This is the accessible text file for GAO report number GAO-06-784 entitled 'Rental Housing Programs: Excluding Servicemembers' Housing Allowances from Income Determinations Would Increase Eligibility but Other factors May Limit Program Use' which was released on July 31, 2006. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to Congressional Committees: July 2006: Rental Housing Programs: Excluding Servicemembers' Housing Allowances from Income Determinations Would Increase Eligibility, but Other Factors May Limit Program Use: GAO-06-784: GAO Highlights: Highlights of GAO-06-784, a report to congressional committees Why GAO Did This Study: Although the Department of Defense (DOD) pays active-duty servicemembers who do not live in military housing a Basic Allowance for Housing (BAH) to help them afford private market residences, expected growth at some military installations has raised concerns about whether nearby communities will have enough affordable rental housing for incoming personnel. In response to a congressional mandate, GAO assessed (1) how excluding BAH would affect servicemembers’ eligibility to apply for federal rental housing programs and (2) factors that could affect their use of the programs in selected communities gaining military personnel. GAO compared servicemembers’ eligibility for the programs as of December 2005 by including and excluding BAH from income determinations and examined factors affecting potential program use near four growing military installations. What GAO Found: Excluding BAH from income determinations for federal rental housing programs would have substantially increased the percentage of servicemembers eligible to apply for the programs as of December 2005, assuming military pay was their only income. To be eligible to apply for rental assistance programs of the Departments of Housing and Urban Development (HUD) and Agriculture (USDA), or to live in units produced by the Internal Revenue Service’s (IRS) Low-Income Housing Tax Credit program, households must have incomes at or below a specific limit, generally 50 percent or 60 percent of the median household income for their area. At the 50 percent income limit, 20 percent of servicemembers who received BAH would have been eligible if BAH were excluded from income determinations, compared with 1 percent with BAH included. Most junior enlisted members would have been eligible if BAH were excluded, as would have small percentages of senior personnel. However, at all levels, many would not have been eligible if their households had even modest income from other sources. Agency and community officials cited factors that could limit the role of federal programs in building housing or helping servicemembers afford existing units near four installations that GAO examined. DOD officials said that servicemembers would be unlikely to need the programs because BAH payments provide for the median cost of market- rate housing. Some community officials said the tax-credit program, which spurs housing production, could be useful if more servicemembers qualified. But developers would have to compete for tax credits, and market factors—such as the financial feasibility of building units that junior enlisted members could afford—could limit their interest. The HUD and USDA programs might help some servicemembers rent existing units, but—because the programs are not entitlements—servicemembers could face lengthy waits, and eligible civilians might wait longer for assistance. Figure: Potential Eligibility for Federal Rental Housing Programs, Assuming December 2005 Military Income Was the Only Household Income: [See PDF for Image] Sources: GAO and DOD. [End of Figure] What GAO Recommends: This report contains a matter for congressional consideration stating that if the primary intent of excluding BAH from income determinations for federal rental housing programs is to increase the supply of affordable rental housing for servicemembers, Congress should consider first applying such a change only to programs that stimulate housing production. DOD commented that servicemembers should be eligible for federal rental housing programs on the same terms as their civilian counterparts. [Hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-06-784]. To view the full product, including the scope and methodology, click on the link above. For more information, contact David G. Wood at (202) 512-8678 or woodd@gao.gov. [End of Section] Contents: Letter: Results in Brief: Background: Excluding BAH When Determining Income Would Extend Eligibility to More Servicemembers, Assuming No Additional Household Income: Eligibility Aside, Lack of Demand and Other Factors Could Limit Servicemembers' Use of Federal Rental Housing Programs: Conclusions: Matter for Congressional Consideration: Agency Comments and Our Evaluation: Appendixes: Appendix I: Objectives, Scope, and Methodology: Objectives: Scope and Methodology: Appendix II: Office of the Under Secretary of Defense: Appendix III: GAO Contact and Staff Acknowledgments: Tables: Table 1: Federal Rental Housing Programs That Support Housing Production or Subsidize Tenants' Rents: Table 2: Amount of Additional Income That Would Have Disqualified Half of the Married Servicemembers, as of December 2005: Table 3: Expected Growth in Military Personnel at Four Selected Installations: Figures: Figure 1: Military Compensation Ranges, by Pay Grade, 2006: Figure 2: Servicemembers' Potential Eligibility, by Pay Grade, for Federal Rental Housing Programs at the 50 Percent and 60 Percent of AMI Limits, as of December 2005: Figure 3: Servicemembers' Potential Eligibility, by Family Size, for Federal Rental Housing Programs at the 50 Percent and 60 Percent of AMI limits, as of December 2005: Figure 4: Percentage of Servicemembers Potentially Eligible for Federal Rental Housing Programs That Were Married or Received Other Military Pay, as of December 2005: Figure 5: Servicemembers Potentially Eligible for Tax-Credit Units at the 60 Percent of AMI limit, as of December 2005: Figure 6: Income Ranges of Housing Choice Voucher Recipients and Servicemembers in Selected Pay Grades, 2006: Abbreviations: AMI: area median income: BAH: Basic Allowance for Housing: BAS: Basic Allowance for Subsistence: BRAC: base realignment and closure : DOD : Department of Defense: HUD: Department of Housing and Urban Development: IRS: Internal Revenue Service: LIHTC: Low-Income Housing Tax Credit: USDA: U.S. Department of Agriculture : July 31, 2006: The Honorable Christopher Bond: Chairman: The Honorable Patty Murray: Ranking Minority Member: Subcommittee on Transportation, Treasury, the Judiciary, Housing and Urban Development, and Related Agencies: Committee on Appropriations: United States Senate: The Honorable Joe Knollenberg: Chairman: The Honorable John W. Olver: Ranking Minority Member: Subcommittee on Transportation, Treasury, and Housing and Urban Development, The Judiciary, District of Columbia, and Independent Agencies: Committee on Appropriations: House of Representatives: As a result of base closings and realignments and other restationing actions, some military installations will be experiencing growth in the next several years, raising concerns about whether the nearby communities will have an adequate supply of affordable housing for incoming military personnel or whether communities can develop such housing. To help make housing affordable to servicemembers, and in keeping with its policy of relying on the private market as the primary source of housing for servicemembers, the Department of Defense (DOD) pays a Basic Allowance for Housing (BAH) to servicemembers stationed in the United States who do not live in military housing. In fiscal year 2005, DOD made about $11.6 billion in BAH payments to approximately 850,000 active-duty servicemembers. DOD sets the housing allowance amounts annually to cover the median cost of rent, including utilities and renter's insurance. The housing allowance amounts vary according to the location and rank (pay grade) of each servicemember, and also vary depending on whether the servicemember has any dependents. Servicemembers may choose to spend more or less than their allowance amounts on their actual housing expenses. Although BAH is intended to allow servicemembers to acquire suitable housing in the private market, some community leaders and housing developers have argued that federal rental housing programs also offer the potential to help provide affordable housing for servicemembers relocating to growing installations. The federal programs include the Low-Income Housing Tax Credit (LIHTC) and tax-exempt multifamily housing bond programs, which are administered by the Internal Revenue Service (IRS) of the Department of the Treasury (Treasury) and the states and support development of rental housing.[Footnote 1] Other programs, including public housing and programs that subsidize tenants' rents, are administered by the Department of Housing and Urban Development (HUD) and the Department of Agriculture (USDA).[Footnote 2] To be eligible to apply for these programs, households must have incomes at or below a specific limit, generally 50 percent or 60 percent of the median household income for their area, adjusted for family size. (These programs are not entitlements, however, and not all eligible households receive assistance.) To help servicemembers meet these income eligibility restrictions, proponents of using federal programs to address the military's housing needs have proposed excluding BAH from servicemembers' incomes when determining eligibility for the programs.[Footnote 3] Noting concerns about the availability of affordable private housing for servicemembers, particularly in rural areas with growing installations, the conference report accompanying the fiscal year 2006 Transportation, Treasury, Housing and Urban Development, the Judiciary, the District of Columbia, and Independent Agencies Appropriations Act directed us to report on the potential effect of excluding BAH from income when determining eligibility for federal rental housing programs.[Footnote 4] Accordingly, this report discusses (1) how excluding BAH from income determinations would have affected the eligibility of servicemembers receiving BAH as of December 2005 and (2) programmatic and market factors that could affect eligible servicemembers' participation in the programs in selected communities gaining military personnel. To address these objectives, we obtained DOD's personnel data for December 2005 and analyzed 702,975 records of servicemembers who were BAH recipients at that time to determine their potential eligibility for federal rental housing programs. We included the following federal programs: HUD's public housing, Housing Choice Voucher, and project- based Section 8; USDA's Section 515 Rural Rental Housing Loans and Section 521 Rural Rental Assistance; and IRS's LIHTC and tax-exempt multifamily housing bonds. We compared potential eligibility for rental housing programs by including and excluding BAH from income eligibility determinations. Because data on spousal income and other sources of income were unavailable, for this analysis we assumed that the primary components of military pay were the only sources of income for the servicemember households. We tested the data that we used in our analysis and found it sufficiently reliable for our purposes. Furthermore, we interviewed and reviewed relevant documentation from installation officials, rental housing program officials, and community organization representatives in the following four communities near installations gaining military personnel: Fort Benning, Georgia; Fort Bliss, Texas; Fort Drum, New York; and Fort Riley, Kansas.