The Homeless Families Program

The Homeless Families Program
A Summary of Key Findings By Debra J. Rog and Marjorie Gutman

The Foundation has made two investments in large national programs directed at alleviating problems facing homeless people in America. The first, Health Care for the Homeless, attempted to increase the availability of health care services for homeless people. It became a model that was cited when the federal government passed the McKinney Act in 1987, providing federal dollars to improve access to health care for homeless people throughout the country. After the Health Care for the Homeless program was completed, the Foundation funded a second program, this time focusing on homeless families. The Homeless Families Program was more ambitious than the first. It attempted to improve not only health care services for homeless families but also a range of other social services generally important to their well-being. The Foundation entered an active and productive partnership with the federal Department of Housing and Urban Development, which made stable housing arrangements available to the families participating in the program. The premise of the program was that both housing and social services (including health care) were needed to get many homeless families back into stable and independent life circumstances.		The Homeless Families Program exemplifies a range of national programs begun by the Foundation in the late 1980s and the start of the 1990s, which emphasized systems reform as a long-range solution to making public investments in social services more productive. The theory held that the problem with social services was not just that more were necessary but that existing resources needed to be better coordinated and better focused. Chapter Ten was written by Debra J. Rog and Marjorie Gutman. They present findings from the formal evaluation of the program that the Foundation funded soon after the program was initiated. This chapter offers insights into the problems faced by homeless families as well as the obstacles faced by program managers trying to bring about system reform. The discussion also addresses the challenges involved in designing and implementing "enriched services" accompanying housing for the homeless. Rog, who is a research fellow at the Center for Mental Health Policy, Institute for Public Policy Studies at Vanderbilt University, has published extensively on the problems of homelessness in America. Gutman, a senior program officer of the Foundation who was in charge of the design and monitoring of this evaluation, has been active in developing and evaluating a number of Foundation programs addressing the needs of vulnerable populations.

Americans have a short attention span. A newly discovered problem receives major national attention from the public and policy makers for a few years, only to be replaced by another pressing problem. Most disturbing, in the case of social issues, the first problem is rarely resolved and the "new" one may even be another manifestation of the same issue.

This is certainly true of homelessness. During the 1980s, homelessness took center stage as a largely unexpected new problem for our society. Homeless people have been found in most times and places, of course, but the increasing appearance of homeless women and children, and even whole families, on the streets and in shelters made the issue highly visible and compelling.¹Best estimates were that women and children totaled one-fifth to one-third of the homeless population.² One heated debate at the time concerned the extent to which these families were homeless because of temporary economic dislocation or because of endemic poverty and other complicating factors.

It was against this backdrop that the Homeless Families Program, a five-year effort, was initiated in 1990. Even as the HFP was starting, national attention on homelessness was already beginning to wane and has remained low ever since. It is true that a small cadre of activists, providers, policy makers, and dedicated volunteers have continued to grapple with the problem, and it does surface now and again in public debate. But by and large the public's attention is captured by current concerns--welfare reform, the "underclass," violence--and it is easy to forget that homeless persons, and especially homeless families, are a small but very important part of the "new" problems the nation is trying to address.

PROGRAM OVERVIEW

The Homeless Families Program, a joint effort of The Robert Wood Johnson Foundation and the Department of Housing and Urban Development, was the first large-scale response to the problem of family homelessness. Started in nine cities across the nation, it had two complementary goals:

The ultimate goal of the Homeless Families Program was to improve the residential stability of families, promote greater use of services, and increase steps toward self-sufficiency. In addition, as a demonstration program the HFP integrated a major evaluation into the initiative at all sites. The evaluation was designed to learn more about the needs of families who struggle with homelessness and other problems, to learn how services and systems might be better organized and delivered to meet those needs, and to examine how housing might be delivered to promote stability and use of services as well as progress toward self-sufficiency.

