urban institute nonprofit social and economic policy research

Evaluation of Continuums of Care for Homeless People

Final Report

Read complete document: PDF


PrintPrint this page
Share:
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: May 01, 2002
Released online: May 01, 2002

This report was prepared by the Urban Institute under U.S. Department of Housing and Urban Development Contract No. DU 100C000018572, Task Order 7. It was published by the Department of Housing and Urban Development in May 2002. The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.


Foreword

A Continuum of Care is a local or regional system for helping people who are homeless or at imminent risk of homelessness by providing housing and services appropriate to the whole range of homeless needs in tl'1e community, from homeless prevention to emergency shelter to permanent housing. In 1995, the Department of Housing and Urban Development (HUD) implemented the Continuum of Care approach to streamline the existing competitive funding and grant-making process under the McKinney-V ento Homeless Assistance Act and to encourage communities to coordinate more fully the planning and provision of housing and services for homeless people.

In 2000, HUD contracted with the Urban Institute and its subcontractor, ICF Consulting, Inc., to conduct an assessment of the Continuum of Care approach. This study examined 25 Continuums of Care in all regions of the country .At the outset of the study, the Department was unsure whether the Continuum of Care concept had achieved detectable results in fostering a coordinated response to homelessness. For this reason, the study intentionally identified communities to study that appeared to have been successful in implementing a Continuum of Care. Because they had been able to convey a sense of progress in meeting their goals, these communities also happened to receive more than a typical share of funding under HUD's competitive allocation process. Among these high performers, the study sought to ensure that the study communities were broadly representative geographically, but the sites cannot be taken as representative of all of the communities that are operating HUD-assisted homeless programs.

The completed study provides a rich array of information on the activities of high-performing Continuums of Care and documents the extent of progress of the studied communities along many dimensions of a comprehensive approach to homelessness prevention and remediation. On balance, the report concludes that for the high-performing communities studied, HUD's implementation of the Continuum of Care funding process stimulated increased communication within local communities in their response to homelessness. As a result, respondents generally agreed that more people have received more services and participate in more and better coordinated programs than before as a consequence of the Continuum of Care approach. In the most advanced communities, the response has moved beyond planning to obtain HUD funding to the far broader goal of attempting to integrate all available funding and services to try to end homelessness.

The Department is continually refining and reassessing its approaches to program delivery, and homelessness is no exception. HUD's FY 2003 budget request includes a proposal to Congress to consolidate the existing McKinney programs and change the way homeless activities are funded. The Department's proposal would ensure that localities are provided with a predictable source of flexible funds through a streamlined, locally- driven process. It is designed to preserve the achievements documented in this report while providing more flexibility to local communities in addressing homelessness, as well as incentives for communities that have been less successful in implementing the Continuum of Care to improve their planning and program delivery.

This report is a significant addition to our knowledge about homelessness. It provides important insights into local responses to the problem and identifies issues that must be resolved as the nation grapples with the difficult and serious problem of homelessness.

Lawrence L. Thompson,
General Deputy Assistant Secretary for Policy Development and Research


Table of Contents

EXECUTIVE SUMMARY
     Purpose of the Research
     Who, What, Where, and When?
     Why This Study Is Important
     Key Findings
          How Communities Organize Themselves to Respond to HUD and to Homelessness
          What Local Homeless Assistance Systems Look Like
          Navigating the Homeless System
          Involvement of Mainstream Agencies and Services
          CoC Planning for and Coordination of Homeless Services
          How the CoC Application Process Has Shaped Planning and Provision of Services
     Implementing Data Systems: A Few CoCs Far Ahead of the Rest
     Implications for Research
     Implications for Practice
     Implications for Policy

CHAPTER 1: INTRODUCTION AND OVERVIEW OF THE STUDY
     Introduction
     Background
          The Growth of America's Shelter Capacity in the 1980s and 1990s
          Growth of the Homeless Population
          The Need for Continuums of Care
          First Efforts Toward Continuums of Care
     Methods
     The Rest of This Report
     Caveats
     A Word on Numbers, or their Absence

CHAPTER 2: POWER AND PEOPLE—"LEAD" AGENCIES AND LEADERSHIP
     No Formal Authority
     Public-Private Arrangements
     Formal Independent Entities
     The Last Community
     People
     Beyond "The Community"
     Implications

CHAPTER 3: LOCAL HOMELESS ASSISTANCE SYSTEMS
     The Meanings of "Continuum of Care"
          Limited View
          Geographically-Influenced View
     Components of CoC Systems
          Prevention
          Emergency Shelter
          Transitional Housing
          Permanent Supportive Housing
          No/Low Demand Shelter and Housing Programs
          Importance of Supportive Services
          Permanent Affordable Housing
     Implications

