ResearchWorks July 2010

 Homelessness Costs and Interventions

Three new studies on homelessness in the United States that examine the cost of first-time homelessness, life after transitional housing for homeless families, and strategies for improving access to mainstream benefits and services are available from HUD's Office of Policy Development and Research.

Costs Associated with First-Time Homelessness for Families and Individuals examines the costs of housing and serving nearly 9,000 individuals and families in six areas of the country Des Moines, Iowa; Houston, Texas; Jacksonville, Florida; Washington, DC; Houston, Texas; Kalamazoo, Michigan; and a large area of upstate South Carolina. Overall, the study shows that the experience of homelessness is diverse and the associated costs vary, depending on the pattern of homelessness, household type, and type of response. This comprehensive research associated with first-time homelessness creates a foundation for comparing the costs of various homeless interventions and establishes that:

• For homeless individuals, emergency shelter is typically the least expensive response and transitional housing is the most expensive.
• For homeless families, emergency shelters and transitional housing programs were equally expensive, usually due to the amount of services families receive in both program models and the higher cost of providing families with accommodations that have a greater degree of privacy than individuals require.
• Permanent supportive housing for both individuals and families is less expensive to the homeless assistance response system, as service costs are borne by other systems, such as Food Stamp or Temporary Assistance for Needy Families programs.

A second study, Strategies for Improving People's Access to Mainstream Benefits and Services, documents how seven different communities (Albany/Albany County, New York; Albuquerque, New Mexico; Metropolitan Denver; Miami-Dade County, Florida; Norfolk, Virginia; Portland, Maine; and Pittsburgh/Allegheny County, Pennsylvania) mobilized to improve homeless people's access to mainstream benefits and services such as Food Stamps, Temporary Assistance for Needy Families, and Medicaid. Results were mixed:

• Communities with the greatest success had a strong central organization intent on improving access of homeless persons to mainstream services.
• Communities were usually able to reduce structural barriers to benefits (such as physical access, complexity and length of application processes) and rules for documenting eligibility.
• Communities were less successful in overcoming barriers beyond their control, such as eligibility requirements for various programs and limited capacity of mainstream service providers.
• Communities have developed innovative ways to overcome barriers to mainstream benefits, but some barriers can only be resolved with state or federal involvement.

A third study, Life after Transitional Housing for Homeless Families, followed 195 families in transitional housing programs across five communities (Cleveland/Cuyahoga County, Ohio; Detroit, Michigan; Houston and Harris and Benton Counties, Texas; San Diego City and County, California; and Seattle/King County, Washington) for 3, 6, and 12 months after leaving the program in an effort to document the impacts of participation in a transitional housing program. The study looks at housing status, employment, and education outcomes associated with the service-intensive transitional housing programs for families with children. The study found:

• Individuals benefited from educational and employment opportunities that help change life circumstances.
• Children benefited from having fewer moves and school changes.
• Families leaving transitional housing moved to their own place, and 60 percent remained in their homes 12 months later.
• No relationship was established between the number of barriers to stability that a family faces, the length of stay in transitional housing, and the outcomes of the stay.
• Families with relatively few challenges remained in transitional housing for long durations and may be using such assistance while waiting for subsidized housing to become available.

The goal of homeless assistance is to create a system of response that is mindful of the many reasons why an individual or family might find themselves at the front door of an emergency shelter. A good response system includes development of effective interventions that provide the necessary amounts and types of housing services to stabilize people, while ensuring that they will be able to obtain and retain housing in the future. At the same time, limited resources for housing and services must be allocated efficiently.

The studies highlighted above can inform the design of a responsive homeless assistance system. For some, supportive services will be critical to ensuring future housing stability. For many, housing assistance is all that is needed to exit the homeless assistance system. By properly targeting the most intensive and highest cost interventions to those most likely to benefit from them and freeing resources needed for other lower cost interventions, such as housing subsidies, families with the greatest need can access transitional housing.

