New Data On The Low-Income Housing Tax Credit

The Low-Income Housing Tax Credit (LIHTC) dispenses more than the equivalent of $3 billion in annual budget authority to subsidize the construction or rehabilitation of housing for low-income families. Support is given to private investors in the form of tax credits that are cashed in over a 10-year period. In exchange, rents for these units must be maintained at affordable levels. The program can be combined with numerous other Federal, State, and local housing subsidies as well as private funds to expand the supply of affordable rental housing. Because of the considerable resources provided, the LIHTC is one of the Federal Government's most important housing programs for low-income families.

Under the supervision of the U.S. Internal Revenue Service, 54 State and local housing finance agencies administer the LIHTC program. These agencies independently allocate tax credits within their jurisdictions and perform various monitoring activities. A decentralized administration makes the LIHTC program responsive to State and local housing needs but also means that little aggregate information is collected about the LIHTC. Because no Federal or State agency has collected comprehensive information on the LIHTC, there are no complete and reliable sources of data available to those individuals wishing to study it. Consequently, many unanswered questions about the LIHTC exist, such as:

• Who resides in tax-credit projects?
• What are their incomes?
• What are their ages?
• What is the racial makeup of LIHTC project tenants?
• How big are LIHTC projects?
• Where are they located?

One of the barriers to analyzing LIHTC data has been the lack of a sampling frame or list of all projects receiving tax credits. Not knowing the total number and basic characteristics of the LIHTC projects in existence precluded designing and drawing a random sample from which to draw statistically valid inferences. To remedy this situation, the U.S. Department of Housing and Urban Development (HUD) contracted with Abt Associates Inc.¹ to collect data on tax-credit projects ready for occupancy (placed in service) between 1990 and 1994. The data were collected by surveying the 54 State and local housing finance agencies.

The LIHTC data collection endeavored to create a national sampling frame of tax-credit projects that can be used to answer a wide range of questions of interest to national, State, and local policymakers, investors, housing advocates, and researchers. This article describes the database and discusses some preliminary analysis done by Abt Associates Inc. and HUD.

The Database

The database contains basic information on 9,785 projects with 339,190 units. The information includes the following:

• Project name and address.
• Name and address of the owner or owner's representative.
• Total number of units.
• Number of low-income units.
• Total number of units by size.
• Year the project was placed in service.
• Year the project received a tax-credit allocation or had tax-exempt bonds issued.
• Type of structure (new construction, rehabilitation, or existing building).
• Credit percentage used (9 percent or 4 percent).
• Nonprofit sponsorship.
• Increase in basis due to location in Qualified Census Tract or Difficult Development Area.
• Use of tax-exempt bonds.
• Use of Rural Housing Service (formerly Farmers Home Administration) Section 515 loans earmarked exclusively for rural housing.

The most complete coverage of LIHTC units and projects placed in service is for the period 1992-94.² For 1992 all 54 State housing agencies submitted the requested data. For 1993 and 1994, all agencies except the Chicago agency reported information. For 1990 and 1991, 46 and 47 agencies, respectively, contributed data. For 1987, 1988, and 1989, the number responding declined to 31, 32, and 34 agencies, respectively.

Each project in the database was geocoded using its address to assign latitude and longitude coordinates and census tract identifiers. Overall, 76 percent of the properties and 78 percent of the units were successfully geocoded. Data displays the location of LIHTC projects in the 48 contiguous States for the years 1987 through 1994. Using geographic information system software, market analysts can use the geocoded location information to provide a snapshot view of the distribution of LIHTC projects in their areas. Geocoding of projects also permitted the inclusion of census tract information in the database, such as population, population density, racial make-up, median income, Qualified Census Tract/Difficult Development Area status, and fair-market rent.

Basic Facts

From 1990 through 1994, an average of nearly 49,500 total units per year were placed in service in connection with the LIHTC program. Of these an average of 44,900 units each year were rent restricted for low-income households. In total the LIHTC added more than 247,000 total units to the rental housing stock, of which 224,446 were low-income units.

Data demonstrates the importance of the LIHTC program by reporting LIHTC units placed in service by their year of completion, the number of rental apartments completed each year as reported by the Survey of Market Absorption, and LIHTC units as a proportion of all new rental units. From 1990 through 1994, more than 36 percent of all rental apartment units completed have been in LIHTC projects. Total completions fell from 1990 through 1993 but rose slightly in 1994, while the number of LIHTC units increased during this period. LIHTC-supported low-income units made up 54 percent of total units completed in 1994.