CHAPTER 4: NAVIGATING THE SYSTEM
     System Entry and Flow
          System Entry: Fragmented, No Wrong Door, and Centralized Approaches
          Reaching Out to the Hard to Serve
          Movement Within the System: Ensuring that Individuals Obtain Needed Services
     Locked Out: Homeless Sub-groups That Have Not Fared Well
          Persons with Mental illnesses, Addictions, or Both
          Youth
          Families
          Ex-Offenders
     Implications

CHAPTER 5: INVOLVEMENT OF MAINSTREAM AGENCIES AND SERVICES
     Introduction
     Strategies for Involving Mainstream Agencies
          Staff with Systems/Service Integration Responsibilities
     vCreating a Local Interagency Coordinating Body
          Having a Centralized Authority
          Co-Locating Services
          Interagency Management Information Systems
     "Reverse" Integration
     Business and Corporate Involvement
          Fort Lauderdale/Broward County, Florida
          Columbus/Franklin County, Ohio
     Implications

CHAPTER 6: PLANNING FOR AND COORDINATION OF SERVICES TO HELP HOMELESS PEOPLE
     The Gaps Analysis
          Data Sources and Methods Used in the Gaps Analysis
          Limitations and Criticisms of the Gaps Analysis
     Setting Priorities in the Gaps Analysis Process
     Ranking Homeless Proposals
          The Proposal Ranking Process
          Who Is Included in the Proposal Ranking Process?
     Scope of Planning for Homeless Programs and Services
          Scope of Planning: Limited to the HUD-funded Continuum of Care
          Scope of Planning: The Larger Homeless Assistance System
          Scope of Planning: Strategic Multi-Year Planning for the Larger Homeless Assistance System
     Planning for Homeless Programs and Services in Relation to Other Planning Efforts
     Stability and Flexibility of the System and the Impact of Renewals
          Background on HUD CoC Funding and Renewals
          Impact of Project Renewals at Sites Visited
     Other Issues and Problems with the Planning Process
     Implications

CHAPTER 7: THE HISTORY OF THE SYSTEM
     Planning for the First CoC Application: Seamless Transitions
          Effect of the CoC Application
     Planning for the First CoC Application: From Informal/Fragmented to Structured/Comprehensive
          Effect of the CoC Application
     Planning for the First CoC Application: From Nothing to Something
          Effect of the CoC Application
     Developments Since the First CoC Application: Shifting Directions
     Next Steps: The Primary Issues Facing These Systems
     Changes That Could Pose Difficulties
     Implications

CHAPTER 8: DATA SYSTEMS
     HMISs in the Most Advanced Communities
          Boston
          Columbus/Franklin County, Ohio
          Balance of Cook County, Illinois
          Memphis/Shelby County, Tennessee
          Montgomery County, Maryland
          Rhode Island
          Shreveport, Louisiana and Nine Surrounding Parishes
          Washington, DC
     Status of HMISs in Other Communities
     Barriers to Development and Implementation
          Lack of Power to Collect Data
          Cost
          Time
          Confidentiality
          Compatibility of Different Data Systems
          Sharing Case Management Information
     Status of Performance Measurement Systems
          Boston
          Columbus/Franklin County, Ohio
          Washington, DC
     Other Types of Data Collection and Limits of an HMIS
     Implications
     Prospects for Functioning Data Systems
     HMIS Implementation
     Use of Performance Information in Funding Decisions

CHAPTER 9: SUMMARY AND IMPLICATIONS
     Question 1: Local Homeless Assistance Networks—Programs and Functioning
     Question 2: Inclusiveness of CoC Planning, Coordination, and Program Support
     Question 3: Integration with Mainstream Programs and Services
     Question 4: Role of Data and Statistics in Planning
     Question 5: Effects of HUD's Application Requirements
     Question 6: Success in Ending or Reducing Homelessness
     Impressions and Implications
          Lack of Affordable Housing
          Steps, Stages, or Levels
          Importance of the Local Environment
          Importance of Philosophy about Homelessness
          Evolving Views of Appropriate Programs and Services
          Effects of CoC Funding Structure on Cooperation and Coordination
          Community Differences and Need for More Information on Rural CoCs
          Possible HUD Actions Suggested by Respondents

REFERENCES

ACRONYMS

APPENDIX A: MOST COMMON SOURCES OF FUNDING FOR HOMELESS ASSISTANCE PROGRAMS FROM THE U.S. DEPARTMENT OF HOUSING AND URBAN DEVELOPMENT
     Emergency Shelter Grant Program (ESG)
     Supportive Housing Program (SHP)
     Shelter Plus Care Program (S+C)

APPENDIX B: DESCRIPTIONS OF THE COMMUNITIES VISITED

NORTHEAST
     Delaware
     Rhode Island
     Southwestern Pennsylvania
     Boston
     Washington, DC
     Essex County, New Jersey
     Montgomery County, Maryland