The Homeless Families Program was an outgrowth of several previous demonstration programs. One main progenitor was the nineteen-city Health Care for the Homeless Program, cofunded by The Robert Wood Johnson Foundation and the Pew Charitable Trusts in 1985.³Under this program, thousands of homeless people received health services, assessments, and referrals through primary care clinics located in shelters. The simple premise of the program was to make health care accessible to homeless people by locating it where they congregate and by tailoring the care to their special needs. The program accomplished its goal of demonstrating the feasibility and the acceptability of health clinics for homeless persons, and it became the template for the hundreds of clinics supported in many cities under the 1987 Stewart B. McKinney Act--the nation's landmark legislation in homelessness.

Additionally, evaluation of the Health Care for the Homeless Program led to the first large multicity dataset on the characteristics of homeless people and their health care needs. This study, along with others conducted at the time, helped establish the fact that young families--consisting mostly of single women with two to three children--made up a significant segment, and the fastest-growing one, of the homeless population.⁴These studies also documented that members of homeless families were experiencing significant health problems, depression, and developmental delays. For example, roughly 33 percent of homeless mothers in the study suffered from psychiatric problems, and roughly 20 percent abused alcohol or illegal drugs. The children in these situations had very low rates of immunization and suffered from extraordinarily high rates of childhood illness. Thirty-five percent of them had recurrent ear infections. The incidence of chronic disorders ran approximately twice the norm. Finally, data from the Health Care for the Homeless Program supported the contention of many researchers that a significant number of these children were at risk for long-term, if not permanent, developmental delay.

Growing recognition and evidence of the more complex needs of subgroups of the homeless, especially these families, led to the development of the Homeless Families Program. The design of the HFP reflects experiences from yet another Robert Wood Johnson Foundation/HUD joint initiative, the Program on Chronic Mental Illness, as well as HUD's Transitional Housing Program under the McKinney Act. Both of these programs reinforced the view that although permanent housing was absolutely necessary, it was not in itself sufficient if substantial segments of the homeless population were to achieve stability and self-sufficiency. Rather, these individuals and families appeared to need more comprehensive, individually tailored benefits involving permanent housing, health, social, and support services. The Program on Chronic Mental Illness,⁵ an effort to create more centralized local systems of care, also furthered the view that, in addition to new ways of delivering services, systemic efforts were needed to help vulnerable populations.

Thus, the first premise of the Homeless Families Program is that whereas some families are homeless for reasons that are primarily economic, others face more complex problems. For them, a lack of housing was not believed to be the sole cause of their homelessness, and housing alone was not the simple solution to it. Such families might need continuing and comprehensive health, housing, and supportive services in order to function in the community. The second premise is that these families need case management to help them get necessary services. Public funding has made a number of services and supports available to young families, through Aid to Families with Dependent Children; Medicaid; Maternal and Child Health; the Supplemental Food Program for Women, Infants, and Children (WIC); Social Service Block Grants; Head Start; and the McKinney Act funds. But these services are fragmented among agencies and may be difficult for homeless mothers to obtain.

The third premise is that although a number of communities do provide many of the services available for homeless families, these efforts are splintered, and a more systemic approach is needed. A modest infusion of grant money and housing subsidies could enable these communities to build comprehensive, coordinated service systems to ensure that these young families get the continuing services they need.

Each of the nine HFP sites received approximately $600,000 in grant money over five years to facilitate systems of care for homeless families and, within that context, to demonstrate a model of services-enriched housing for a group of families. The projects were led by either a city or a county public agency, a coalition or task force for the homeless, or another nonprofit provider. Guided by the HFP National Program Office and HUD, each project developed a memorandum of understanding with the local public housing authority. Through this agreement, each project received an allotment of approximately 150 Section 8 housing certificates⁶ from HUD to allocate to families with multiple problems, many of whom had not been on the existing waiting list of the public housing authority. For each family receiving a Section 8 certificate, the HFP lead agency was to provide or obtain services through case management. Robert Wood Johnson Foundation support of the program totaled $4.7 million, while HUD's contribution totaled $30 million in rental subsidies over five years.

EVALUATION DESCRIPTION

An evaluation was designed to answer three major questions:

Preliminary results of the evaluation are reported below.

