The research forming the basis of this report was conducted by the Division of Program Evaluation in the Office of Policy Development and Research, U.S. Department of Housing and Urban Development.

Acknowledgments

This report was made possible by the extraordinary commitment to the needs of America's homeless by hundreds of nonprofit organizations and general units of government. Furthermore, no evaluation findings would be possible without the full cooperation and candid sharing of experiences by the HUD-assisted providers of homeless shelters and services.

The conceptualization and oversight of this report, including preparation of the executive summary, were the responsibility of James Hoben ofthe HUD Office of Policy Development and Research. Mr. Hoben also supervised or performed four of the six homeless evaluations reported upon. The body of this report, including the synthesis of all six evaluations of HUD-administered homeless programs,was ably performed by Michelle Bussard, a consultant to HUD.

Foreword

The array of programs administered by HUD under the McKinney Act have made important contributions to the fight against homelessness, assisting homeless individuals and families and helping build and support a nationwide network of private and public shelter and service providers. Today, as we stand at a turning point in homeless assistance policy, I am pleased to offer this Review of Stewart B. McKinney Homeless Programs Administered by HUD, which draws lessons that can help further increase the effectiveness of HUD's efforts.

Collectively, these programs have targeted resources to many types of homeless assistance, including emergency shelter, transitional and permanent housing, and homeless prevention, as well as a wide range of supportive services. Some programs, such as Emergency Shelter Grants (ESG), have enabled many types of assistance providers to offer basic services across a broad spectrum of homeless populations; other programs have focused on providing limited types of assistance or on serving clienteles with special housing needs.

However, the sheer number and variety of HUD's McKinney Act programs have sometimes created barriers to their efficient use. Differences among the target populations, eligible activities, application requirements, and selection criteria for the various programs have made this assistance difficult to obtain and coordinate. Overlapping sets of regulations and reporting requirements have further complicated grant administration. Moreover, the unpredictability of competitive grants, appropriation levels, and the varying lengths of grant awards have frustrated attempts to implement longer term, comprehensive strategies for eliminating homelessness.

The challenge now facing HUD is clear -- to forge an approach that will empower local communities with the flexibility to develop comprehensive, integrated homeless assistance strategies that are responsive to their needs and resources. The experience of the McKinney Act programs offers crucial insights into how HUD can foster this continuum of care. For example, recent evaluations of HUD's McKinney Act programs have shown that block grant funding works well and that local grantees have been able to make excellent use of program flexibility, demonstrating judgment and creativity in developing comprehensive asssistance strategies.

These and other lessons from the first generation of McKinney Act programs have helped guide President Clinton and the Congress in forging a bipartisan approach to reforming Federal homeless assistance programs. By building on the successes of the past with programs that are more flexible, consistent, and responsive, HUD can help local communities provide the multifaceted, long-term assistance needed to finally break the cycle of homelessness.

Executive Summary

This report synthesizes the findings of six evaluations of homeless programs administered by the U.S. Department of Housing and Urban Development (HUD). The report responds to a congressional mandate that HUD "conduct a comprehensive review and evaluation of each program [it administers] under Title IV of the Stewart B. McKinney Homeless Assistance Act." The principal conclusion of the report is that the time is right to consolidate and simplify the McKinney programs.

BACKGROUND

The Stewart B. McKinney Homeless Act of 1987 (P.L. 100-77) was the Nation's first attempt at a comprehensive response to homelessness among individuals and families with children. The Act originally created 20 programs for the homeless that were administered by 9 Federal agencies. Appropriations, which in FY89 were $0.5 billion, have been increased each year. Appropriations totaled $1.35 billion in FY94. HUD is responsible for directing six major programs that together expend nearly 70 percent of all appropriations.

Five years after passage of the McKinney Act, Congress requested, in the Housing and Community Development Act of 1992, Section 1409, that the Secretary of HUD report in 1994 on the results of the McKinney programs administered by HUD. This report responds to that request by summarizing the findings contained in evaluations of the following six programs administered by HUD:

1. Emergency Shelter Grants Program (ESG).
2. Supportive Housing Demonstration Program (SHDP).
3. Section 8 Moderate Rehabilitation Assistance for Single-Room Occupancy Dwellings (SROs).
4. Shelter Plus Care (S+C).
5. Supplemental Assistance to Facilities to Assist the Homeless (SAFAH).
6. Single Family Property Disposition Initiative (SFPDI).