SOUTH
     Shreveport, Louisiana and nine surrounding parishes
     Fort Lauderdale/Broward County, Florida
     Orlando/Orange County, Florida
     Memphis/Shelby County, Tennessee
     Winston-Salem/Forsyth County, North Carolina

MIDWEST
     Chicago, Illinois
     Balance of Cook County, Illinois
     Lake County, Illinois
     Madison/Dane County, Wisconsin
     Washtenaw County/Ann Arbor, Michigan
     Columbus/Franklin County, Ohio
     St. Paul/Ramsey County, Minnesota

WEST
     Alameda County, California
     Denver, Colorado and five surrounding counties
     San Francisco, California
     Long Beach, California
     Phoenix/Maricopa County, Arizona
     Tacoma/Pierce County, Washington


Executive Summary

Purpose of the Research

The purpose of this project was to examine Continuums of Care for homeless people throughout the United States, to understand their development, their current structure, and their likely future. A Continuum of Care (CoC) is, ideally, a system for helping people who are or have been homeless or who are at imminent risk of homelessness. A full CoC includes prevention, outreach and assessment, emergency shelter, transitional housing, permanent supportive housing, and affordable housing, plus supportive services in all components. HUD has promoted the CoC concept through much of the 1990s, and has structured its competitive funding under the McKinney-Vento Homeless Assistance Act to further CoC development.

This study sought to answer several questions about the ways in which local communities are organized into CoCs to address homelessness:

  • What do local homeless assistance networks look like, how do they work, and whom do they serve?
  • Are all the important players, or their representatives, included in planning the local CoC and coordinating their programs and services?
  • How well are homeless and mainstream services integrated?
  • What goals is each jurisdiction trying to accomplish with its CoC—helping homeless people, ending homelessness, or some combination—and how does its concept of its "continuum" further those goals?
  • What role does data or statistics about homeless people, services, and program performance play in the planning process and in decisions about what to support?
  • How has the HUD requirement for a coordinated community-wide application affected development of CoCs, client access to and receipt of needed programs and services, inclusion of relevant homeless-specific and mainstream players, and data-based decision making?

Who, What, Where, and When?

In 2000, the Department of Housing and Urban Development's Policy Development and Research Office funded the Urban Institute and its subcontractor, ICF Consulting, Inc., to conduct a study to answer the research questions. This study examined CoCs in all regions of the country. Telephone interviews were first conducted with key informants in each community. Site visits by two-person teams followed, during which researchers interviewed local homeless assistance system planners, providers, case managers, funders, and consumers. Additional documentation included applications for HUD funding, local studies and reports, and local statistics if available.

A sample of 25 of the more than 300 applicants for CoC funding was selected for the study. These applicants could be single jurisdictions or combinations of jurisdictions. Throughout this report we refer to these applicants as "communities." These communities are distributed throughout the United States, and represent a wide variety of community configurations. They are:

NORTHEAST MIDWEST
1. Delaware
2. Rhode Island
3. Southwestern Pennsylvania
4. Boston
5. Washington, DC
6. Essex County, New Jersey
7. Montgomery County, Maryland
13. Chicago
14. Balance of Cook County, Illinois
15. Lake County, Illinois
16. Madison/Dane County, Wisconsin
17. Washtenaw County/Ann Arbor, Michigan
18. Columbus/Franklin County, Ohio
19. St. Paul/Ramsey County, Minnesota
SOUTH WEST
8. Shreveport, Louisiana and 9 surrounding parishes
9. Fort Lauderdale/ Broward County, Florida
10. Orlando/Orange County, Florida
11. Memphis/Shelby County, Tennessee
12. Winston-Salem/Forsyth County, North Carolina
20. Alameda County, California
21. Denver, Colorado and 5 surrounding counties
22. San Francisco
23. Long Beach, California
24. Phoenix/Maricopa County, Arizona
25. Tacoma/Pierce County, Washington

The final sample includes two whole states and four multi-county regional applicants. It also includes one applicant representing an urban county minus its largest jurisdiction and one other that were eligible to apply on their own and did so (Cook County minus Chicago and Evanston). Four applicants were counties; two did not have any cities eligible to apply in their own right, and two had four or five such cities that had opted to apply jointly with the county. Lastly, we had five cities on their own and ten city/county jurisdictions. This sample is skewed in several important ways. First, we deliberately selected communities whose CoC applications received high ratings from HUD (22 of the 25 were in the top 10 percent of FY 2000 applicants). Second, many communities were selected specifically because they exemplified something unique and/or potentially interesting that might provide valuable lessons for other CoCs. Third, due to resource constraints, the sample underrepresents both the most and the least populous communities. With the exception of Chicago and Cook County, we did not select the nation's largest jurisdictions, nor did we have adequate representation of truly rural areas.

