Low-income housing credit

Originally published in the NTA Encyclopedia of Taxation and Tax Policy, Second Edition, edited by Joseph J. Cordes, Robert D. Ebel, and Jane G. Gravelle. The encyclopedia is available in paperback from the Urban Institute Press. Order online at www.uipress.org or call toll-free 1-877-847-7377

Leonard E. Burman
The Urban Institute, The Tax Policy Center, and Georgetown University
Alastair McFarlane
Department of Housing and Urban Development

Designed to encourage the acquisition, construction, and rehabilitation of housing for low-income families, this federal tax credit, which is criticized for its inefficiency and complexity, is a major source of low-income housing support.

Enacted as part of the Tax Reform Act of 1986, the credit replaced a variety of incentives for investment in low-income housing with a tax subsidy aimed more directly at lowerincome households. The original low-income housing credit expired in 1989, but it was extended three times before the Omnibus Budget Reconciliation Act of 1993 made it permanent.

Overview

Under Section 42 of the Internal Revenue Code, qualifying individuals and corporations can claim a tax credit over a 10-year period equal to a percentage of the depreciable costs (excluding land) incurred to provide low-income housing. For new construction and costs of renovation on qualified housing that does not receive additional federal subsidies, the credits may have a present value of up to 70 percent of the depreciable basis in low-income units. A reduced credit with a present value of up to 30 percent is available for housing with other federal subsidies, such as tax-exempt bond financing, and for the cost of purchasing existing housing that is renovated. The corresponding maximum annual credit percentages were 8.07 percent and 3.46 percent, respectively, in August 2004. The Internal Revenue Service (IRS) sets these credit percentages every month based on current interest rates; they tend to remain near 8.2 percent for the larger credit and 3.5 percent for the reduced credit. The tax credit received depends on the proportion of units set aside for low-income tenants. Additionally, developers are eligible for a further 30 percent of the tax credit amount if the project is located within a low-income or high construction cost area.

To qualify for the credit, either 20 percent of units in a housing project must be rented to tenants with incomes below 50 percent of the area’s median income, adjusted for family size, or 40 percent of units must be rented to tenants with incomes below 60 percent of the median income. The rent on a low-income unit is limited to 30 percent of the qualifying income level, assuming a family size equal to 1.5 times the number of bedrooms in the unit. The qualifying income level is 50 percent of the area median family income for projects that meet the "20-50" threshold or, more commonly, 60 percent for those that meet the "40-60" threshold. HUD’s estimated median area income for a family of four in FY 2004 was $57,400. For such a family, the qualifying income limit would be either $34,450 (60 percent standard) or $28,700 (50 percent standard), and the restricted rents would be $861 or $718, respectively. The qualifying income (and thus the rent limit) is adjusted for family size.

Past tax credits are recaptured with interest if the project fails to comply with the rent limits and set-aside requirements during the first 15 years. The percentage of credits recaptured phases out in the 11th through the 15th years. In addition, project owners must agree to provide low-income units for at least 30 years, although owners may terminate this commitment after 15 years under certain conditions.

Cost and allocation of the credits

State housing authorities allocate credits for housing projects. Credit allocations for each project are supposed to be limited so that investors and project organizers do not reap excessive profits. In 2004, each state was allowed to allocate up to $1.80 per capita for projects that were not financed with tax-exempt bonds, with a minimum of slightly more than $2 million per state. The per capita limit was indexed to inflation in 2003. Total credits allocated for bond-financed projects are only limited by states’ authority to issue tax-exempt bonds. For a more detailed description, see Guggenheim (2003).

The Joint Committee on Taxation projects that annual credit allocations will increase from $4.3 billion in 2004 to $5.3 billion in 2008, because of population growth and inflation. As a result, they estimate that the tax expenditure on low-income housing credits will total $23.8 billion from 2004 to 2008 (Joint Committee on Taxation 2003). Thirty percent of credits are claimed by individuals, and 70 percent by corporations. Note that credits paid in any year go to projects placed in service not just in that year, but also during the preceding 10 years.