DESCRIPTION OF PARTICIPATIONG FAMILIES

Of the 1,670 families accepted into the program and included in its management information system, 1,298 entered services-enriched housing. These families, the focus of the evaluation, were similar demographically to families described in other studies of homeless families and welfare recipients. The average family was headed by a woman in her late twenties or early thirties, with two children, at least one of whom was less than three years old. Ethnicity and education varied across the sites, most often reflecting the characteristics of the particular site. Service needs appeared to be more pervasive and severe for the HFP families than has been the case in other studies of homeless families and families receiving welfare.⁸

EVALUATION FINDINGS

Because the program targeted homeless families with multiple problems, there are limitations to the extent to which these data can be applied to the full population of homeless families. However, the results do clarify the complexity and the character of a segment of the population and help anchor an understanding of the broader population. Perhaps most important, taken together with the outcomes, we can learn what is possible even for families with the most complex challenges and needs.

FAMILIES IN NEED OFTEN PRESENT A WEB OF INTERRELATED AND DEEP-SEATED CHALLENGES. Families served through the HFP struggled with homelessness and residential instability for some time. On average, families experienced their first time without their own home about five years before they entered the program. In the eighteen months just before entering the HFP, most of the families spent time either in a shelter (approximately five months) or doubled up (five months). Families spent about seven months in their own homes, but rarely consecutively. The families moved frequently--about every three and half months in the eighteen-month period--before entering the HFP.

Housing instability is only one of the areas of difficulty for the families who entered the program. All areas of need were pronounced (Table 10.1) for the population of women served, especially in comparison to the general population. Additionally, the majority of families had multiple service needs in the areas of physical and mental health, substance abuse, education and training, and others. In fact, nearly a quarter were found to have current needs in all three major areas examined--human capital, physical health, and mental health-related--and 80 percent had needs in two or more categories.

MENTAL HEALTH NEEDS AND DOMESTIC VIOLENCE SURFACE AS TWO DISABLING FACTORS FOR THE FAMILIES. Perhaps the most marked areas of need that these women have tackled through much of their lives fall within the category of mental health. Even as children, more than half of them would have been considered at risk for future mental health problems because they had been placed in foster care, had run away for a week or longer, or had experienced physical and/or sexual abuse.

As adults, the vast majority of women in the HFP had one or more indications of mental health need. A sizable percentage, 15 percent, were hospitalized one or more times for a mental health problem, 3 percent in the year before they came into the program. More than half were considered psychologically distressed and in need of further evaluation for depression.

One of the most troubling findings in this area is the extent to which these women have been victims of abuse and, in turn, have tried to hurt themselves. Nearly all (81 percent) reported some type of abuse by a former partner, and 65 percent reported one or more severe acts of violence by a past partner. In addition, more than a third cited domestic violence as a reason for moving in the five years before they entered the program.

Reports of suicide attempts by these women were almost ten times as frequent as they were for the general population (28 percent versus 3 percent). Of those who reported ever attempting suicide, more than half of this group--57 percent--reported multiple attempts. Drug overdosing was the most common method used. The seriousness of the suicide attempts is highlighted by the fact that 43 percent of the most recent attempts resulted in a hospitalization and, in fact, account for the majority of the mental health hospitalizations reported.

THROUGH THEIR PARTICIPATION IN THE HFP, FAMILIES INCREASE THEIR ACCESS TO AND USE OF AN ARRAY OF SERVICES. A critical discovery made as families entered the Homeless Families Program was that the majority of them were not receiving needed services.⁹In fact, the vast majority--70 percent or more of those in need--were not receiving mental health, dental, and alcohol treatment services. A smaller but still sizable percentage of families, 58 percent, were not receiving needed drug services.

Through the program, access to services appeared to improve (Table 10.2). The biggest increases were experienced in mental health services, followed by alcohol and drug services. Because families had reported relatively high access before they entered the HFP, health services did not change while they were in the program. Access to dental services increased slightly, but adult family members continued to have a high level of unmet need.¹⁰

FAMILIES ACROSS THE SITES HAVE MADE CONSIDERABLE GAINS IN RESIDENTIAL STABILITY. Despite years of instability, families achieved substantial residential stability after they entered the program. At eighteen months after entering the program, more than 85 percent of the families were still stably housed in the six sites that provided data. This represents more than a doubling of the time the families spent in permanent housing for the same period before they entered the program. The remainder of families either lost their Section 8 certificate because of one or more violations--fraud, for example--or voluntarily returned the certificate (for example, to move to another state).