Once site visits were completed, the study team summarized and integrated findings for each research issue across the 25 communities, and wrote this report. Where communities are mentioned by name, the text has been reviewed for accuracy by the primary contact we had for that community.

Why This Study Is Important

This study takes the first systematic look at Continuums of Care for homeless people as they have developed in communities across the United States. It examines how development has been affected by the availability of federal funding for homeless assistance programs and HUD's requirement that to get these funds, communities must organize themselves to provide coherent systems of care. Its findings reflect what communities have accomplished and what factors have been most important in stimulating and supporting CoC development. They provide guidance for HUD and for the communities themselves in promoting future development to serve homeless people and to prevent and reduce homelessness.

In 1987, Congress passed the first federal law specifically addressing homelessness. The Stewart B. McKinney Homeless Assistance Act of 1987, later named the McKinney-Vento Homeless Assistance Act, provides federal financial support for a variety of programs to meet many needs of homeless persons. The housing programs it authorizes are administered by HUD's Office of Special Needs Assistance Programs. McKinney-Vento has been a major federal avenue for supporting the development of homeless assistance systems. Funding in recent years has been slightly more than $1 billion a year.

Emergency Shelter Grant funds support emergency shelter and, to some extent, activities that prevent households at imminent risk from becoming homeless. Emergency Food and Shelter Program funds may also be used for these purposes, as well as for emergency feeding programs. Programs like these were the mainstay of homeless assistance through most of the 1980s.

McKinney-Vento added other program types for which experience with homeless individuals and families during the 1980s had indicated the need. The intention was to offer a federal incentive to local communities to develop comprehensive homeless assistance systems. These systems were envisioned as being able to offer the array of services and skills necessary to address the needs of all types of homeless people—from first-time emergency situations to long-term chronically homeless substance abusers, people with severe and persistent mental illness, people with HIV/AIDS, and other disabling conditions. New program types included transitional housing for families and people with disabilities and permanent supportive housing for people with disabilities through the Supported Housing Program. They also included Shelter Plus Care funds to develop more permanent supportive housing for people with disabilities, and Section 8 Moderate Rehabilitation SRO funding to increase the supply of this type of housing and give homeless individuals priority for its units. In addition, McKinney-Vento monies can be used to fund supportive services either by themselves or attached to the various shelter/housing programs. Also part of the initial premise was that once programs begun with federal funds had proved their merit locally, local governments and other funders would step up and assume funding responsibility for their survival. While McKinney-Vento has stimulated a great deal of local funding as part of its required financial match, most programs begun with these federal funds continue to rely on them, thus creating tensions between the need to continue funding for existing programs and the desire to develop new programs and approaches.

For the first six or seven years of federal funding for homeless assistance programs, HUD did not impose any requirements for systematic planning and comprehensiveness at the local level. Emergency Food and Shelter Program funds and Emergency Shelter Grant funds were and still are distributed on a formula basis to eligible communities. They do not require competitive applications, nor are communities receiving them required to demonstrate how these funds will be used in ways that complement other federal homeless assistance dollars.

Funds from the remaining programs are distributed by the mechanism of national competitions. In 1987 when the first McKinney Act passed, individual programs applied to these national competitions. Only a few communities made serious attempts to plan or structure their homeless programs and services. Most communities could not be described as having "a system," and providers developed programs for which they saw a need and could find money, without regard to the larger pattern of services in the community.

Just as there was no comprehensive planning or intentional structure to the array of homeless assistance programs and services before the mid-1980s, initial uses of funds under McKinney and later McKinney-Vento increased capacity but were not structured to "complete" a service system. Nor were they structured to assure that homeless people received all the care they needed that the system had to offer. Since 1996, in contrast, the Department of Housing and Urban Development (HUD) has used a competitive application process for the supportive housing programs authorized by McKinney-Vento to promote the development of Continuums of Care (CoCs) for homeless people in communities throughout the nation. The hope has been that by requiring communities to come together to submit a single comprehensive application, HUD could stimulate them to move toward greater structure and a more strategic vision of their programs and services for homeless people. This increased system structure and rationality was expected to improve services for homeless people and increase the chances that their needs would be met.

This study was designed to assess the degree to which these hoped-for changes have taken place.

Key Findings

How Communities Organize Themselves to Respond to HUD and to Homelessness

Trying to categorize, or even characterize, the nature of leadership in the 25 communities we visited stimulates a great appreciation for the truly enormous variety of arrangements that different communities have evolved to plan for and deliver homeless assistance services. Not only are the arrangements many, but in most cases the arrangements seem to be effective in their own setting at achieving the goals the local communities set for themselves. Given the varied organizational patterns and the importance of housing CoC organizing in an enthusiastic entity, federal pre-specification of a CoC planning configuration would run the risk of weakening CoC planning in many communities.