[Footnote 5] We cannot generalize the information from these installations to all installations that will gain military personnel. We conducted our work in and around Washington, D.C., and Junction City and Manhattan, Kansas, between November 2005 and July 2006, in accordance with generally accepted government auditing standards. Appendix I contains a more detailed description of our scope and methodology. Results in Brief: Excluding BAH from income when determining servicemembers' eligibility for federal rental housing programs would have substantially increased the percentage that would have been eligible to apply for the programs as of December 2005, assuming the primary components of military pay were their only sources of income. More specifically: * Using an income limit of 50 percent of the area median, approximately 20 percent of the servicemembers would have been eligible, compared with the less than 1 percent that were eligible with BAH in income determinations. * Using a higher limit of 60 percent of area median (which applies only to some LIHTC and tax-exempt multifamily housing bond properties) approximately 40 percent of the servicemembers would have been eligible. * Some increase in eligibility would have occurred at all servicemember pay grades, but the greatest impact would have been at the lower pay grades. For example, 65 percent or more of servicemembers in the lowest pay grades would have been eligible to apply for the federal programs, whereas very small percentages of senior servicemembers and officers would have been eligible. * The increase in eligibility would have affected servicemember families of all sizes, but those with the largest families (nine or more persons) would have been somewhat more likely to be eligible because the programs' income limits increase with family size. Although we lacked data on servicemember household incomes from nonmilitary sources, DOD data show that at least 80 percent of the potentially eligible servicemembers were married; thus, these households could have had additional income earned by a spouse. Using an income limit of 50 percent of the area median, we estimated that additional annual income of about $4,000 would have disqualified half of the married servicemembers who were potentially eligible for the federal programs, even if BAH were excluded from income determinations. In the four communities we examined, programmatic and housing market factors--for example, a potential lack of demand for the programs among servicemembers, the costs of developing new housing, and the limited availability of rental assistance--may limit the extent to which the federal programs would help developers increase housing supply or help servicemembers afford existing housing. DOD officials said that servicemembers would be unlikely to need federal rental housing programs because BAH payments cover median local housing costs and would adjust annually to reflect any increases in market rents that resulted from increased demand for housing near growing installations. Yet, some community officials cited the LIHTC program as a potential tool to build more housing for which incoming servicemembers might qualify, but only if BAH were excluded when determining eligibility so that more lower-income servicemembers could live in the new tax-credit units. However, even if more servicemembers were eligible, LIHTC-funded development still might not occur near the installations because the state agencies that award available tax credits have a variety of priorities. Furthermore, market conditions could make developers more or less likely to consider using LIHTC to finance their projects. For example, developers might be more disposed to seek LIHTC financing for projects near Fort Riley, Kansas, because a substantial number of incoming servicemembers likely will be low-ranking personnel with families--a population that might have more difficulty affording market- rate units than the more senior servicemembers. In contrast to the LIHTC program, the HUD and USDA programs generally do not support production of new housing, but rather subsidize rents for tenants of existing units. However, because these programs are not entitlements, the limited availability of assistance may preclude eligible servicemembers from using them, especially in areas with long lists of civilian applicants already awaiting assistance, according to agency and community officials. Finally, to the extent that more servicemembers applied for these programs, lower-income civilians might face longer times on waiting lists for rental assistance because of the larger pool of applicants. If the primary intent of excluding BAH from income determinations for federal rental housing programs is to increase the supply of rental housing that servicemembers with the lowest incomes could afford, Congress should consider first applying such a change only to programs intended to stimulate production of such housing, such as LIHTC and tax- exempt multifamily housing bonds. We provided a draft of this report to DOD, HUD, IRS, Treasury, and USDA. DOD, HUD, and IRS provided technical comments, which we incorporated in the final report as appropriate. USDA and Treasury declined to comment on the draft report. DOD commented that BAH does an excellent job of achieving the objective of providing servicemembers with the same quality and quantity of housing that their civilian counterparts can afford, and the department noted that in cases of sudden shortages, BAH rates would increase to allow servicemembers to bid competitively for housing. DOD also stated that servicemembers should be eligible for federal rental housing programs under the same terms as their civilian counterparts. DOD's comments are discussed in the Agency Comments and Our Evaluation section of this report, and its written comments appear in appendix II. Background: BAH, one of several components of military compensation, is intended to provide servicemembers with an allowance to enable them to obtain suitable housing when military-owned housing is not provided. Accordingly, BAH payments reflect the cost of housing where servicemembers are stationed, and the payments change annually in response to increases or decreases in local housing costs. Still, the most recent base realignment and closure (BRAC) process, among other restationing actions, will cause movement of large numbers of military personnel to communities that initially may lack enough private housing that is affordable to most servicemembers. Several HUD, USDA, and IRS rental housing programs that are intended to make housing affordable to low-income households count BAH as income when assessing the eligibility of active-duty servicemembers. Military Compensation Incorporates Basic Pay, Allowances, and Bonuses: BAH is one of several elements of regular military compensation.[Footnote 6] Regardless of whether they live in military- owned housing or receive BAH, servicemembers receive basic pay and a Basic Allowance for Subsistence (BAS). BAH and BAS are not subject to federal income tax. With the addition of average BAH payments to the other two pay elements, regular military compensation in 2006 starts at $26,401 for the lowest-ranking enlisted servicemembers and culminates at $183,196 for the highest-ranking officers, excluding consideration of any tax advantage because the allowances are not subject to federal income tax (see fig. 1). Figure 1: Military Compensation Ranges, by Pay Grade, 2006: [See PDF for image] Sources: GAO and DOD. Note: This figure reflects the cash components of regular military compensation and does not include the value of the tax advantage that accrues to servicemembers because BAH and BAS are not subject to federal income tax. [End of figure] In addition to the primary elements of military compensation shown in figure 1, servicemembers with duty stations in more than 55 continental U.S. locations, where nonhousing expenses exceed the national average by at least 8 percent, receive a cost-of-living allowance. Servicemembers also may receive other types of pay, allowances, or bonuses, depending on their professional backgrounds, skills, or duties. For example, servicemembers may receive special pay for hardship duty or exposure to hostile fire, allowances when they are separated from their families because of a change in station or a temporary duty assignment, and bonuses for enlistment and reenlistment.[Footnote 7] According to DOD officials, in March 2006, about 950,000 personnel lived in private housing (including privatized military family housing) and received BAH--including roughly 70 percent of active-duty servicemembers in the United States, as well as some activated reservists and servicemembers stationed overseas whose dependents lived in the United States.[Footnote 8] DOD generally requires enlisted servicemembers in the lowest ranks who do not have dependents to live on base in furnished living quarters, commonly referred to as barracks. These enlisted servicemembers do not receive BAH. Each year, DOD sets BAH rates (i.e., the allowances servicemembers receive monthly) that are based on the median local monthly cost of housing, including current market rents, utilities, and renter's insurance. The amounts that servicemembers receive also are based on their pay grades and whether they have dependents. To calculate BAH rates for different pay grades, DOD uses six standard categories of housing--ranging from an one-bedroom apartment to a four-bedroom, single-family detached house--that are intended to match the housing normally occupied by civilians with comparable incomes. DOD applies separate categories to servicemembers with and without dependents, but the number of dependents does not affect the BAH amount. BAH rates have increased since 2000 as DOD implemented an initiative to reduce servicemembers' out-of-pocket housing costs. Prior to 2005, the BAH rate for each area and pay grade was the local median monthly housing cost minus a percentage of the national median monthly housing cost. That deduction represented the amount that servicemembers would have to pay out of pocket if their actual housing costs exactly matched the median local housing cost for their pay grade. In 2000, the deduction was 19 percent of the national median housing cost. DOD gradually reduced the deduction so that, by 2005, BAH rates equaled the median housing cost for each area and pay grade. Furthermore, while the housing allowance is calculated on the basis of the rental market, servicemembers may choose to apply their allowance toward purchasing a home, and they are free to spend more or less than their allowance on housing. We reported in April 2006 that the increases in BAH rates had made it possible for more servicemembers to afford private housing in the local market, thus reducing the need for privatized housing at installations.