Official estimates are not available, but Abt Associates, Inc. (2003), estimates that 1,300 projects and 90,000 units were placed in service annually from 1995 to 2001. Sixty-two percent of the projects placed in service were newly built; 37 percent of the projects were rehabs; and 1 percent of the projects were composed of both newly constructed and rehabilitated buildings. The average project size is 68 units and most (95 percent) units within a project are Low-Income Housing Tax Credit (LIHTC) program units. These tax credit units represent approximately one-fifth of all new multifamily rental unit completions from 1995 to 2001.

Cummings and DiPasquale (1999) find that tax credits cover 46 percent of the average development cost of a project; 38 percent is covered by first mortgage loans and the rest by gap financing provided primarily by local governments. They report that 78 percent of tax credit projects had positive cash flows. Moreover, the perception of risk associated with the program appears to have declined over time. Cummings and DiPasquale estimate that the internal rate of return of tax credit projects declined from 20.5 percent in 1987 to 11.8 per- cent in 1994. Part of the decline in the internal rate of return may also reflect better management of credit allocations by state housing authorities.

The GAO (1997) also concludes that tax credit projects became more efficient over the first 10 years of the program, for several reasons: increased participation of private banks as lenders in order to comply with the Community Reinvestment Act; equity funds developed by states that compete for tax credits and drive down the "price"; and increasing investment by corporations encouraged by tax laws that exempt them from passive loss rules that apply to individual investors.

Analysis

The question "Is the low-income housing credit an efficient way to help poor people meet their housing needs?" can be dissected into two parts. First, how efficient are government subsidies to the supply of housing compared with demandbased subsidies? Second, if supply subsidies are called for, how efficient is the low-income housing credit compared with other subsidy mechanisms? The debate on both of these questions has been lively and remains unresolved.

Efficacy of housing supply programs

If the supply of low-income housing is very elastic in the long run, then production of limited amounts of subsidized housing will simply replace other housing that would otherwise have been provided. Housing supplied or subsidized by the government might increase the average quality of housing available to low-income tenants, but it would have little lasting effect on the quantity or price of housing available to poor people. (See Weicher and Thibodeau 1988 for a discussion of the effects of subsidized housing on the housing market as a whole.) Moreover, because new and substantially rehabilitated housing is expensive to produce, it is likely to be worth far less to tenants than an equal cash supplement, such as housing vouchers. Furthermore, DiPasquale, Fricke, and Garcia-Diaz (2003) estimate that the average cost of producing a tax credit unit exceeds the cost of the average voucher unit by 19 percent. (For an early comparison of housing vouchers with low-income housing credits, see Congressional Budget Office 1992.)

Advocates of supply subsidies argue that, in certain housing markets, the supply of housing is inelastic, even in the long run. Apgar (1990) argues that shortages of decent lowrent housing have been getting worse over time. As a result, housing produced or subsidized by the government will permanently increase the stock of housing available. Moreover, because supply is inelastic, supply-based programs can be more efficient than demand subsidies because they generate pecuniary externalities. The augmented supply lowers the price of housing, generating benefits not only for the tenants in the new housing, but also for others in the market (Coate, Johnson, and Zeckhauser 1994). In contrast, a program like housing vouchers, which expands the demand for housing without affecting the supply, raises the price, harming low-income tenants who do not have vouchers.

But supply is likely to be elastic in markets with high vacancy rates even in the short term (because there is a surplus of units available to rent). According to the U.S. Census Bureau’s Annual Survey of Housing Vacancies and Homeownership, the vacancy rate for rental housing was 9.8 percent in 2003, which is relatively high by historical standards. By comparison it was 7.4 percent in 1993 and 5.7 percent in 1983. Vacancy rates have been on a steady upward trend since 1981 and reached the record level of 10.4 percent by the first quarter of 2004. National rental vacancy rates from the decennial census are 7.1 percent, 8.5 percent, and 6.8 percent for 1980, 1990, and 2000, respectively. In the 75 largest metropolitan areas, the average vacancy rate was 9.6 percent in 2003. Only four metropolitan areas had vacancy rates below 5 percent.