Three general organizational patterns could be identified:

  • The CoC leadership in 10 communities has no formal authority. These communities relied on voluntary cooperation, as no entity is really in a position to make anyone else do anything.
  • Another 10 communities have public-private arrangements with both a board and at least one government agency playing a strong role. In some of these communities it was reasonably easy to identify which entity was in the lead and which had a support function, but in others it was not.
  • In 4 communities, formal independent entities orchestrate homeless planning. These independent entities may or may not be the formal recipient of the HUD CoC dollars and may or may not enjoy support from mainstream agencies.

The final community in our sample cannot be categorized into any of the foregoing groups. Formal control rests with a government agency, which had long handled the CoC application without paying much attention to input from the larger community or the homeless assistance network, although such input was offered. In recent years the process has been opened up and involvement of the larger community is growing.

Several positive results were evident from the CoC process. The CoC process has stimulated significantly increased communication and information-sharing among homeless service providers, and often also among homeless-specific entities and mainstream entities. Information-sharing has, in turn, led to increased coordination of programs and services, both because people know more about the services available from different components of the homeless assistance network and because people who talk together often go on to develop coordinated or joint programming. We found well-functioning year-round larger planning and coordination processes in all configurations, even those with no formal powers.

Other contributors to a CoC's success are worth mentioning. Leadership is essential to effective CoCs, and should be accepted and appreciated wherever it is found. The local environment beyond homeless-specific services plays a key role in determining how well CoC planners and participants will be able to go in fulfilling a vision of a comprehensive CoC or, even better, of ending homelessness. For example, business and corporate interests can be strongly supportive, both conceptually and financially, or indifferent, or resistant.

What Local Homeless Assistance Systems Look Like

The 25 local homeless assistance systems fall into one of three categories:

  • Most of the 25 localities view the CoC as a single system covering a locally defined area and including all of the following components—prevention, outreach and assessment, emergency shelter, transitional housing, appropriate supportive services, permanent supportive housing, and permanent housing.
  • Eight of the 25 communities have a limited view of a CoC, seeing it mainly in terms of the activities that are eligible for CoC funding from HUD. Their focus is almost entirely on obtaining HUD funding.
  • Geographical considerations have led a few communities in this study to have what could be characterized as sub-CoC systems serving different regions within the larger CoC geographical area described in their CoC applications.

In addition to prevention, outreach and assessment, emergency shelter, transitional housing, supportive services, permanent supportive housing, and permanent affordable housing, low/no demand programs play a key role in the provision of homeless services for many communities, and will probably play an even greater part in the future. These alternative approaches complement outreach, emergency shelter, and permanent supportive housing efforts, providing a haven for those whose needs may not be addressed by other CoC components. Finally, respondents consistently said that the principal challenge facing their community in preventing and eradicating homelessness continues to be centered on the lack of permanent affordable housing.

Navigating the Homeless System

Each community uses its own unique approach and method to deliver services to homeless persons. Entry into homeless assistance networks may be broadly classified as:

  • Fragmented—homeless people may directly approach any provider in the network, may (or may not) gain entry, and may or may not get connected to other programs and services. About three-quarters of our 25 communities would fall into this category, although for four of these only individuals face this fragmentation because family intake is centralized.
  • "No wrong door"—homeless people gain access by approaching any program, after which program staff augment these first contacts with shared knowledge of what is available and systematic linkages that help clients get to the right programs and services. A few of our communities fall into this category, and a few more would do so if we expanded our concern to include an integrated approach to assessment and service delivery through multi-service centers, in addition to intake.
  • Centralized—one or a few linked points of entry. According to proponents, centralized entry minimizes prolonged and misdirected searches for emergency shelter and services, and allows for uniform intake and assessment, which helps ensure equity of access to services. Five of our communities have centralized family intake, while one has a centralized intake mechanism in place for everyone.

This summary of intake approaches in these very high scoring CoCs indicates that many communities still have not streamlined homeless people's access to programs and services. Families are more likely than single individuals to benefit from a centralized or "no wrong door" approach, but even they do not have such access in most of this study's communities. The most common situation is that both families and single individuals must rely on the case managers at their particular program to help them gain access to the services they need.

Based on access to resources, local philosophies and priorities, and community need, each of the 25 continuums faces challenges in serving certain sub-populations. Each community identified at least one of the following groups as being hard to serve with existing resources—chronically homeless persons with mental illnesses and/or substance abuse problems, youth, large families and/or families with teenage sons, and ex-offenders.

Involvement of Mainstream Agencies and Services

HUD urges communities to take maximum advantage of mainstream services in responding to the needs of homeless people—not just for planning, but for coordination of services, and supportive services to accompany the housing components. Mainstream programs commonly mentioned in this regard include public housing and vouchering programs, food stamps, SSI, SSDI, general assistance, TANF, job training, health care, mental health care, substance abuse treatment, and veteran programs.