[Footnote 9] This has recently contributed to lower-than- expected occupancy rates at some privatized housing projects. If some privatized projects persistently experience lower-than-expected occupancy rates, they could encounter financial difficulties or, at worst, failures. To avoid such concerns in future privatization projects, we recommended that DOD determine how increased BAH rates would affect installations' housing requirements and provide guidance on how the services should incorporate this information into their assessments of the need for privatized family housing. More Than 100 Installations Will Gain Personnel through the 2005 BRAC Process: The National Defense Authorization Act for Fiscal Year 2002 authorized a new BRAC process in 2005. This was the fifth such process in the last two decades, but the first since 1995.[Footnote 10] As in previous processes, Congress enacted the legislation to close unneeded bases and realign others. On November 9, 2005, Congress accepted in their entirety the most recent BRAC recommendations for base closings and realignments. DOD has 6 years, or from 2005 until September 15, 2011, to implement these recommendations. The 2005 BRAC process affects a substantial number of communities surrounding installations that are expected to experience considerable growth in military personnel. While scores of installations will gain or lose military personnel, more than 20 installations each are expected to gain between 2,000 and 21,000 military, civilian, and mission-support contractor personnel.[Footnote 11] For the most part, installations with the largest gains are located in predominantly urban counties. However, some installations are in rural areas that may have less housing available, raising the possibility that incoming personnel initially could face a shortage of nearby housing that is affordable to them.[Footnote 12] The installations that will gain the most personnel through BRAC are Department of the Army installations, with their gains attributable to actions such as the consolidation of various activities and the return of personnel from overseas locations under DOD's integrated global presence and basing strategy. In addition to shifts related to BRAC, the Army is realigning personnel as it changes its force structure. Many Federal Rental Housing Programs Include BAH in Income Determinations: Various HUD, USDA, and IRS rental housing programs are intended to make housing affordable for lower-income renters. None of the federal agencies that administer these programs maintain data on the number of participating servicemembers. The programs either support the production of new or rehabilitated rental housing for eligible families or subsidize tenants' rents to make existing units affordable (see table 1). Specifically: * Among the production programs, LIHTC and Section 515 Rural Rental Housing Loans require property owners to restrict the rents that eligible tenants pay. The rent on each tax-credit unit generally cannot exceed 30 percent of the applicable income limit, adjusted for the number of bedrooms. Tenants pay 30 percent of their adjusted incomes toward the rent on Section 515 units. The tax-exempt multifamily housing bonds program requires units to be set aside for eligible families, but the rents on these units generally do not have to be restricted. * Rental assistance programs make payments to property owners to make up the difference between an eligible tenant's rent contribution (generally, 30 percent of adjusted monthly income) and a unit's total rent. The Housing Choice Voucher program offers tenant-based rental assistance that tenants can use to rent privately owned apartments or single-family homes, and that they can transfer to new residences if they move. In contrast, the project-based Section 8 and Section 521 Rural Rental Assistance programs offer project-based rental assistance, which is attached to specific properties and is available to tenants only when they are living in units at these properties. * Public housing also subsidizes tenants' rents. However, rather than making rental assistance payments to owners that are keyed to tenants' rent payments, HUD provides public housing agencies with annual operating subsidies that are based partly on the property's projected overall rental income. Table 1: Federal Rental Housing Programs That Support Housing Production or Subsidize Tenants' Rents: Federal agency: HUD; Program: Housing Choice Voucher; Program type: Rental assistance; Description: Provides tenant-based rental assistance for households with extremely low; very low; and, on an exception basis, low incomes. Federal agency: HUD; Program: Public housing; Program type: Rental assistance; Description: Provides subsidized housing operated by public housing authorities for households with extremely low to low incomes. Federal agency: HUD; Program: Project-based Section 8 rental assistance; Program type: Rental assistance; Description: Provides project-based rental assistance in multifamily properties for households with extremely low, very low, and (in certain properties) low incomes. Federal agency: IRS; Program: Low-Income Housing Tax Credit; Program type: Production; Description: Through the allocation of federal tax credits by state agencies, provides an incentive for investment in development of multifamily housing that sets aside a portion of units for households with very low incomes or incomes at or below 60 percent of the area median income. Federal agency: IRS; Program: Tax-exempt multifamily housing bonds; Program type: Production; Description: Allows government entities to issue tax-exempt bonds whose proceeds finance private rental properties that set aside a portion of units for households with very low incomes or incomes at or below 60 percent of the area median income. Federal agency: USDA; Program: Section 515 Rural Rental Housing Loans; Program type: Production; Description: Makes loans for the construction and rehabilitation of rural multifamily properties for households with very low to moderate incomes. Federal agency: USDA; Program: Section 521 Rural Rental Assistance; Program type: Rental assistance; Description: Provides project-based rental assistance to tenants with very low and low incomes in Section 515 and other USDA- financed multifamily properties. Sources: GAO, HUD, IRS, and USDA. [End of table] All of these federal programs use a common definition of income as set out in a HUD regulation.[Footnote 13] Under this definition, incomes of servicemember households include all regular pay, special pay, and allowances (including BAH) of the servicemember, except special pay to servicemembers who are exposed to hostile fire. Each program determines households' eligibility to apply by comparing their incomes with an income limit, expressed as a percentage of the area median.[Footnote 14] The income limits are adjusted for family size, with higher limits for larger families. In addition, the HUD and USDA programs use tenant income (with certain adjustments) to determine how much of a unit's rent the tenant will pay. The programs generally target various categories of households, defined according to the relationship between a household's income and the local area median income (AMI): extremely low (household income is no more than 30 percent of AMI), very low (no more than 50 percent of AMI), low (no more than 80 percent of AMI), and moderate (no more than $5,500 above 80 percent of AMI). In addition to these categories, the LIHTC and tax-exempt multifamily housing bond programs can target households with incomes that are no more than 60 percent of AMI. For purposes of this report, we focused on the 50 percent and 60 percent of AMI limits because they generally apply to new applicants for the two largest federal rental housing programs, Housing Choice Voucher and LIHTC.[Footnote 15] The federal rental housing programs are not entitlements and, as a result, do not assist all households that HUD has identified as having housing needs--that is, households with very low incomes that pay more than 30 percent of their income for housing, live in substandard housing, or both. According to HUD data for 2003, federal rental housing programs assisted an estimated 4.3 million households, or 27 percent of all renter households with very low incomes.[Footnote 16] Over 9 million renter households with very low incomes (about 59 percent) did not receive federal assistance and had housing needs. Of these 9 million households, over 5 million had what HUD terms "worst case" needs--that is, they paid over half of their income in rent, lived in severely substandard housing, or both. Excluding BAH When Determining Income Would Extend Eligibility to More Servicemembers, Assuming No Additional Household Income: Assuming that the primary components of military pay were the only sources of servicemembers' household incomes, excluding BAH payments from income when determining servicemembers' eligibility for federal rental housing programs would have substantially increased the percentage that would have been eligible to apply for the programs as of December 2005. Specifically, most junior enlisted members would have been eligible for the programs, as would have much smaller percentages of senior servicemembers. In addition, although few in number, servicemembers with the largest families (nine or more persons) would have been somewhat more likely to be eligible for the programs than those with smaller families. However, to the extent that servicemembers' households had income from nonmilitary sources, fewer of them would have been eligible for the federal programs. We lacked data on servicemember household incomes from nonmilitary sources, but at least 80 percent of the potentially eligible servicemembers were married, and income earned by spouses would likely have disqualified many of these households. Most Junior Enlisted Members and Some Senior Members Would Have Been Eligible if Military Pay Were Their Sole Source of Income: Assuming that the primary components of servicemembers' military pay were their only sources of household income in 2005, we found that by excluding BAH from income determinations, 19.9 percent of servicemembers of all grades would have been eligible for federal rental housing programs that used an income limit of 50 percent of AMI, compared with less than 1 percent of servicemembers with BAH included.[Footnote 17] Similarly, at the 60 percent of AMI limit, 39.3 percent of the servicemembers would have been eligible if BAH were excluded when determining income, compared with 4.8 percent if BAH were included (see fig. 2).