Data summarizes some of the most interesting findings contained in the database. An average of 1,228 projects were placed in service annually during the period 1990 to 1994. The vast majority of LIHTC projects were relatively small. More than two-thirds of LIHTC projects placed in service during this time contained 50 units or fewer, and more than 20 percent had fewer than 10 units. The average LIHTC project placed in service during this period contained 40.6 units.

86 percent of LIHTC units placed in service between 1990 and 1994 contained 2 bedrooms or fewer. The American Housing Survey reports that 62.4 percent of the total rental units built in the years 1990-93 contained 2 bedrooms or fewer, and 28 percent were 3-bedroom units. Thus, the typical LIHTC unit is smaller than its counterpart in the broader rental housing market.

As noted above, data shows the geographic distribution of LIHTC projects. Within the 4 census regions, 43.3 percent of LIHTC units are located in the South, 27.0 percent in the Midwest, 12.5 percent in the Northeast, and 17.2 percent in the West.

More LIHTC units are located inside metropolitan central cities than outside. About 53.6 percent of LIHTC units are in central cities, 26.6 percent are in suburbs, and 19.8 percent are in nonmetropolitan areas. The sizable rural component of the tax-credit program is probably linked to the 26.8 percent of LIHTC units financed through Rural Housing Service Section 515. Therefore, it is fair to characterize the LIHTC as a significant source of affordable housing in both rural and urban America.

To qualify for tax credits, projects must set aside a specific proportion of units for lower-income households. Owners may elect to set aside at least 20 percent of the project's units for households at or below 50 percent of the area's median income or at least 40 percent of the project's units for households with incomes less than 60 percent of the area's median income. Rents in set-aside units are limited to no more than 30 percent of the elected 50- or 60-percent income limit. Thus, owners of LIHTC projects have some discretion over the number of low-income tenants. Between 1990 and 1994, nearly 91 percent of units in LIHTC-supported projects were set aside for low-income families. LIHTC projects placed in service during that period were composed almost entirely of rent-restricted units reserved for low-income use. In fact, only about 5 percent of units are in projects where LIHTC units compose 40 percent or fewer of the total units. Almost 90 percent of LIHTC projects have at least 80 percent of their units receiving LIHTC assistance.

When potential maximum rents for LIHTC projects are compared with HUD's local area Section 8 Fair Market Rents (FMRs) based on the rent of the 45th-percentile market-rate unit, 76.5 percent of LIHTC units would have a higher maximum rent than the local FMR under the 60 percent of median-income standard. Under the 50 percent of median-income standard, 40.9 percent of units would have maximum rents higher than the FMR.

In establishing the tax-credit program, Congress required that 10 percent of each State's LIHTC dollar allocation be set aside for projects with nonprofit sponsors. The percentage of units with nonprofit sponsors rose from 18 percent in 1992 to 24 percent in 1993 to 27 percent in 1994, for an average of 23 percent across the 3 years. Nonprofit organizations are increasingly using the tax credit in their efforts to help supply affordable housing.

Database To Support More Research

Although the database contains useful information on some important issues related to LIHTC, more research will be needed to answer many other important questions. The major purpose of the LIHTC data collection is to provide a sampling frame that will enable additional research on LIHTC by both HUD and outside researchers. HUD has made the LIHTC database and the accompanying report available to the general public on the Internet at http://www.huduser.gov/datasets/lihtc.html, and HUD will periodically update the database.

LIHTC data collection and locational analysis are part of a broader effort that is under way at HUD to gather and make available data on all major housing and urban economic development programs. In cooperation with the research community, HUD seeks to better understand these programs to improve program efficiency and effectiveness.

Notes

1. Abt Associates Inc. prepared a report, Development and Analysis of the National Low-Income Housing Tax Credit, summarizing the database and many preliminary findings. This information is available to the public on the Internet: http://lihtc.huduser.gov/

2. The General Accounting Office (GAO) is conducting a similar study of the LIHTC focusing on the 1992 to 1994 period. There are slight differences between HUD's database and GAO's for this period.