Despite recognizing the importance of mainstream services, relatively few of the 25 communities in this study were successful in fully integrating mainstream agencies and systems whose clients include but are not limited to homeless people. On the other hand, a number of the 25 study communities have taken actions to integrate mainstream services and systems into their homeless assistance system. They have used several different strategies, singly or together, which have achieved varying levels of mainstream agency involvement. These strategies, all of which the CoC approach has encouraged and sometimes funded, include:

  • Having staff with the responsibility to promote systems/service integration,
  • Creating a local interagency coordinating body,
  • Having a centralized authority for the homeless assistance system,
  • Co-locating mainstream services within homeless-specific agencies and programs, and
  • Adopting and using an interagency management information system (MIS).

Communities that recognize that mainstream agencies need to be seriously involved in broader-scale planning and coordination efforts and spared the details of the CoC application process have experienced greater success. Key components to successful integration are strong leadership in the homeless assistance system and a commitment from both mainstream agency leadership and homeless-specific program and service providers to work together.

CoC Planning for and Coordination of Homeless Services

Some communities in this study engaged in extensive planning for and coordination of their homeless assistance programs and services, while others did only what was necessary to get HUD CoC funding.

With respect to the CoC application, HUD requires that communities present a "gaps analysis" showing the gap between "need" and "current inventory" of programs and services to meet that need, and then propose projects for HUD funding that meet at least part of the identified need. The 25 communities in this study usually divided this work into several distinct components. First, a few people did the preliminary work needed to gather the data to determine need and current inventory for the gaps analysis. Second, there was usually broad participation in meetings to accept the gaps analysis statistics and assign high, medium, and low priorities to identified gaps. Finally, agencies developed proposals for submission to HUD and a committee ranked all proposals for the final HUD application. Participation in this final ranking committee was often less inclusive than participation in the gaps analysis process. Many communities take great care that the ranking process uses clear criteria and is fair to all applicants. To this end many exclude providers with current HUD funding, or those applying for HUD funding, from the ranking process, and also try to include representatives from business, government, and local funders to supply a broad community viewpoint.

The simplest level of planning, pursued by about one-third of the 25 communities studied, focused primarily on the CoC application and the steps necessary to complete it. These communities paid most attention to the types of projects that HUD CoC funds could support (transitional and permanent supportive housing and supportive services), and gave less attention to the larger spectrum of homeless assistance from prevention and outreach to affordable housing. They also tended to concentrate their efforts in the few months leading up to submission of the HUD application, although a few did year-round planning.

Another one-third of the communities in this study operated at a second level of planning, which they pursued year-round planning with a focus on the larger system of homeless programs and services, including the integration of at least some mainstream programs and services. The final group of communities engaged in multi-year, strategic planning for homeless programs and services well-integrated with mainstream services, and usually with a goal of ending homelessness.

One conclusion from this research is that limited planning produces limited results. By focusing on three key areas—broadening the scope of planning, incorporating additional funding sources, and increasing the participation of mainstream agencies—local CoCs will be better able to meet the constantly changing needs of homeless persons and increase their ability to be flexible in the years to come. Planning and coordination of programs and services that is part-year or limited to CoC funding, while meeting HUD's requirements, does not benefit from coordinated funding and long-term planning. Likewise, year-round planning and coordination that only include limited participation by mainstream providers or do not include multi-year planning efforts, do not reap the same rewards as multi-year, comprehensive, strategic planning. Comprehensive, broad-based planning encompassing mainstream services and multiple funding sources benefits from sharing the burden of solving homelessness among many players and relying on having access to a broader array and higher level of resources.

Broader planning is becoming especially important as communities face pressures to develop new programs and services while maintaining what they have. Many of the communities in this study needed their entire allocation of CoC funding just to maintain existing programs. This situation is being referred to as the "renewal burden," and it is having a profound impact on the flexibility of CoC systems in all communities in this study. HUD could promote additional flexibility with administrative changes related to renewals. For example, HUD could elect to hold CoCs harmless for generating mainstream support for services. That is, any amounts of HUD funding currently going for services (as opposed to housing) that a community could replace with local resources would stay in the community and be available for alternative uses within the CoC. This would both allow communities to broaden their base of financial support for planning, coordination, and service delivery, as well as increase the role of mainstream providers.

How the CoC Application Process Has Shaped Planning and Provision of Services

When HUD introduced the idea of the continuum of care in several stages from 1993 through 1995, the new requirements were imposed on existing local relationships and power dynamics. The relationships, planning structures, and coordination linkages that existed before the CoC funding approach, or the lack of them, have had a major impact on current homeless systems as well as on the applications to HUD.