After thirty months, rates of stability continued to be high, but more differences emerged among program sites. In three of the six sites for which data are available, more than 80 percent of the families were known to be in permanent housing, typically with the original Section 8 certificates. In the other three sites, fewer than 65 percent were known to be residentially stable in permanent housing. Analyses to date have identified few stable predictors of housing loss within and across sites. When the data are examined through the use of more complex statistical analysis, they show that the loss of housing is related to mothers' reports of current severe violence as well as being pregnant or having an infant at the time they entered the program.¹¹ These two predictors are important and troubling, and both will continue to be explored in future analyses.

FAMILIES HAVE MADE LITTLE AND ERRATIC PROGRESS TOWARD SELF-SUFFICIENCY AND CONTINUE TO HAVE CONSIDERABLE DEPENDENCE ON FEDERAL AND STATE SUPPORT. Participation in education, job training, and employment has fluctuated throughout the Homeless Families Program. The data from the management information system do not permit a sensitive and complete analysis of the changes in these areas, but for those sites where data are available, it appears that about 40 percent of the primary parents have attended school at least once during the program, and about half received some employment or vocational service, including counseling and job training. About half of the primary parents also reported working some time during this period.

When families leave the program, having received a year or so of case management services, 20 percent of the primary parents are known to be working, compared with 13 percent when they entered.¹² The increase is slight and not uniform, which means that some individuals who were employed when they entered the program were not working when they left it. Although some sites showed increases in the number of people with jobs, in most cases the increases were not statistically significant. In addition, a few reportedly had a job lined up (5 percent); about 9 percent were in job training, and 16 percent reported being in school. Taken together, 39 percent were working, preparing to work, or obtaining further education.

A number of factors may explain the unstable movement toward self-sufficiency shown by most families in the HFP. One factor is that the projects focused more on helping families remain residentially stable than on becoming self-sufficient; another is that a year or so after entry into the program may still be too early to determine accurately how successful a family will be at self-sufficiency. Indeed, residential stability and self-sufficiency may go hand in hand, since part of the rationale behind the program is that residential stability is required before anyone in a family can begin to think about holding a full-time job, continuing schooling, and moving off public assistance. Even after thirty months, the data on residential stability indicate that the vast majority of families who remain in permanent housing were still receiving subsidies, and the available information on those who returned the Section 8 certificates or lost them suggests that these families were most likely in unstable living and working situations. Few appeared to have moved off public assistance, and very few off Section 8.

Lack of child care may be a major barrier to achieving self-sufficiency. Across the sites, 72 percent of the families in the program were reported to need child care services at intake and/or during the program. Only 41 percent of the families needing child care services were reported to have received them at least once throughout the program.

It is also important to recognize the range and the magnitude of the problems faced by most of the families in the HFP. Their past lives often have been challenged by economic and personal traumas. Even though the data suggest that the HFP has given them access to some services, many of the problems are long-standing and unlikely to disappear instantly. Despite increased housing stability, many families remain vulnerable to the ordinary challenges of life, let alone to broader reforms. The primary parents in these families often cycle through jobs, education, and services in response to other events in their lives--reuniting with a former batterer, for instance, or returning to drug or alcohol abuse. The overwhelming majority of HFP families received multiple public benefits (AFDC, food stamps, school lunches) and were likely to be substantially affected by changes in welfare and related benefit programs. Whether the new programs can adequately prepare these families to enter the workforce within the time frame stipulated remains to be seen. Families' involvement in similar efforts in the past appear to have limited benefit, at least in the short run.

The evaluation was designed to examine how service systems can be developed, organized, and sustained to respond to the needs of homeless families, especially those with multiple needs.

THE SERVICE "SYSTEM" FOR HOMELESS FAMILIES IS ILL-DEFINED AND FRAGMENTED AND INVOLVES MULTIPLE SERVICES AND SYSTEMS. In fact, the term homeless system is a misnomer; for families in particular, there are often multiple systems that provide services. Each of these systems has its own level of fragmentation, and the connections among the various systems are even looser.