BACK TO THE TOP

 A Portrait of Homelessness in 2009

On a single night in January 2009, more than 643,000 people were homeless in the U.S. Approximately two-thirds were in shelters and over one-third spent the night without refuge on the streets, in abandoned buildings, or in other places not intended to be inhabited. Sixty-three percent of homeless persons on that January night were individuals; the rest were members of homeless families.

To form a portrait of homelessness, this point-in-time count is coupled with data from a 12-month period spanning October 2008 through September 2009, when nearly 1.6 million individuals spent at least one night in emergency or transitional housing. Most (77%) of these homeless persons used emergency shelters, while others used transitional housing, and a small percentage used both. This sheltered homeless population was primarily located in principal cities (68%), with the rest residing in suburban or rural areas. Thirty-nine percent of the total homeless population resided in three states that account for only 25 percent of the national population: California, New York, and Florida, this concentration likely due to high housing costs.

Two excerpts of data provide a sense of who comprised the sheltered homeless population. One described the typical homeless person: a middle-aged, adult male who was a member of a minority group and was by himself. More precisely, of these homeless persons:

• 78% were adults;
• 61% were male;
• 62% were members of a minority group;
• 38% were 31 to 50 years old;
• 64% were lone individuals; and
• 38% had a disability.

However, accounts from local service providers indicated that attributes of the homeless varied across the country. Relative to the national sheltered homeless population, young, white women (between 18 and 30) were more common in Seaside and Monterey Counties on the central California coast. In Idaho, the population of homeless individuals was mostly comprised of young, white men. Homeless seniors were increasingly common in Detroit, which also had a significant proportion of homeless teens.

A second way to describe the homeless is by household type, as shown in this chart.

These data are from the most recent overview of homelessness, presented by HUD in The 2009 Annual Homeless Assessment Report to Congress. Point-in-time snapshots accounted for both sheltered and unsheltered homeless persons on a single night, usually at the end of January, and revealed trends about the seasonality of homelessness. The second source was over-time counts taken from the Homeless Management Information System (HMIS) that provided more detailed information on shelter users over the course of a full year.

Although point-in-time estimates are not comparable (due to dramatic changes within cities, inclement weather, adjustments in localities' reporting methods, etc.), the HMIS data on individuals and families is useful in detecting emerging trends in the sheltered homeless population. For example, between 2007 and 2009, the number of homeless individuals declined by 7 percent and the number of homeless families swelled by 30 percent. The aging of the adult homeless population is also brought into focus with longitudinal data, as is a steady decrease (from a high starting point) in the percentage of sheltered homeless African American families and individuals.

Looking forward, future annual assessments will illuminate patterns of homelessness and long-term impacts of the recession on this population. The 2010 study will add another year of data to the national perspective of homelessness, and will be the first to report on utilization of permanent supportive housing and trends among homeless veterans. The Homelessness Pulse Projectreports based on data from nine communities contributing to the annual reports will continue to highlight changes that provide timely information on local economic conditions and homelessness.

Read the National Strategic Plan to Prevent and End Homelessness

BACK TO THE TOP

 LIHTCs Boost Affordable Rental Housing Supply

Over 1.4 million rental units entered service between 1995 and 2007 with the help of Low-Income Housing Tax Credits (LIHTCs). With the Tax Reform Act of 1986, Congress created LIHTCs to incentivize the private market to invest in affordable rental housing. Each year, the Internal Revenue Service allocates low income housing tax credits to state agencies usually state housing finance agencies that award the credits to affordable housing developers, giving priority to projects that serve the lowest income families and remain affordable for the longest periods.

From 1995 through 2007, developers used LIHTCs to raise capital for the construction and rehabilitation of 18,865 affordable rental housing projects nationwide. Investors who purchase these tax credits receive dollar-for-dollar federal tax credits annually for a period of 10 years. The purchases help reduce the amount of money a developer must borrow to finance a construction or rehabilitation project and result in lower, more affordable rents for low-income families. Most (82%) of these rental properties are composed entirely of low-income units, although a mix of affordable and market units is permitted. Rents and utilities for low-income units are restricted for a minimum of 30 years and adhere to one of two low-income occupancy threshold requirements:

• The 20-50 Rule - At least 20 percent of the units must be rent-restricted and occupied by households with income at or below 50 percent of area median income; or
• The 40-60 Rule - At least 40 percent of the units must be rent-restricted and occupied by households with incomes at or below 60 percent of area median income.

