, but sometimes unavailable, included education, employment training, mental health services, followup support, and legal assistance.


The PH program was expressly designed to provide independent living arrangements for homeless individuals who were mentally or physically disabled. The program annually served an estimated 1,515 persons in 1,356 households. The largest single group served by PH programs were white, non-Hispanics (73 percent) who were single adult males (60 percent) between 31 to 50 years of age (51 percent). Eighty-three percent of all households were unaccompanied adults without children. The largest group of PH projects served persons with severe mental illness (56 percent) or developmental disabilities (31 percent). In most PH arrangements, however, residents were characterized by multiple disabilities, deficits, and disorders that might include alcohol and substance abuse, AIDS or HIV infection, or victimization of domestic violence.

An important PH program goal was to provide long- term stable housing and supportive services to residents. The majority (69 percent) of PH residents remained in their housing arrangements for at least 1 year. In addition, nearly one-half (48 percent) of the residents who left a PH project were able to secure stable housing arrangements. There were variations among PH projects serving different primary populations. One year later for example, among the severely mentally ill (SMI), 68 percent of residents remained in their PH housing units. Of residents who were developmentally disabled, approximately 78 percent remained in their units.

The employment and income patterns of PH project residents did not seem to relate to achievement of stable housing. Of residents who had been in a PH project for at least 1 year, employment increased by only 5 percent, from 24 percent at entry to 29 percent.

PH projects did not provide significant followup for residents that either left voluntarily or were dismissed. The followup services that were provided were usually through phone calls and continued meetings with case managers. In most cases, however, project sponsors reported that they lacked the funds to provide long-term, meaningful followup.

The success of PH projects was commonly attributed to the case management that provided effective shelter and service linkages, along with interagency collaboration and the availability of supportive services. PH sponsors identified a need to involve and educate the community, to build linkages, and to garner relationships with community institutions and support among local businesses and civic groups.

Table of Contents 

Publication Categories: Homelessness     Publications    


All Publications
Search for Publications
Search for Ongoing Research