[Footnote 18] At both income limits, most junior enlisted members (for our purposes, E-1 through E-4) would have been eligible for the programs if BAH were excluded.[Footnote 19] Specifically, at the 50 percent of AMI limit, substantial majorities of E-1s (92.4 percent), E-2s (78.7 percent), and E-3s (65.2 percent) would have been eligible. At the 60 percent of AMI limit, virtually all E-1s (99 percent) and E-2s (97.6 percent) and substantial majorities of E-3s (90.2 percent) and E-4s (64.6 percent) would have been eligible. Figure 2: Servicemembers' Potential Eligibility, by Pay Grade, for Federal Rental Housing Programs at the 50 Percent and 60 Percent of AMI Limits, as of December 2005: [See PDF for image] Sources: GAO and DOD. Note: This figure shows the potential eligibility to apply for federal rental housing programs among servicemembers who were receiving BAH as of December 2005, based on the payments they received that month for basic pay; BAS; BAH; and, where applicable, a cost-of-living adjustment for servicemembers living in high-cost areas. These data were from DOD's Active Duty Pay and Active Duty Military Personnel Master File systems. This analysis assumes that the primary components of military pay were servicemembers' only sources of income; income from other sources would reduce the percentage of eligible servicemembers. [End of figure] In addition, using the same assumption that household income included only the primary components of military pay, some senior enlisted members and officers would have been eligible for the programs if income determinations excluded BAH. Specifically, at the 50 percent of AMI limit, 19.2 percent of E-5s and 9.4 percent of E-6s would have been eligible, as would have very small percentages of servicemembers in pay grades E-7 through E-9 (see fig. 2). The percentage of eligible officers also would have been very small, as follows: 1 percent using the 50 percent of AMI limit, and 2 percent using the 60 percent of AMI limit. Excluding BAH, More Servicemembers with Families of All Sizes Would Have Been Eligible, Especially Larger Families: Again assuming that the primary components of military pay were the only sources of household income, by excluding BAH from income determinations, considerable percentages of servicemembers with families of all sizes would have been eligible for the programs, using either the 50 percent or 60 percent of AMI limit. However, because the programs' income limits increase with family size, servicemembers with larger families (although relatively few in number) generally would have been more likely to be eligible than those with smaller families (which were much greater in number). For example, with BAH in income determinations, 6.6 percent (59) of the largest families (those with nine or more persons) would have been eligible for programs using the 50 percent of AMI limit, compared with 0.5 percent (866) of the smallest (two-person) families (see fig. 3).[Footnote 20] With BAH excluded, 40.6 percent (361) of the largest families would have been eligible, compared with 23.7 percent (45,262) of the smallest families. The same general pattern held true for programs using the 60 percent of AMI limit. For example, 63.8 percent (568) of the largest families and 44.5 percent (84,999) of the smallest families would have been eligible if BAH were excluded from income determinations. Figure 3: Servicemembers' Potential Eligibility, by Family Size, for Federal Rental Housing Programs at the 50 Percent and 60 Percent of AMI limits, as of December 2005: [See PDF for image] Sources: GAO and DOD. Note: This figure reflects the potential eligibility of servicemembers to apply for federal rental housing programs as of December 2005, assuming that the primary components of military pay were the only sources of household income. [End of figure] Additional Household Income Would Reduce the Number of Eligible Servicemembers: To the extent that servicemembers had additional sources of household income, their actual eligibility for the federal rental housing programs would have been less than the percentages shown in our analysis. Additional sources of household income could include income on assets (such as savings accounts or mutual funds), employment of other household members, or types of military pay that we did not include in our analysis.[Footnote 21] For example, figure 4 shows that- -at both program income limits and with BAH included in or excluded from income determinations--at least 80 percent of the potentially eligible servicemembers were married and, thus, could have had additional income earned by a spouse. In addition, at least 9 percent of the potentially eligible servicemembers received other types of military pay. Figure 4: Percentage of Servicemembers Potentially Eligible for Federal Rental Housing Programs That Were Married or Received Other Military Pay, as of December 2005: [See PDF for image] Sources: GAO and DOD. Note: Although we knew the percentage of servicemembers that received other military pay in December 2005, we excluded such payments from our analysis because we lacked a reliable method to estimate the total amount of other military pay received in 2005 using data from a single month. [End of figure] To illustrate how additional sources of household income could affect eligibility for the federal rental housing programs, we calculated the amounts of additional income it would take to disqualify married servicemember households that would have been eligible on the basis of their military incomes alone. We found that, among the married servicemembers who were potentially eligible with BAH included in income determinations, income from even part-time, minimum-wage work by their spouses likely would have disqualified many from the federal programs. The same was true even if BAH were excluded from income determinations. For example, with BAH included, spousal income of $2,004 would have been enough to disqualify half of the married servicemembers that were potentially eligible for programs using the 50 percent of AMI limit (see table 2). With BAH excluded, spousal income of $4,044 would have been enough to disqualify half of the married servicemembers that were potentially eligible. At the 60 percent of AMI limit, $3,108 in spousal income would have disqualified half of the potentially eligible married servicemembers with BAH included in income determinations, compared with $6,180 if BAH were excluded. As shown in table 2, these amounts represent part-time work of 24 hours per week or less at the federal minimum wage. Table 2: Amount of Additional Income That Would Have Disqualified Half of the Married Servicemembers, as of December 2005: Income limit: 50 percent of AMI limit; BAH included in or excluded from income determinations: Included; Income that would have disqualified half of the married servicemembers[A]: $2,004; Hours per week at minimum wage[B]: 8. Income limit: 50 percent of AMI limit; BAH included in or excluded from income determinations: Excluded; Income that would have disqualified half of the married servicemembers[A]: 4,044; Hours per week at minimum wage[B]: 16. Income limit: 60 percent of AMI limit; BAH included in or excluded from income determinations: Included; Income that would have disqualified half of the married servicemembers[A]: 3,108; Hours per week at minimum wage[B]: 12. Income limit: 60 percent of AMI limit; BAH included in or excluded from income determinations: Excluded; Income that would have disqualified half of the married servicemembers[A]: 6,180; Hours per week at minimum wage[B]: 24. Sources: GAO and DOD. [A] For married servicemembers who were potentially eligible at each income limit, we calculated the median amount of additional income (such as spousal income) that would have disqualified them. This table presents the amount of spousal income that would have disqualified half of the married servicemembers. [B] We calculated the hours per week (assuming 50 weeks of work per year at the federal minimum wage, that is, $5.15 per hour) needed to earn the additional income that would have made half of the married servicemembers ineligible. v Eligibility Aside, Lack of Demand and Other Factors Could Limit Servicemembers' Use of Federal Rental Housing Programs: Agency officials and representatives from the four communities we examined described factors that may limit the role of federal rental housing programs in increasing the supply of housing or helping servicemembers afford existing housing, regardless of how BAH affects their eligibility. DOD officials said that servicemembers would be unlikely to need federal rental housing programs because BAH rates cover median local housing costs and would adjust annually to reflect any increases in market rents that resulted from increased demand for housing near growing installations. Yet, some community officials said that the LIHTC program could be used to build more affordable housing if more servicemembers were eligible. However, states would have to award tax credits to projects in these communities, and housing market factors--such as the financial feasibility of building market-rate units with rents that low-ranking servicemembers could afford--could affect developers' interest in using the LIHTC program. Furthermore, although HUD and USDA programs could help some eligible servicemembers rent existing units, the programs are not entitlements; the limited availability of this rental assistance may preclude servicemembers from using the programs. Also, if more servicemembers applied for these programs, eligible lower-income civilians might face longer times on waiting lists. Servicemembers May Not Need Programs because BAH Rates Cover Median Costs and Are Adjusted for Changing Rents: According to DOD officials, servicemembers would be unlikely to need federal rental housing programs to obtain affordable housing near growing installations because BAH rates cover local housing costs and would adjust for any increases in market rents that resulted from personnel gains. As of 2005, BAH rates fully cover the median local cost of housing at each installation. Officials noted that DOD's recent initiative to reduce servicemembers' out-of-pocket housing costs had resulted in substantial increases in BAH rates nationwide, including at the four selected installations we reviewed (Forts Benning, Bliss, Drum, and Riley). In addition, the officials said that, if increased demand for housing near a growing installation caused upward pressure on housing costs, DOD would adjust BAH rates upward as part of the annual rate-setting process, allowing servicemembers to obtain market- rate housing without additional federal assistance. However, if vacant units were not available in the communities immediately surrounding a growing installation, DOD officials acknowledged that some servicemembers might have to seek housing in outlying communities until the private market responded with new construction closer to the installation. Furthermore, the National Defense Authorization Act for Fiscal Year 2006 authorized the Secretary of Defense to prescribe temporary increases in BAH rates in disaster areas or areas that contain one or more installations that are experiencing a sudden increase in the number of servicemembers assigned to the installation.