The evaluations were funded and directed by the HUD Office of Policy Development and Research (PD&R). They were initiated prior to the 1992 congressional request, in response to earlier requests for specific program evaluations by Congress in the cases of SRO and SHDP; the President's Office of Management and Budget (OMB) for ESG, S+C, and SAFAH; and the HUD assistant secretary responsible for each program.

EVALUATION FINDINGS

The McKinney Programs are an array of distinct programs that are challenging to use in a coordinated, comprehensive manner. In 1987, Congress created a set of programs for the homeless that addressed the needs of several different homeless populations and supported a variety of program responses, among them:

• Different client populations. The SAFAH and SHDP Transitional Housing programs supported all homeless persons, but gave special preference to families. The SRO program focused on the single homeless. The S+C and SHDP Permanent Housing programs served homeless persons with disabilities mainly those with severe mental illness, chronic substance abuse, and/or AIDS.
• Different applicants. States and local governments, Indian Tribes, Public Housing Authorities, and nonprofits could apply for SHDP Transitional Housing, and SAFAH. Only States could apply for SHDP Permanent Housing. Public housing authorities (PHAs) often with nonprofit or for- profit partners were permitted to apply for SROs. States and local governments, and Public Housing Authorities with nonprofit sponsors could apply for S+C. General units of government or nonprofits were eligible for surplus property for the homeless.
• Different eligible activities. The ESG, SHDP, and SAFAH programs supported acquisition and rehabilitation of shelters or housing, funded supportive services, and paid operating costs. In addition, ESG funds could be used for homeless prevention. S+C and SROs provided rental assistance only. Neither program funded supportive services, although S+C required a match of locally provided supportive services for each dollar of rental assistance provided and SRO encouraged the a provision of services. The SFPDI program provided properties at subsidized lease rates or discounted purchase prices, but no funds for operations or services.
• Types of homeless programs supported also varied. ESG provided emergency shelter, services and prevention; SHDP Transitional Housing was for transitional housing and services; and SHDP Permanent Housing, S+C, and SROs supported permanent housing for the homeless.
• Different funding systems and varying criteria. One program, ESG, was an entitlement and was awarded as a block grant using a formula. Four programs, SHDP, S+C, SRO, and SAFAH, were competitively awarded, each in accordance with its own selection criteria. The SFPDI relied on requests and negotiated agreements.
• Confusing timeframes and regulations. The duration of grant awards differed among programs from 1 or 2 years under SAFAH, to 5 years for SHDP, and to 10 years for SRO and elements of the S+C program. The timing of grant awards differed for each program from year to year. A major frustration for governments planning comprehensive homeless strategies was the unpredictability of competitive awards. Once as few as 3 percent of applicants (42 in 1,400) were funded because of large needs and limited appropriations.

Some of the complexities listed above were addressed by Congress in the FY92 consolidation of SAFAH and SHDP programs into the Supportive Housing Program (now a permanent, rather than a demonstration, program).

McKinney programs have had measurable impacts

In spite of challenges and complexities, McKinney grantees have achieved measurable results toward ending or preventing homelessness. Most emergency shelters have gone from shelter only to include supportive services. As points of first care, ESG grantees have helped approximately 7 percent of all assisted households find permanent housing and referred others to transitional housing.

Transitional housing sites supported under the SHDP and SAFAH programs reported that approximately 70 percent of the households that complete the transitional program go on to permanent housing within an average of 9 months. Half of permanent housing was unsubsidized and half was subsidized. Grantees using SHDP Permanent Housing funds for the disabled homeless reported that 85 percent of households were still in the SHDP residence after 1 year or were living in an equivalent permanent housing setting. SRO grantees reported positive results in providing permanent, affordable (average monthly rent of $300) housing for the single homeless, though there was significant turnover. SHDP Transitional Housing grantees reported a doubling of households employed or in training from 18 percent at entrance to 38 percent at exit, and a doubling of households earning $900 per month from 6 percent to 11 percent. Costs for providing the assistance were reasonable. The average cost of transitional housing, with support services, was $30 per person per day. The average cost for permanent housing including support services was $45 per person per day.

McKinney funds supported comprehensive programs and were well leveraged

Sponsors of homeless programs that used McKinney funds created comprehensive programs for their specific populations and provided a diverse array of needed support services. For example, SHDP Transitional Housing projects provided most of 28 important supportive services. Because far more money is needed than is provided under McKinney, project sponsors exceeded the maximum McKinney dollar to local funding matches of 1:1. On average, ESG dollars were matched 1:10, SHDP dollars were matched 1:3, and S+C grantees pledged to match rental assistance by in-kind supportive services at 1:2. Sixty percent of SRO development costs came from community sources. Matching resources were obtained from State and local governments, foundations, businesses, churches, and individual donations. The ESG, SHDP, and SAFAH McKinney funds, though often a modest portion of an organization's homeless budget, were critical because they could be used to leverage other funds or, as a last resort, could finance otherwise unavailable services or housing.