Examining more detailed data on housing characteristics for 2003, the vacancy rate for older housing (built before 1980, and thus more likely of lower quality) is 8.6 percent, compared with 13.3 percent for housing built during or after 1980. Although the market for older housing is relatively tight, these vacant units-especially the inadequate ones- are likely to leave the stock if better quality housing is supplied (Weicher 1990). In other words, tax credits may not increase markedly the overall supply of housing.

In the long run, the supply of unsubsidized housing is likely very elastic because housing investments have to earn rates of return competitive with other capital. Malpezzi and Vandell (2002) find a very high rate of substitution between LIHTC and unsubsidized units based on a cross-state regression. Thus, even in tight markets, subsidized housing starts are likely to displace unsubsidized housing over the long run. However, adjustment to this long-run equilibrium may be slow. Thus, targeted supply subsidies might speed adjustment to equilibrium. For this reason, some analysts advocate supply subsidies as a complement to demand subsidies in tight housing markets (Struyk 1990).

Supply subsidies might also enhance efficiency if they mitigate market failures. The market for low-income housing might fail for several reasons. Lenders might be unwilling to lend to investors in neighborhoods where many poor people live because of class or racial biases (a form of discrimination often referred to as "redlining"). To the extent that discrimination in lending occurs, a capital subsidy might improve the allocation of capital.

In addition, when an area falls into decline, decayed housing creates a negative externality that discourages investment in housing on a small scale. A significant part of property value arises from the characteristics of the neighborhood (indeed, externalities are sometimes referred to as "neighborhood effects" in the economics literature). Thus, if all or most of the housing in a neighborhood were improved, it would be worth more per unit than would one or two improved buildings. As a result, government investment, directly or indirectly, might generate social benefits by improving a neighborhood.

Indirect evidence of such market failures might be the poor quality of housing available for rent to poor families. The Joint Center for Housing Studies of Harvard University (2003) reports that 15 percent of renter households in the lowest income quintile lived in households with moderate or severe structural inadequacies in 2001, compared with 10.5 percent of the second quintile and approximately 8 percent for each one of the top three quintiles. However, a simpler competing explanation is that low-income households cannot afford decent housing, even if it is available. In 2001, 55 percent of renter households in the lowest income quintile faced a severe cost burden, compared with an average of 4 percent for the top four income quintiles. In most housing markets, the primary problem seems not to be market failure, but affordability.

Advantages and disadvantages of the low-income housing credit

Supposing policymakers decide that the government should build or subsidize the creation of affordable housing, how does the low-income housing credit compare with other available mechanisms for increasing supply? The credit has obvious political advantages. First, the private sector builds the housing, rather than state or federal governments. Second, the program devolves almost total control over allocation of the credits to state housing agencies, a sterling example of how fiscal federalism can be implemented through the tax code. Third, new spending done through the tax code (see tax expenditures) is viewed as a tax cut, rather than increased government spending.

These advantages come at a high price. The equity capital raised for investment generally comes from syndicates of individual investors or from corporations. Syndicating limited partnerships is technically complex, and expensive. Moreover, because the credit is very complex and risky to investors, who might become ineligible for credits and have to repay past credits with interest if their project fails to comply with the restrictions of the law, investors require high after-tax rates of return. Stegman (1991), using an estimated internal rate of return of 15 percent, estimates that housing produced by syndicates of individuals costs the government nearly twice what a direct capital grant to the project sponsor would cost.

Corporate investors do not have to form syndicates to raise capital, thereby lowering overhead costs, but this may translate into higher profits rather than smaller subsidies. Lampert (1993), who calls the low-income housing credit "perhaps, the last true tax shelter remaining-especially for the widely held corporation," claims that "lofty internal rates of return, ranging from 15 percent to 25 percent, are possible in today’s market."