[Footnote 22] Specifically, a temporary increase in BAH rates would be based on the amount by which area housing costs increased because of the disaster or influx of service members and would apply until new rates for the next calendar year took effect. According to DOD officials, no installations had requested an increase in BAH rates because of installation growth, and the Secretary had not used this authority as of June 2006. If an installation requests a temporary increase in BAH rates because of installation growth, the officials said that DOD would review local market conditions to determine whether an increase was warranted. Ability of the LIHTC Program to Produce Needed Housing Could Be Affected Not Only by Servicemembers' Eligibility, but Also by the State Allocation Process and Market Conditions: To varying degrees, officials in the four communities (near Forts Benning, Bliss, Drum, and Riley) that we examined described a need to build more private housing for incoming servicemembers. Some officials indicated that, under certain conditions, the LIHTC program could help address their anticipated housing needs. According to officials at the selected installations, expected gains in military personnel ranged from about 4,500 at Fort Benning to about 19,500 at Fort Bliss (see table 3). The rural installations--Fort Drum and Fort Riley--expected more substantial growth relative to their existing supply of housing than did the urban installations. The communities generally did not yet have precise data on the expected number of servicemembers that would be most likely to seek private housing (servicemembers with families and those in higher pay grades who do not have dependents) or required to live in barracks (servicemembers in junior pay grades who do not have dependents). However, community officials in the Fort Riley area estimated that at least 9,000 more housing units would be needed, considering both the estimated number of incoming military personnel and the expected growth in the civilian employment at the installation. Similarly, community officials near Fort Drum estimated the need for approximately 2,000 additional units. Community officials in the Fort Benning and Fort Bliss areas had not developed such estimates, but they also anticipated that some new construction would be necessary to accommodate installation growth as well as other population increases. Table 3: Expected Growth in Military Personnel at Four Selected Installations: Installation: Fort Benning, Georgia; Primary counties: Muskogee and Chattahoochee; Urban or rural: Urban; Pregrowth military personnel: 15,000[A]; Expected growth in military personnel: 4,500[A]; Area's private housing units, 2002: 81,500; Growth as a percentage of private units: 6%. Installation: Fort Bliss, Texas; Primary counties: El Paso; Urban or rural: Urban; Pregrowth military personnel: 9,500; Expected growth in military personnel: 19,500; Area's private housing units, 2002: 232,000; Growth as a percentage of private units: 8. Installation: Fort Drum, New York; Primary counties: Jefferson; Urban or rural: Rural; Pregrowth military personnel: 11,000; Expected growth in military personnel: 6,000; Area's private housing units, 2002: 54,500; Growth as a percentage of private units: 11. Installation: Fort Riley, Kansas; Primary counties: Geary and Riley; Urban or rural: Rural; Pregrowth military personnel: 11,500; Expected growth in military personnel: 6,500 - 7,500; Area's private housing units, 2002: 36,000; Growth as a percentage of private units: 18 - 21. Sources: GAO, U.S. Census Bureau, and selected installations. Note: The installations are gaining servicemembers as a result of the BRAC process, the restationing of servicemembers from overseas, the Army's initiative to change its force structure, or a combination of these initiatives. We determined whether an installation was rural or urban using the Rural-Urban Commuting Area codes developed by USDA's Economic Research Service and the Office of Rural Health Policy, Department of Health and Human Services. We compiled the most recent Census estimates of the total number of private housing units in the primary counties within each installation's Army-defined market area. We considered a county to be primary if most of its land area fell within the market area. We present the installations' growth as a percentage of the primary counties' housing units only to illustrate the relative magnitude of each installation's growth. This table is not intended to suggest the number of servicemembers that would seek private housing, the number of existing units that might be available for rent or purchase, or the number of existing units that the Army considers suitable for servicemembers. [A] The figures for Fort Benning exclude its military student population because these personnel generally live on base and, thus, would not seek private housing. [End of table] Officials in some of these communities indicated that, under certain conditions, the LIHTC program could help address their anticipated housing needs. In particular, officials in the rural communities surrounding Fort Drum and Fort Riley said that the LIHTC program could help them build more affordable housing in response to installation growth, but only if more servicemembers would qualify to live in tax- credit units (see sidebar).[Footnote 23] Assuming that the primary components of military pay were the only sources of household income, modest percentages of servicemembers at Fort Drum and Fort Riley in December 2005 might have qualified for tax-credit units using the 60 percent of AMI limit even under the program's existing income definition, but much larger percentages (about 37 percent and 26 percent, respectively) would have been eligible if BAH were excluded from income determinations (see fig. 5). In contrast, almost none of the servicemembers at Fort Benning and Fort Bliss would have been eligible under the existing income definition, and modest percentages (about 14 percent and 10 percent, respectively) would have been eligible if BAH were excluded from income determinations. The variation in servicemembers' eligibility across installations reflected differences in the percentages of servicemembers in the lowest pay grades. Figure 5: Servicemembers Potentially Eligible for Tax-Credit Units at the 60 Percent of AMI limit, as of December 2005: [See PDF for image] Sources: GAO and DOD. Note: This figure reflects the eligibility of servicemembers at selected installations as of December 2005, assuming that the primary components of military pay were the only sources of household income. Our analysis does not take into account anticipated gains in the number of servicemembers at these installations. [End of figure] Although these data, which pertain to personnel already located at these installations as of December 2005, do not indicate how many incoming personnel might be eligible to live in tax-credit units, they suggest that substantial percentages of those at the rural installations might become eligible if BAH were excluded from income determinations. In light of that possibility, community officials near Fort Drum and Fort Riley stated that excluding BAH could create opportunities to use the LIHTC program. Specifically: * Community officials near Fort Drum indicated that some developers were interested in building new rental housing but faced obstacles in financing projects because of an estimated gap between current market rents, which incoming junior enlisted personnel likely could afford, and the higher rents that developers would need to charge to make new apartments financially feasible without government subsidies. The officials had been working with developers to seek financing assistance through state programs, including New York's low-income housing tax credit program, which serves households with incomes up to 90 percent of AMI. However, because the state programs are relatively small, the officials said that increasing servicemembers' eligibility for the larger federal LIHTC program would provide more financing options for developers. * Community officials near Fort Riley noted that servicemembers make up a substantial portion of the current and expected future rental market in the area, particularly in the community of Junction City just outside of the installation. They said that while some developers of tax-credit projects have expressed interest in building more units in the area, they would only do so if the pool of potential tenants included more incoming servicemembers, because the demand for additional tax-credit units among civilian families is limited. However, even if BAH were excluded from income when determining eligibility and if developers proposed building tax-credit units, LIHTC- funded development might be limited near growing installations because the state agencies that award available tax credits have a variety of: priorities.[Footnote 24] By law, each state must prepare an annual plan that identifies its criteria for distributing its allocation of credits among proposed developments. A state would have to weigh how a proposed property would address the housing needs near growing installations against the state's priorities and selection criteria. States must give preference to projects serving the lowest-income tenants and projects that would serve qualified tenants for the longest periods of time. The states' selection criteria also must include other considerations, such as tenant populations with special housing needs. For example, the priority housing needs in Kansas's plan for allocating tax credits in 2006 include projects in communities with populations of fewer than 5,000; preservation of housing with Section 8 or Section 521 project- based rental assistance; projects for special-needs populations, such as the homeless or people with disabilities; and projects whose units would offer below-market-rate rents. Projects addressing these priorities would receive extra points in the scoring process used to evaluate proposals. Furthermore, officials in the four communities described market factors that could influence whether developers would try to use the LIHTC program to build housing near growing installations.