The amount of tax credits available to a project depends on the development cost (excluding land), the proportion of affordable units set aside, and the credit rate (which varies based on the development method and whether other federal subsidies are used). Credits provide benefits with a present value equal to either 30 or 70 percent of a property's qualifying basis. The 30-percent tax credit is for acquisition or for federally financed rehabilitation and new construction. The 70-percent tax credit is for nonfederally financed rehabilitation or construction.

In 1989, Congress added provisions to the program to encourage production of affordable housing units in hard-to-serve areas. Specifically, the act permits projects located in Difficult Development Areas (DDAs) or Qualified Census Tracts (QCTs) to claim 30 percent more in tax credits than identical projects outside of these areas. Designated by HUD, DDAs are metropolitan or nonmetropolitan areas in which construction, land, and utility costs are high relative to incomes; QCTs are census tracts in which at least half of the households have incomes that are less than 60 percent of the area median income or have a poverty rate of at least 25 percent.

Although HUD does not administer this subsidy program for the construction and rehabilitation of low-income rental housing, the Office of Policy Development and Research compiles and tracks information about LIHTC projects and their impact on the nation's rental housing supply. New units put into service in 2007 are in the most recent update of the database, available to the public at https://www.huduser.gov/portal/datasets/lihtc.html.

Also, see U.S. Rental Housing Characteristics: Supply, Vacancy, and Affordability

BACK TO THE TOP

 Snapshot of Worst Case Housing Needs in the U.S.

Since 1991, the U.S. Department of Housing and Urban Development's Office of Policy Development and Research (PD&R) has identified households experiencing "worst case needs" across the United States. PD&R's most recent biennial report, Worst Case Housing Needs 2007: A Report to Congress, uses American Housing Survey data to highlight the demographics of families experiencing worst case needs, the availability of affordable housing stock to these families, and the poverty levels and residential patterns of neighborhoods where these renters live.

A family experiences worst case needs if it does not receive housing assistance and meets two criteria. First, the family must have very-low income, earning less than 50 percent of the area medium income (AMI), or extremely low income, earning less than 30 percent of AMI. Second, the family must have a severe rent burden (paying more than 50 percent of the family's income towards rent) and/or live in severely inadequate conditions with regard to plumbing, heating, electrical, or upkeep.

Nearly 6 million households (5.91) experienced worst case needs in 2007. This is an 18 percent increase from 2001 when only 5.01 million households faced this difficulty. Of these households, the report highlighted the following:

• 93 percent have severe rent burdens (5.48 million households), the primary cause of worst case needs.
• 73 percent have extremely low incomes.
• 37 percent were families with children, 20 percent were elderly, 10 percent were non-elderly disabled, and 32 percent were "other."
• Almost half (46%) of households with children had full-time employment.
• 49 percent were non-Hispanic white; 21 percent were Hispanic; and 23 percent were non- Hispanic black.

In other findings, worst case needs households were more likely than the larger universe of renters to live in poorer neighborhoods; were far more concentrated in higher poverty neighborhoods when they lived in central cities; and were less likely to live in high poverty neighborhoods when located in suburban and non-metropolitan areas, but were unable to access affordable housing. The study found that the availability of housing stock across the nation is insufficient for the lowest income groups. For every 100 extremely low income households, there are only 76 affordable rental units available (those costing 30% or less of a household's income). This lack of affordable and available rental units and severe rent burdens are the largest barriers to families experiencing worst case housing needs.

Worst Case Housing Needs 2007: A Report to Congress, can be found at https://www.huduser.gov/portal/publications/affhsg/wc_HsgNeeds07.html.

Visit the American Housing Survey website

BACK TO THE TOP