The costs of the low-income housing credit include the costs of administration by federal and state housing agencies and by the IRS. Adequate monitoring by state housing agencies and the IRS would be expensive, but, without such monitoring, credits might be allocated to fraudulent claimants, to those who do not comply with the income or rent restrictions of the law, or to investors who otherwise manage to extract unwarranted economic rents. The GAO (1997) observes that the cost control measures of state agencies to ensure the reasonableness of tax credits vary by state from a limited verification of cost and financing data to frequent site visits and independent audits of developers. GAO estimates that for about 14 percent of the projects, the states lacked complete information on the uses of funds. Although the IRS has improved its oversight since 1995, the year of the study, the IRS still faces obstacles in collecting reliable data on taxpayer compliance. Both the costs of noncompliance and the costs of administration must be added to the total cost of the tax credit approach.

ADDITIONAL READINGS

• Abt Associates, Inc. "Updating the Low-Income Housing Tax Credit Database: Projects Placed in Service through 2001." Paper prepared for the U.S. Department of Housing and Urban Development. Cambridge, MA: Abt Associates, Inc., 2003.
• Apgar, William C., Jr. "Which Housing Policy Is Best." Housing Policy Debate 1 (1990): 1-32.
• Coate, Stephen, Stephen Johnson, and Richard Zeckhauser. "Pecuniary Redistribution through In-Kind Programs." Journal of Public Economics 55 (September 1994): 19-40.
• Congressional Budget Office [CBO]. "The Cost-Effectiveness of the Low- Income Housing Tax Credit Compared With Housing Vouchers." Washington, DC: CBO staff memorandum, April 1992.
• Cummings, Jean L., and Denise DiPasquale. "The Low-Income Housing Tax Credit: An Analysis of the First Ten Years." Housing Policy Debate 10, no. 2 (1999): 251-307.
• DiPasquale, Denise, Dennis Fricke, and Daniel Garcia-Diaz. "Comparing the Costs of Federal Housing Assistance Programs." Federal Reserve Bank of New York Economic Policy Review (June 2003): 147-66.
• Guggenheim, Joseph. Tax Credits for Low-Income Housing. 12th ed. Glen Echo, MD: Simon Publishers, 2003.
• Guttman, George. "Despite Assurances to Congress, IRS Slow to Track LIHC Compliance." Tax Notes 67 (June 26, 1995): 1715-17.
• Joint Center for Housing Studies of Harvard University. The State of the Nation’s Housing 2003. Cambridge, MA: President and Fellows of Harvard College, 2003.
• Joint Committee on Taxation. "Estimates of Federal Tax Expenditures for Fiscal Years 2004-2008." Washington, DC: U.S. Government Printing Office, December 22, 2003.
• Lampert, Peter M. "Corporate Investment in the Low-Income Housing Credit." The Journal of Taxation 79 (December 1993): 344-50.
• Malpezzi, Stephen, and Kerry Vandell. "Does the Low-Income Housing Tax Credit Increase the Supply of Housing?" Journal of Housing Economics 11, no. 4 (2002): 360-80.
• Stegman, Michael A. "The Excessive Costs of Creative Finance: Growing Inefficiencies in the Production of Low-Income Housing." Housing Policy Debate 2 (1991): 357-73.
• Struyk, Raymond J. "Comment on William Apgar’s ‘Which Housing Policy Is Best.’ " Housing Policy Debate 1 (1990): 41-51.
• U.S. General Accounting Office [GAO]. Tax Credits: Opportunities to Improve Oversight of the Low-Income Housing Program. Washington, DC: GAO, 1997.
• Weicher, John C. "Comment on William Apgar’s ‘Which Housing Policy Is Best.’ " Housing Policy Debate 1 (1990): 33-39.
• Weicher, John C., and Thomas G. Thibodeau. "Filtering and Housing Markets: An Empirical Analysis." Journal of Urban Economics 23 (January 1988): 21-40.