[Footnote 25] In general, developers would have limited incentive to compete for tax credits if conditions for building market-rate housing were favorable, such as in areas having a higher-income population. Generally, market- rate housing allows developers to charge whatever rents the market will bear, without other restrictions. In contrast, applicants for tax- credit financing must agree to limit the rents charged for tax-credit units for at least 30 years and must comply with other federal requirements for 15 years or risk losing the right for investors to claim the tax credits. Thus, developers might be less likely to propose new tax-credit units near a growing installation that expected to receive more senior servicemembers with relatively high incomes than near one that expected more junior members with relatively low incomes. For example: * Aside from military students who would live on base, most of the incoming military personnel at Fort Benning are associated with the planned realignment of a training school with primarily senior-ranking personnel. Community officials said that because these personnel likely could afford to pay market rates for housing, they did not expect developers to focus on providing new housing through the LIHTC program. * In contrast, on the basis of preliminary estimates from Fort Riley officials, roughly 45 percent of the servicemembers that would eventually be stationed there might be married members in pay grades E- 1 through E-6. As of early 2006, the communities near Fort Riley had substantial market-rate development under way or in the planning stages. However, community officials anticipated that enough additional low-cost housing would be needed for servicemembers in these lowest pay grades to justify building tax-credit units for them (assuming they were to become eligible). In addition, developers might be more disposed to seek LIHTC financing in areas where the cost to build new housing was high relative to the incomes of junior enlisted members. For example: * Whereas officials in the Fort Benning area expected that developers could build new market-rate housing within the price range that incoming servicemembers could afford, officials in the rural Fort Drum and Fort Riley areas stated that increasing construction, labor, and infrastructure costs could make new market-rate units too expensive for junior enlisted members or could make it difficult to secure financing for market-rate units.[Footnote 26] For example, the cost of bringing materials and, perhaps, workers into a rural area can contribute to relatively high development costs. * Near Fort Bliss, El Paso city officials said that the LIHTC program might be an attractive financing alternative for developers if they could not otherwise build housing that servicemembers with the lowest incomes could afford. However, the officials did not yet know whether developers might need subsidies. They planned to study the issue by considering the expected incomes of servicemembers who would be arriving at Fort Bliss, the supply and price of existing housing, and the development costs and rents that would be charged for new market- rate housing. Waiting Time and Scarcity of Larger Units May Preclude Servicemembers from Renting Existing Units through HUD and USDA Programs: Even if more servicemembers were to become eligible for HUD and USDA rental housing assistance programs, waiting lists for units and the limited availability of large units might limit servicemembers' participation in these programs, according to officials from HUD, USDA, and the four selected communities. Rather than financing new rental housing near growing installations, HUD's Housing Choice Voucher, public housing, and project-based Section 8 programs and USDA's Section 515 and Section 521 programs primarily would help servicemembers rent existing units if they obtained the programs' assistance, typically by making up the difference between their required contribution (generally 30 percent of adjusted monthly income) and a unit's total rent.[Footnote 27] However, these programs are not entitlements, and many of the HUD, USDA, and community officials said that the limited number of units or limited supply of rental assistance may deter eligible servicemembers from applying for these programs, especially in areas with long lists of applicants already awaiting assistance. If they did join the programs' waiting lists, servicemembers might find other private, military-owned, or privatized housing; relocate to a different installation; or become ineligible for the program because of a promotion before they rose to the top of a list. In all four of the communities we reviewed, the Housing Choice Voucher and public housing programs had waiting lists, with times ranging from a few months to 2 years, according to officials from HUD field offices and the housing authorities that maintain the lists.[Footnote 28] For example, in Columbus, Georgia, near Fort Benning, the waiting list for vouchers was long enough that it was closed as of March 2006 and was not expected to open to new applicants until 2008. In addition, servicemembers with large families may face obstacles to using rental assistance programs because of the limited availability of units with three or more bedrooms, according to some HUD, USDA, housing authority, and installation officials.[Footnote 29] In the four communities, properties with project-based Section 8 assistance and public housing developments offered relatively few units with three or more bedrooms, thereby limiting the options for families of five or more persons. For instance, in the Fort Drum area, of 690 project-based Section 8 and public housing units intended for families, 172 had three bedrooms and 43 had four bedrooms; the remaining 475 had fewer than three bedrooms. Similarly, although voucher recipients can seek housing in the broader private rental market, some of the HUD field office officials noted that larger families could have a hard time finding a sufficiently sized apartment or house that would meet the program's quality and cost standards. If servicemembers did join the programs' waiting lists, HUD headquarters and field office officials noted that housing authorities could adopt preferences that would reduce servicemembers' wait for vouchers or public housing, but some officials said that such a step could be controversial. The housing authorities that administer vouchers and public housing developments may establish local preferences for selecting families from waiting lists, on the basis of local housing needs and priorities. However, HUD and housing authority officials said that such preferences--for example, for victims of domestic violence or a single homeless person--have sometimes met opposition from those who would face longer waits because they did not qualify for these preferences. Similarly, some HUD officials said that increasing servicemembers' eligibility for the programs or giving them preference on waiting lists could create tensions with lower-income civilians who might have to face even longer waiting periods for rental assistance as a result. On the basis of the 2006 pay rates for the primary elements of military compensation, servicemembers in all pay grades would have substantially more income than most existing Housing Choice Voucher recipients, even without their BAH payments (see fig. 6).[Footnote 30] In light of that difference in incomes, some of the officials also cited potential concerns about balancing any advantages for servicemembers with the programs' current emphasis on targeting assistance to households with extremely low incomes.[Footnote 31] Figure 6: Income Ranges of Housing Choice Voucher Recipients and Servicemembers in Selected Pay Grades, 2006: [See PDF for image] Sources: GAO and DOD. [End of figure] Conclusions: As some military installations gain servicemembers, nearby communities face opportunities for growth as well as potential challenges in providing an adequate supply of housing that incoming servicemembers can afford. Many of these incoming personnel may not have problems finding housing they can afford--for example, junior enlisted members without dependents generally live in barracks; DOD has the ability to raise BAH rates for other servicemembers to reflect any increases in housing costs near the growing installations; and many servicemembers may have additional resources, such as spousal income, that they can put toward housing costs. Where communities lack enough housing for incoming personnel or where rents are expensive for married junior personnel, federal rental housing programs might help provide affordable housing for servicemembers through the production of additional housing or through rental assistance for existing housing. By excluding BAH from servicemembers' incomes when determining eligibility, many of the lowest-ranking servicemembers could qualify to apply for these programs. However, the effects of such a change are uncertain and could involve trade-offs that warrant attention. For example, the LIHTC program (or, similarly, tax-exempt multifamily housing bonds) could help increase the supply of affordable rental housing for incoming servicemembers, if more of the members were eligible to live in tax-credit units. However, even if more servicemembers were eligible, the extent to which the LIHTC program would play a role in increasing the supply of affordable housing near growing installations would depend on local housing market conditions, the income distribution of incoming servicemembers, and the decisions of state agencies regarding whether to allocate tax credits to projects near growing installations or to projects that might address other state housing priorities. Furthermore, the rental assistance programs are not entitlements and already do not assist all eligible households. While some servicemembers might be deterred by the prospect of a lengthy wait from applying for HUD and USDA rental assistance for existing units, those who did apply would expand the pool of those waiting for a limited supply of available assistance. Thus, making more servicemembers eligible by excluding BAH from income determinations could cause these programs to serve more servicemembers at the expense of eligible civilians. Matter for Congressional Consideration: If the primary intent of excluding BAH from income determinations for federal rental housing programs is to help increase the supply of rental housing that servicemembers with the lowest incomes could afford, Congress should consider first applying such a change only to programs intended to stimulate production of such housing, such as LIHTC and tax-exempt multifamily housing bonds. Agency Comments and Our Evaluation: We provided a draft of this report to DOD, HUD, IRS, Treasury, and USDA for their review and comment. Treasury and USDA did not comment on the draft report. DOD, HUD, and IRS provided technical comments, which we incorporated where appropriate. DOD also provided comments in a letter from the Acting Deputy Under Secretary for Military Personnel Policy (see app. II). DOD commented that BAH does an excellent job of achieving the objective of providing servicemembers with the same quality and quantity of housing that their civilian counterparts can afford. However, DOD also noted that servicemembers may have difficulty finding adequate housing if there are substantial changes in the supply of or demand for housing in a local area, at least until the private market has had time to adjust to the changing conditions. DOD also observed that servicemembers with large families, who seek larger housing than an average size family, may have difficulty finding adequate housing using their BAH payments alone and may apply for federal rental housing programs. However, DOD also stated that servicemembers should be eligible for federal housing subsidies under the same terms as their civilian counterparts. Furthermore, DOD commented that excluding BAH from income determinations might transfer existing scarce resources from low-income civilians to the military and generate ill-will among civilians toward the military. Finally, DOD stated that, while our draft report showed that excluding BAH from income determinations might not have the desired effect of increasing the supply of rental housing for servicemembers, there might be other ways in which the government could assist the private market in responding to housing shortages. Our draft report discussed the particular difficulties of large families-- even those receiving rental assistance--in finding suitable housing. The draft report also addressed the potential role of existing programs, particularly the LIHTC program, in stimulating production of affordable housing near growing installations. However, examining other possible federal strategies for increasing the supply of private housing was beyond the scope of this study. We are sending copies of this report to other interested congressional committees; the Secretaries of the Departments of Agriculture, Defense, Housing and Urban Development, and the Treasury; and the Commissioner of Internal Revenue. We will make copies available to others upon request. This report will also be available at no charge on GAO's Web site at [Hyperlink, http://www.gao.gov.] Please call me at (202) 512-8678 if you or your staff have any questions about this report. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. See appendix III for key contributors to this report. Signed by: David G. Wood: Director, Financial Markets and Community Investment: [End of section] Appendix I: Objectives, Scope, and Methodology: Objectives: Our objectives were to determine (1) how excluding the Basic Allowance for Housing (BAH) from income determinations would have affected the eligibility of servicemembers receiving BAH as of December 2005 and (2) programmatic and market factors that could affect eligible servicemembers' participation in the programs in selected communities gaining military personnel. Scope and Methodology: The federal rental housing programs in our scope include the Department of Housing and Urban Development's (HUD) public housing, Housing Choice Voucher, and project-based Section 8 programs; the Department of Agriculture's (USDA) Section 515 Rural Rental Housing Loans and Section 521 Rural Rental Assistance programs; and the Low-Income Housing Tax Credit (LIHTC) and tax-exempt multifamily housing bond programs, which are jointly administered by the Internal Revenue Service (IRS) of the Department of the Treasury (Treasury) and the states.[Footnote 32] To determine how excluding BAH would have affected the eligibility of active-duty servicemember households receiving BAH as of December 2005, we compared the incomes of servicemembers who received BAH in December 2005 with the federal rental housing programs' income limits in effect at that time.[Footnote 33] We obtained personnel and pay data from the Department of Defense's (DOD) Active Duty Personnel Master and Active Duty Pay files for December 2005.[Footnote 34] We analyzed data on 702,975 active-duty servicemembers who received BAH payments that month. We calculated servicemembers' annual incomes on the basis of their December 2005 payments for basic pay; BAH; Basic Allowance for Subsistence; and, where applicable, a cost-of-living adjustment for servicemembers in certain high-cost areas. We assumed that these elements of military pay were servicemembers' sole sources of household income because data on other income sources, such as spousal income, were unavailable.[Footnote 35] However, we analyzed how other sources of income might have reduced servicemembers' eligibility by calculating the median additional annual income needed before married servicemembers' families would have exceeded the income eligibility limit.[Footnote 36] To assess the reliability of the data used in our analysis, we interviewed DOD officials who were familiar with the data, reviewed relevant documentation, and tested the data for missing and apparently erroneous values. DOD provided data on 708,548 active-duty members of the Departments of the Army, Navy and Air Force and the Marine Corps. On the basis of our tests of these data, we excluded 5,573 (about 0.8 percent) records because we could not match the servicemembers' zip codes to the geographic areas for which income limits were defined, because data on family size were missing, or because anomalies in the monthly pay data prevented us from calculating an annual income amount.[Footnote 37] As a result, our servicemember population was 702,975 for this analysis. We concluded that these data were sufficiently reliable for our purposes. Nonetheless, our analysis was limited because it provided a snapshot of servicemembers' potential eligibility to apply for the programs on the basis of their incomes in December 2005. We could not predict the effects of a future change in income determinations, because potential changes in servicemembers' incomes or duty locations and annual adjustments to programs' income limits would also affect eligibility. To determine the programmatic and market factors that could affect eligible servicemembers' participation in the programs in selected communities gaining military personnel, we interviewed and reviewed relevant documentation from military installation officials; rental housing program officials (including officials from HUD and USDA field offices, public housing authorities, and state housing finance agencies); and local government or community organization representatives in four communities near installations that will gain military personnel as a result of the 2005 base realignment and closure (BRAC) process or other military initiatives. We selected Fort Benning, Georgia; Fort Bliss, Texas; Fort Drum, New York; and Fort Riley, Kansas. We selected these installations on the basis of their expected net gains of military personnel and preliminary information indicating that the surrounding communities had initiated planning to address the housing needs of incoming servicemembers.[Footnote 38] We also sought a balance between urban and rural locations. Our selection of four Army installations reflects that this service generally expected the largest personnel gains. We visited the Fort Riley, Kansas, area and contacted the other areas by telephone. We cannot generalize the information from these four installations to all installations that will gain military personnel. In addition to our local contacts, we also discussed factors that affect the use of federal rental housing programs with headquarters officials at the Army, DOD, HUD, USDA, IRS, and Treasury. We conducted our work in Washington, D.C; Arlington, Virginia; and Junction City and Manhattan, Kansas, between November 2005 and July 2006, in accordance with generally accepted government auditing standards. [End of section] Appendix II: Office of the Under Secretary of Defense: Office Of The Under Secretary Of Defense: 4000 Defense Pentagon: Washington, D.C. 20301-4000: Jul 19 2006: Mr. David G. Wood, Director, Financial Markets and Community Investments: U.S. Government Accountability Office: 441 G Street N. W. Washington, D.C. 20548. Dear Mr. Wood: This is the Department of Defense response to the GAO report 06-784, 'Rental Housing Programs: Excluding Servicemembers' Housing Allowances from Income Determinations Would Increase Eligibility, but Other Factors May Limit Program Use,' dated June 12, 2006 (GAO Code 250264). Our proposed general comments to the draft report are enclosed. My point of contact for this matter is Dr. Saul Pleeter, telephone number 703-695-9371. Although there are no recommendations for the Department, we appreciate the opportunity to respond to the report. Signed by: William J. Carr: Acting Deputy Under Secretary: (Military Personnel Policy): Enclosures: As stated: Department Of Defense Comments GAO Draft Report GAO Code 250264/GAO-06- 784: Subject: GAO Draft Report, "Rental Housing Programs: Excluding Service members' Housing Allowances from Income Determinations Would Increase Eligibility, but Other Factors May Limit Program Use," dated June 12, 2006 (GAO Code 250264/GAO-06-784): General Comments: Service members are provided a Basic Allowance for Housing (BAH) that is based upon current market rents and generates allowances equivalent to the median cost of housing in that local area. Allowances vary by pay grade and whether the member has dependents and rates are designed to provide members with the same quality and quantity of housing that their civilian counterparts can afford. In general, the BAH does an excellent job of accomplishing this objective. There are however, situations in which it is difficult for members to find adequate housing. Largely, these instances are the result of substantial changes in either the supply and/or demand for housing in a local area. An extreme example of such a change would be the destruction of the housing stock resulting from Hurricane Katrina. If a housing shortage were to 'suddenly' appear, either through a reduction in supply or an increase in demand, such as would occur from a base realignment, newly arrived members would have a difficult time finding adequate housing. BAH rates would increase with rental rates to allow members to be competitive in bidding for the scarce housing, but, practically speaking, members, as well as civilians, would have to move farther out or purchase less or more housing than they otherwise would want. This disequilibrium would continue until the market - driven by higher than market rents --has had the time to adjust. Another fundamentally different situation in which the member may have difficulty finding adequate housing occurs when a junior member with a large number of dependents looks for larger housing. Since the BAH is not a function of the number of dependents, and is based upon average size families, a member that requires a large amount of housing would probably seek housing on-base or devote an increased share of their budget to housing. A third option would be to seek assisted or low- income housing, on the same terms as his/her private sector counterpart. The difference between the two situations is that the former is driven by sudden changes in the supply and/or demand for rental housing while the latter is the result of personal decisions made by an individual. Individuals with large families and limited incomes are eligible for federal and state programs that subsidize housing and, in our view; military members should also be eligible for this subsidy under the exact same terms as their civilian counterparts. Any preference given to a military member, such as not counting a portion of the member's income, necessarily is at the expense of a civilian that is, in an income sense, more deserving. The objective of