Three types of service systems are relevant to homeless families: homeless services, mainstream services, and coordinating services. Homeless services are specifically designed for individuals and families who lack a regular and permanent place to stay. In most places, these services include shelter and transitional housing, as well as food, clothing, and furniture assistance.

Mainstream services refer to those needed by most low-income families, homeless or otherwise. These include housing, income support, child care, health and dental care, mental health services, counseling for domestic violence, alcohol and drug treatment, and others. Many mainstream service systems are well established but were not designed to serve a homeless population. An examination of the systems across the nine HFP sites revealed few absolute gaps in services for homeless families. Every site had some type of child care available, for example, and some form of mental health services. What was more common and amazingly consistent across the sites were gaps in specific types of services resulting from limitations in capacity, eligibility restrictions, high costs, and constraints on accessibility. This finding conflicted with a key program premise, that services for homeless families did exist.

Common service gaps include affordable housing, with a desperate need for larger units and subsidies to ensure affordability; residential alcohol and drug programs that permit mothers to keep their children with them; affordable child care; mental health services for adults and children who are not seriously and persistently mentally ill but are trying to cope with issues of domestic violence, depression, and other problems stemming from their past instability; general legal assistance; and dental services for adults beyond extractions and fillings. Often, transportation is not available or not affordable, which creates an additional barrier to services.

Coordination services for families, if available, typically entail some form of case management. When case management was available in shelters and transitional housing, it was generally in short supply and of limited duration, with little if any follow-up once a family moved into permanent housing. Some mainstream services, such as job training, had case management attached to them, but it was also generally of limited duration and involved only services brokering. At the HFP sites, case management was not routinely available to homeless families moving into permanent housing. Therefore, unless families were in the HFP, they were largely on their own to negotiate the web of systems and services when they moved into permanent housing.

THE HFP HAS LED TO SOME FIXING OF THE SYSTEM. A goal of the HFP was to change the systems for dealing with the homeless. By and large, however, the system activities of the HFP projects did not result in broad-based changes but rather in some temporary or small-scale fix to improve service delivery for the families needing the service. The most common HFP system activities were project fixes: filling service gaps for the families in the HFP services-enriched housing. Mental health services, for example, were a critical gap for families participating in the program. Unable to obtain timely access for families needing therapy and other services to cope with the effects of domestic violence or other problems, a number of the HFP sites responded by hiring individual therapists to work with their families. The HFP efforts thus improved the accessibility of mental health services for families in the project, but they did not change how the overall mental health system relates to homeless families.

Other HFP activities created "system fixes" in which services were increased or improved for homeless families other than those in the demonstration. For example, the HFP National Program Office developed ways to use the Federally Qualified Health Center provision of Medicaid to create additional resources to cover the costs of case management and other services. This typically involved working with a single community health center or a Health Care for the Homeless clinic. Although the effort had the potential to spread systemwide, it was relatively circumscribed and independent of other reforms within the health care system.

System changes--enduring and far-reaching reformulations or modifications in the structure of a system--were rare in the Homeless Families Program. The one exception involved changes in the role of the public housing authority. Through their participation in the program, several housing authorities increased their awareness of, and sensitivity to, the needs of homeless families. They became more active participants in developing supportive housing for this population. The efforts of The Robert Wood Johnson Foundation and HUD appear to have been key factors in facilitating this change.

THE COMPLEXITY AND THE FRAGMENTATION OF SERVICE SYSTEMS FOR FAMILIES TRANSITIONING FROM HOMELESSNESS MAKES TRUE CHANGE DIFFICULT, IF NOT IMPOSSIBLE, FOR SMALL PRIVATE INITIATIVES. Although ambitious, the efforts of the Homeless Families Program to reform systems were in many ways overpowered by the complexity of the systems that needed restructuring. First, restructuring systems to meet the needs of homeless families meant dealing not with just one system but with several systems. Most of these systems are large and complex and serve a variety of people, of whom homeless families generally make up a small and relatively invisible fraction. Significant changes in mainstream systems are unlikely to be driven by the needs of this subpopulation of clients.