We identified three categories to describe the transitions in the 25 communities:

  • Seamless transitions—7 communities had already established a formal, system-wide planning process for homelessness, and therefore experienced few changes in the way they plan for homeless programs and services as a consequence of CoC requirements.
  • From informal/fragmented to structured/comprehensive—10 CoCs had done at least some system-wide planning for homelessness before the HUD CoC requirements took effect.
  • From nothing to something—the CoC application process has had the greatest impact on 8 communities that previously had little or no system-wide planning process in place for homeless programs and services. For the first time, these communities brought together providers, advocates, representatives from local government, and other community members who had previously been working on the issue independently, or with minimal collaboration.

As expected, the level of planning that existed before HUD introduced the current CoC funding process has had a great deal to do with subsequent planning for homelessness as well as the application to HUD. Specifically, communities that did significant, system-wide planning for homelessness prior to the HUD CoC requirements have not only had an easier time meeting those requirements, but also tend to have more comprehensive homeless programs and services, more stable planning structures, and more participants satisfied with the planning process and its outcomes. Furthermore, those jurisdictions with stable leadership and firmly entrenched planning structures tend to be more successful in establishing comprehensive, system-wide planning for the application and in providing a reasonably well-coordinated and full continuum of programs and services.

Implementing Data Systems: A Few CoCs Far Ahead of the Rest

Ideally, every community receiving CoC funding would have in place a system that routinely collected data about homeless people and services on an ongoing basis. While the majority of communities have individual providers collecting data, they have not progressed very far toward developing and implementing a comprehensive Homeless Management Information System (HMIS). At this stage in the ongoing development of CoC data systems, there is a long way to go before most communities reach this ideal. This is not due to lack of recognition of the potential benefits of HMIS, but is due to significant barriers including:

  • Lack of power to enforce data gathering in cases where the lead agency for the CoC application does not control CoC funds and does not enter into contracts with homeless providers;
  • Cost of implementing and operating an HMIS, such as software and hardware purchase and installation, initial training of current homeless provider staff, the salary of a data systems manager, software and hardware upgrades and installation, subsequent training of new homeless provider staff, and ongoing technical assistance as problems arise or specific accommodations are needed;
  • Lack of paid or volunteer staff time to spend on implementing and running an HMIS system;
  • Concerns about confidentiality of client data;
  • Compatibility of different data systems within the homeless service provider network and among mainstream providers; and
  • Sharing case management information electronically is a concern for time, confidentiality, and adequacy reasons.

While several of the CoCs we visited have overcome these barriers, stakeholders in many other communities insist that they are insurmountable. Information sharing among different communities at different stages of implementing HMIS may provide the most convincing argument for those communities facing the most barriers to HMIS development.

Of the 8 communities with the most advanced data systems, 5 have a CoC lead agency that not only is charged with developing the CoC application, but also becomes the grantee, receiving CoC funding directly from HUD. This allows the lead agency to enforce data collection through the contracts it writes with homeless providers.

We also found little evidence of well-developed performance measurement systems among the 25 communities visited, with the exception of three communities that were also well advanced in the systematic collection of other information.

Implications for Research

This study was not able to give sufficient attention to issues in truly rural communities (those beyond easy access to a major metropolitan area). Nor did it examine approaches used by states to promote CoCs throughout their boundaries. Issues that would be of interest in states that have developed regional or statewide CoCs include their strategies for identifying appropriate geographical boundaries for regions, strategies for organizing regions, and what happens when regional and state levels of government are added to the CoC development and HUD application process. Some states have been very active in creating, motivating, and supporting regional applicants with technical and other assistance, but we do not know much about how they do it, the decisions they face, and the ways they resolve them. These issues would make a good focus for a follow-up research project similar to the present one.

In addition, many questions focus on who needs transitional housing, in what form, and for how long? Likewise, for whom is transitional housing a disservice and placement in permanent units with supportive services to be favored? Research should be undertaken to address these issues.

Finally, it would be good to learn much more about the characteristics of permanent supportive housing programs that work as safe haven/low barrier approaches, and to identify program characteristics that may need to be different for substance abusers, people with severe mental illnesses, and dually- and triply-diagnosed people.

Implications for Practice

  • Based on our impressions of the 25 CoCs visited for this study, several steps, stages, or levels, some more sophisticated than others, describe the various stages in which communities may find themselves in developing CoCs and in addressing the ultimate goal of ending or reducing homelessness:
    • Step 1: Community planning, and perhaps also coordination, activities cover only the homeless-specific programs and services that HUD will fund through the CoC application. Little if any participation occurs from non-homeless programs, or even from emergency shelter and prevention providers.

      Step 2: Community planning and coordination extends to all homeless-specific programs and services, including emergency shelter, outreach, and prevention, incorporating planning for Emergency Shelter Grant monies and perhaps also Emergency Food and Shelter Program funds in the same process.