McKinney flexibility was used responsively and effectively

SAFAH, and to a lesser extent ESG and SHDP, demonstrated the importance of permitting project sponsors to use McKinney funds where most needed. SAFAH grantees took advantage of the flexibility of that program to fund gaps in homeless programs. In addition, one-third of the SAFAH projects were considered innovative.

ESG grantees initially used funding to expand and improve shelter, but once done, they shifted funding to supportive services and operations. When homeless prevention activities became eligible for funding, ESG grantees used the maximum allowable amounts for preventive programs. In contrast, grantees of the less flexible S+C and SRO programs complained that the limitation of those funds to rental assistance meant they sometimes could not obtain vitally needed supportive services, such as case management.

A homeless formula grant will work, and nonprofit homeless providers will continue to play a role in shelter and service provision

The ESG program is a formula grant program, and it has worked well. States and local governments have praised the predictability of funding, though they have been frustrated by the uncertainty of annual appropriations. They appreciated that they need not waste time writing competitive grant applications, and that they did not face the uncertainty of winning or losing. States and local governments were prompt in reallocating ESG funds to homeless providers.

Though State and local governments and PHAs are eligible grantees and could directly operate homeless facilities and services, most of them passed McKinney funding through to nonprofit organizations. Nonprofit organizations provided 80 percent of homeless housing and services. There were several reasons for reliance on nonprofits. First, they could more readily integrate housing and services because they were unencumbered by a government bureaucracy and could draw on a wide variety of other community resources. Second, local governments that relied on nonprofits could avoid increasing government staffing and the burdens that implied.

Successful programs focused on the individual's specific needs, required individual responsibility, and were comprehensive

The four characteristics that stood out in every successful homeless assistance program, regardless of the cause of homelessness, were:

1. treatment of each person as unique and valued;
2. an operating principle that respects individual rights and requires individual responsibility;
3. availability of stable housing and existence of a comprehensive set of assistance services; and
4. a continuing challenge to each individual to be as independent as possible.

McKinney administrative funds for grantees need to be increased

Virtually all project sponsors said that McKinney expenditures permitted for administrative activities were inadequate. Sponsors said that considerable uncompensated time and cost were expended creating and maintaining comprehensive housing and services because most funding and services came from multiple local sources. Coordination costs were considerable, and costs for outreach and intake of the homeless were substantial. Recordkeeping on household progress was time consuming, and many reported that preparing the required reports for multiple funders, including HUD, was a major undertaking.

Because of tight administrative budgets, most sponsors admitted that, although they should have, they did not track homeless households once the households left a program. S+C represented an extreme case of inadequate administrative funding. Because S+C grantees had to obtain all supportive services from local providers, they were forced to spend substantial amounts of time locating, coordinating, and overseeing the services and yet were not allowed to use S+C funds for that critical function.

HUD staff need to provide more technical assistance and monitor programs more closely

Grantees praised HUD staff efforts to help them with applications and project implementation, but said they still needed more help. HUD staff were limited in both the amount of technical assistance they could provide and in the amount of time available to monitor projects. All grantees were eager to learn what their peers were doing and to hear about alternative program approaches.

As Congress and HUD move toward a consolidated and more flexible McKinney program, there will be an even greater need for both HUD-provided and other technical assistance to ensure effective program administration and results.

CONCLUSION

The programs evaluated in this report demonstrate that immense capacity exists at the local level for going beyond present efforts, if a better Federal system of assistance is put in place. The McKinney-supported projects that have demonstrated the most positive client outcomes have succeeded by building comprehensive community care systems -- a continuum of care -- that includes outreach systems, prevention programs, emergency shelter with strong service components, transitional supportive housing, and permanent housing placement.

The time has come to advance the opportunity for similar successes in all parts of the Nation. A consolidated approach to homeless assistance that improves coordination and eliminates fragmentation simply makes sense. That approach will place a high priority on helping people to "help themselves in a relationship of mutual rights and responsibilities." (Priority Home! The Federal Plan to Break the Cycle of Homelessness, U.S. Department of Housing and Urban Development, March 1994, p. 51.)