Second, the positioning of the Homeless Families Program in each community rarely gave it the clout needed to restructure or build a homeless service system. In some cases, the HFP was a program within a city or county health agency. At best, it could call upon the agency leadership to coordinate the efforts of other agencies. In other cases, the program was located in coalitions or task forces for the homeless that might have had the influence to bring groups together and to identify needed changes but were not in a position to make the changes happen.

Third, the resources that the HFP brought to the communities were too small to allow for major restructuring. The modest funding provided to each site was used to support a program director and case management or program staff. Although the Section 8 certificates contributed from HUD for this initiative were not inconsequential, they were designated as housing subsidies, not flexible funds that could be used to create new systems or strategies for action. Ironically, just having these certificates caused projects to focus much of their limited resources on the development and implementation of the services-enriched housing and less on the more nebulous goal of creating systems change.

SYSTEMS CHANGE AT THE LOCAL LEVEL MAY BE SPURRED BY FEDERAL AND NATIONAL LEADERSHIP The one area where systems change appeared most consistently involved the public housing authorities. Across the sites, the housing authorities became stronger and more vocal participants in supportive housing for families. In addition to providing concrete advice that spurred collaboration between housing authorities and the HFP lead agencies, HUD and the Foundation required them to work out a Memorandum of Understanding. The framework for these memoranda, developed by HUD and the HFP National Office, stipulated the nature of collaboration that was expected between the housing authority and the HFP over the five years of the program. The memorandum was often used, particularly in the early stages of the program, as a tool to prod the housing authorities to modify procedures, cut red tape, and institute other changes needed to get the program off the ground. In addition, by having a written memorandum, the projects were less susceptible to internal changes (such as changes in executive directors of the housing authorities) that could otherwise threaten the agency's involvement in the HFP.

THE ABILITY TO ENGINEER AND MEASURE SYSTEMS CHANGE IS HAMPERED BY THE LACK OF A THEORY OF SYSTEMS CHANGE. The Homeless Families Program had no articulated theory of systems change. Absent from the projects was a perspective of what the ideal system should be for families in order to break the cycle of homelessness. The goals were nonspecific, and the steps needed to achieve the goals were not detailed. None of the projects was preceded by an assessment of the needs, gaps, and strengths in the system. To fill this gap, the evaluation developed a framework of the ideal system to use as a tool for benchmarking the varied activities of the sites.

In addition, there was no explicit strategy for bringing about change. The desired outcome was a system that would be coordinated, accessible, and comprehensive. The general thinking was that the projects would work with other agencies to determine the changes that were needed in the current system, and either reorganize it or identify and leverage additional resources for new and enhanced services, or both. Ironically, the one very positive initiative, the Memorandum of Understanding, was not consciously viewed as a vehicle for systems change but as a mechanism to ensure that the Section 8 certificates would be dedicated to this program. The system changes stimulated by the memoranda were an unexpected by-product.

The evaluation was also designed to provide a detailed look at the implementation of services-enriched housing. Little is known about how best to meet the needs of families who have been homeless and have had other problems for years. One of the more widely touted strategies has been to provide services, particularly case management, to families for some period of time after they move to permanent housing. The Homeless Families Program was the first large-scale attempt to provide services-enriched permanent housing to families. HUD's Shelter+Care and Supportive Housing Programs, which also combine case management services and housing, are a major part of HUD's continuum-of-care strategy for homeless families and individuals. Despite the increasing use of case management, there has been little explicit study of its effectiveness with homeless families. The HFP provided an unparalleled opportunity to study case management and other key aspects of services-enriched housing.

DESPITE A COMMON MODEL OF INTERVENTION, THE AMOUNT OF CASE MANAGEMENT PROVIDED TO HOMELESS FAMILIES VARIES DRAMATICALLY There was a remarkable similarity in the background and training of the case managers hired at the nine HFP sites, in the types of activities they conducted, and in the services they provided.¹³ The vast majority of case managers working in this program were women in their thirties or forties, with a bachelor's or higher degree in social work or a related field, who had been working as case managers for five or fewer years. Most reported that their time with families was spent arranging services, making routine visits or calls to families, and working with families on skills development issues, such as budgeting or problem solving.