      Step 3: Significant involvement of mainstream agencies for service delivery to people involved with homeless-specific programs and services. These mainstream representatives may include agencies with responsibilities for services related to alcohol, drug, and mental health problems; public health; employment and training; public schools; welfare and cash benefits; public housing; and Veteran Affairs. They may be public or private nonprofit agencies, but usually are public, or at least publicly funded.

      Step 4: Once communities achieve Step 3, which is no small feat, and experience its benefits for several years, they may realize that just as many people are coming into the system as ever. They may then rethink their approach. When they do so, the results are usually an even greater involvement of mainstream agencies, this time geared to prevention and expanding the availability of affordable housing. The prevention aspect usually focuses on mental health, substance abuse, and even corrections agencies doing a better job of assuring that people released from their facilities have stable housing and any supportive services that are needed to prevent them from ending up homeless. In addition, resources to avert eviction, pay rent or utility arrearages, do some financial counseling, and otherwise handle situations of imminent homelessness, usually for families, often play a significant role on the prevention side. The affordable housing aspect means doing something serious about the often impossibly tight, and expensive, housing market facing homeless and other very poor people, and requires a very long-term vision and commitment.

      • Communities will not be likely to progress much beyond Steps 1 and 2 without the support of mainstream agencies, local (and sometimes state) governments, and the local business community and other power brokers.
      • Local, state, and regional attitudes toward public responsibilities also make a huge difference, as does the basic level of resources. Our 25 communities are located in states that (1) have many resources and are committed to use them, (2) have (or could have) reasonable levels of resources but are opposed to both taxation and public action, and (3) have few resources but are willing to use whatever they have. It should be clear that communities in states of the second sort will get no help from outside, while those in states of the first and even the third sort have distinct advantages.
      • Even with strong mainstream agency support and the local power structure on board, it makes a huge difference what one is trying to do. A few of our communities are very well organized and have strong local support, but focus their energies primarily on helping people who are already homeless. A few others have spent a number of years in that situation and have now begun to move toward stopping the flow into homelessness, creating permanent supportive housing for the chronically homeless people in their community, and addressing the desperate need to create more permanent affordable housing for all very poor people.
      • The ability of no/low barrier permanent supportive housing programs to create stable housing environments for chronically homeless people taken right off the streets suggests that policymakers and practitioners should be paying more attention to learning how this works where it succeeds, and worry less about strict adherence to theories of rehabilitation and recovery that may have the effect of excluding significant numbers of homeless people from programs and services.

    Implications for Policy

  • In every community we visited, respondents cited the severe lack of affordable housing as affecting both the level of homelessness and the ability of programs to move people from shelter back into permanent housing. Problems cited included rental vacancy rates of 1 percent or less, average rents 15 to 25 percent above Fair Market Rents, a lack of apartments accepting Section 8 vouchers, zoning and other legal restrictions on SRO and other low-income housing developments, and destruction of public housing units. Respondents were near-unanimous in feeling that their community will not be successful in ending homelessness until it can produce more housing and make it affordable to very poor people, including single people with disabilities.


  • Whatever research ultimately shows about the utility of transitional housing, respondents brought up many issues with regard to the inconsistencies in the length of services allowable for people in different configurations of transitional housing. These could be anywhere from 30 months at the most to less than 12 months, depending on how long clients live in a transitional housing facility before leaving. It was suggested that, to eliminate current inequities, HUD might allow programs to offer clients up to 24 months of transitional supportive services, regardless of where they live.


  • Effects of the CoC funding structure and implications for policy.


  • There was little question among respondents in most of the communities in this study that HUD's CoC funding structure has moved them toward significantly greater planning. Further, that planning has come over the years to encompass a broader scope, and in many communities to involve more players. Where community respondents felt the CoC structure had not made much difference, they were already very well organized before 1996 and tried to cover the range of homeless assistance from prevention to permanent supportive housing.


  • In addition to planning, the CoC process has resulted in significantly higher levels of shared knowledge about what is available in the community, more program and service coordination, better referral networks, and development of new joint projects. It has also produced a good deal of cross-jurisdiction cooperation that never happened before, or never happened with respect to homelessness. Eligible geographical areas such as smaller cities and counties have been drawn into larger planning and service systems, and are not either "going it alone" or doing nothing about homelessness.


  • We conclude that HUD's adoption in 1996 of a community-wide approach to distributing its competitive homeless assistance dollars has moved communities further in the direction of broad planning and program development than would have happened without the CoC approach. The ensuing networks of programs and services have been able to offer more support to homeless people, with more cohesion, than would otherwise have been possible. Of course there are never enough programs and services to meet all needs, but the communities in this study come closer to doing so in part as a result of the CoC approach. The mechanism that has brought communities together to accomplish this result should be maintained and strengthened.
  • This report is available in its entirety in the Portable Document Format (PDF), which many find convenient when printing.



    Topics/Tags: | Cities and Neighborhoods | Housing | Nonprofits


    Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

    If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

    Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

    Email this Page