Despite the similarities, the differences in the amount of case management provided to families are striking. Although the average family received fifteen hours of contact during their first twelve months in the program, 33 percent of the families received less than six hours, whereas nearly 20 percent received more than twenty-four hours, and 5 percent received more than fifty hours.¹⁴

Sites differed greatly with respect to the intensity of the case management offered; statistically they can be grouped into four levels. At the high end, families at one site received an average of fifty-two hours of case management during their first year in the program, or about an hour a week. All the other sites had relatively less intensive case management, with case managers in most projects meeting with each family about one hour every two to three weeks. In two sites, families met with their case manager less than one hour a month.

IMPLEMENTING INTENSIVE CASE MANAGEMENT FOR HOMELESS FAMILIES REQUIRES A DIFFERENT APPROACH THAN THAT USED IN THE HFP. Intensive case management was to be one of the cornerstones of the Homeless Families Program, with case managers spending as much time as needed with families. Although no explicit definition of intensive case management was provided, it was generally understood to mean at least one hour a week of face-to-face contact with each family. A key insight gained from this evaluation is that even when the HFP projects were implemented as designed--a caseload of one manager to twenty families, working with families a year or longer, visiting families in their home--the expected level of intensity could not be achieved.

With the need to spend time on paperwork, phone calls, meetings, travel, and so on, case managers spent little over an hour a month with each family, and they generally had one-quarter of the day to meet face-to-face with families. Assuming a one-to-twenty caseload, only if case managers had at least half of each day to meet with families could they reach the program goal of an hour a week in direct contact with each family. In order to achieve the desired level of intensity, the caseload and the work responsibility of case workers needs to be reduced, and they should be teamed with lay helpers and other support mechanisms.

INDIVIDUALS SKILLED IN LOCATING HOUSING AND WORKING WITH LANDLORDS MAY BE AN IMPORTANT SERVICE FOR FAMILIES WHO HAVE NOT HAD PREVIOUS SUCCESS WITH HOUSING, ESPECIALLY IN TIGHT HOUSING MARKETS. An innovation at several project sites was the appointment of a housing locator. This person identifies promising housing, recruits landlords, helps families find housing, and performs other related activities. A locator has typically been used at sites where the low-income housing market is especially tight and landlords willing to take Section 8 certificates are not numerous. For most of the public housing authorities, this position was new and welcome; in at least one site, the housing locator was continued within the housing authority so that he could work directly with any family needing assistance. At another site where the housing market was beginning to tighten over the last year of the HFP, the housing authority was seriously exploring the possibilities of hiring a housing locator to help families find landlords willing to take HUD Section 8 certificates.

IMPLICATIONS OF THE EVALUATION FINDINGS FOR HOMELESS FAMILIES AND BEYOND

The findings outlined below are pertinent not only for initiatives directed toward homeless families but alsofor a broader set of initiatives aimed at building and changing systems.

CONCLUSION

Homeless families, especially those with multiple problems, are challenged by the reforms under way in welfare, health, and housing. The gains in residential stability achieved by the families in the Homeless Families Program are encouraging, particularly in view of the long histories of housing instability and other life struggles they have endured. However, families' reliance on federal support for their basic needs and their lack of steady progress in employment raise questions about how long their situations will remain stable. Moreover, the HFP findings suggest an ominous situation for other families who are currently homeless, particularly those who mirror the profile of the HFP families. Since these families have consistently fallen out of the system, the only real gain they have experienced in the last five years is staying in permanent housing. The Section 8 housing subsidy, in particular, appears to have pushed the majority of families above the threshold. Few have gone beyond that, however, and most continue to lack jobs, child care, and often sufficient education. Cutbacks in welfare present a formidable challenge for these families. The lack of consistent employment among families during their stay in the HFP suggests that these families continue to remain at risk of homelessness and that, unless major changes are also provided in the employment environment, the risk increases with welfare reforms. If Section 8 reforms also limit the time Section 8 certificates can be used, a return to homelessness for many families seems inevitable.

Endnotes