this paper, to examine the implications of excluding the BAH from income determination for federal rental housing programs is, in effect, trying to solve a problem generated by a sudden change in demand and/or supply - the former problem --by giving preference to military members to apply for federally subsidized housing - the latter problem. Not counting the BAH might inappropriately stimulate the supply of subsidized housing, transfer existing scarce resources from low-income civilians to the military and conceivably generate ill-will among civilians toward the military. A broader objective of the study should be to look at ways in which a sufficient supply of rental housing can be provided to service members. As the GAO report shows, arbitrarily reducing military incomes by excluding BAH will make more members eligible for federal rental housing programs, but may not have the desired effect of increasing the supply of such housing. While it is practically certain that the market will respond to the housing shortage, perhaps a 'better' issue to examine is the ways in which the government can assist the private market in responding to the shortages in the most effective way. Although the GAO Report did not make any recommendations, we feel it is important to make the distinction. [End of section] Appendix III: GAO Contact and Staff Acknowledgments: GAO Contact: David G. Wood, (202) 512-8678 or WoodD@gao.gov: Staff Acknowledgments: In addition to the contact named above, Steve Westley, Assistant Director; Julianne Stephens Dieterich; Alison Martin; Bettye Massenburg; John McGrail; John Mingus; Marc Molino; David Pittman; and Barbara Roesmann made key contributions to this report. (250264): FOOTNOTES [1] The LIHTC program was established in 1986 to spur the production of rental housing for lower-income households at rents they can afford. Tax-credit units are financed in part by investors who receive federal income tax credits. In exchange, the units are restricted to households with incomes below certain limits for a fixed number of years. [2] Generally, these programs make up the difference between 30 percent of tenants' adjusted incomes and an existing unit's total rent. [3] H.R. 3186, which was introduced on June 30, 2005, would exclude BAH from being treated as income for the purposes of determining eligibility for any HUD or other federal housing assistance program. [4] H.R. Conf. Rep. No. 109-307, at 131 (2005); S. Rep. No. 109-109, at 152 (2005). [5] We selected two rural (Fort Drum and Fort Riley) and two urban (Fort Benning and Fort Bliss) installations on the basis of their expected military personnel gains and indications that the surrounding communities had initiated planning to address the housing needs of incoming servicemembers. [6] Congress developed the concept of "regular military compensation" to compare military and civilian compensation. Regular military compensation includes basic pay, BAH, the Basic Allowance for Subsistence, and the federal income tax advantage that accrues because the allowances are not subject to federal income tax. The tax advantage is the additional income military personnel would have to earn to receive their current take-home pay if the allowances were taxed. Because federal rental housing programs do not consider the value of the tax advantage when determining eligibility, we did not factor it into our analysis. [7] GAO, Military Personnel: Active Duty Benefits Reflect Changing Demographics, but Opportunities Exist to Improve, GAO-02-935 (Washington, D.C.: Sept. 18, 2002). This report describes the types of military pay, allowances, bonuses, and benefits. [8] Privatized housing normally is located on military installations. Private developers own, operate, and maintain the housing. As do families who live in private housing in the local community, families in privatized housing use their BAH to pay rent and typical utility costs. Families and servicemembers without dependents in military-owned housing receive housing and utilities in lieu of BAH. [9] GAO, Military Housing: Management Issues Require Attention as Privatization Program Matures, GAO-06-438 (Washington, D.C.: Apr. 28, 2006). This report is the latest in a series of GAO reports on DOD's housing privatization program. [10] Pub. L. No. 107-107, Title XXX (Dec. 28, 2001). [11] Defense Base Closure and Realignment Commission, 2005 Defense Base Closure and Realignment Commission Report to the President (Washington, D.C.: Sept. 8, 2005). [12] Although this report focuses on the housing needs of servicemembers near growing installations, civilians would also be affected by any housing shortages or other changes in housing market conditions that resulted from installation growth. In addition, problems of housing availability could affect urban, as well as rural, communities. [13] The HUD regulation (24 C.F.R. § 5.609) implements a provision of the United States Housing Act of 1937, as amended, which broadly defines "income," for the purposes of public housing, project-based Section 8, and Housing Choice Vouchers, as income from all sources from each member of a household (42 U.S.C. § 1437 a(b)(4)). By statute, USDA's Section 515 Rural Rental Housing Loans and Section 521 Rural Rental Assistance programs and IRS's LIHTC and tax-exempt multifamily housing bond programs also follow HUD's method of calculating income (§ 515(q) of the Housing Act of 1949, as amended (42 U.S.C. § 1485(q)) and § 42(g)(4) and § 142(d)(2)(B) of the Internal Revenue Code of 1986, as amended (26 U.S.C. § 42(g)(4) and § 142(d)(2)(B))). HUD's HOME Investment Partnerships program--which provides funds to states and localities for construction or rehabilitation of affordable housing, rental assistance, and homeownership assistance--allows participating jurisdictions to use one of three income definitions, including this regulatory definition. [14] HUD annually estimates median incomes for metropolitan areas and nonmetropolitan counties. By definition, half of the households within an area have incomes below the median, and half have incomes above it. [15] Although the very low income limit (household income at or below 50 percent of AMI) generally applies to Housing Choice Voucher applicants, the program admits some new tenants with low incomes on an exception basis. However, at least 75 percent of new participants each year must have extremely low incomes. (Similarly, each year, at least 40 percent of new participants in public housing and project-based Section 8 must have extremely low incomes.) [16] U.S. Department of Housing and Urban Development, Office of Policy Development and Research, Affordable Housing Needs: A Report to Congress on the Significant Need for Housing (Washington, D.C.: December 2005). [17] We used the 50 percent of AMI limit because it generally applies to HUD's Housing Choice Voucher program and other federal rental housing programs, as well as to tax-credit properties that set aside units for households at this income level. [18] The 60 percent of AMI limit only applies to properties financed with tax-credits or tax-exempt multifamily housing bonds that set aside units for households at this income level. [19] We calculated the annual incomes of 702,975 servicemembers who were receiving BAH, using their December 2005 payments for basic pay; BAH; BAS; and, where applicable, a cost-of-living adjustment for servicemembers in high-cost areas. We compared their 2005 military pay with the program income limits that were in effect at the end of 2005. Because we did not have data on other sources of income, such as spousal pay, we assumed that the servicemembers' military pay was the only source of household income. Our analysis cannot be used to determine the specific impact of a future change in income determinations. Changes in servicemembers' incomes or duty locations and annual adjustments to programs' income limits would affect their future eligibility. [20] Figure 3 shows that some servicemembers without dependents (i.e., a family size of one) who were receiving BAH also would have been eligible for the programs. Junior enlisted members without dependents who were required to live in barracks were not included in our analysis because they did not receive BAH. [21] As previously noted, servicemembers can earn various types of special pay, allowances, and bonuses (e.g., additional pay for certain skills, such as proficiency in a foreign language). [22] Pub. L. No. 109-163 (37 U.S.C. § 403(b)). [23] Although the community officials that we contacted focused on a potential role for the LIHTC program, tax-exempt multifamily housing bonds also could be used to help finance new rental housing. Like the LIHTC program, the tax-exempt bond program requires properties to set aside a minimum percentage of units for households at the 50 percent or 60 percent of AMI limit, but it generally does not require them to restrict rents. However, properties financed with bonds may have to restrict units' rents if they also participate in other federal rental housing programs with such requirements, including the LIHTC program. [24] If a project receives 50 percent or more of its financing through tax-exempt multifamily housing bonds, it does not have to compete to receive credits through the LIHTC program, and those credits do not count against the state's annual tax-credit allocation. Such projects receive fewer credits than those that compete for an allocation. However, states can issue only a limited volume of tax-exempt bonds, so projects would still have to compete for bond financing. [25] We did not evaluate how these factors might affect the prospects for using the LIHTC program in the four communities, nor could we generalize about their applicability in other communities. Rather, this information is intended to illustrate factors that might make it more or less likely that developers would use the LIHTC program to address housing needs near growing installations. [26] We did not collect data on development costs in these communities or evaluate the feasibility of market-rate development. [27] The Section 515 Rural Rental Housing Loans program also finances new rental housing, but the loans are available only in USDA-eligible rural areas (generally towns no larger than 20,000 people), and USDA annually designates a limited number of these communities to receive loans for new construction. Because the program now finances relatively few new units, we focused on the availability of existing Section 515 units. [28] HUD and USDA offices do not maintain information on which properties in the project-based Section 8, Section 515, or Section 521 programs have waiting lists or the length of wait times. However, HUD field office officials generally described the four communities' project-based Section 8 properties for families as fully occupied or nearly so. USDA officials reported that some Section 515 properties near rural Fort Drum and Fort Riley had vacancies, while others were fully occupied and had waiting lists. [29] Installations can accommodate some servicemembers with large families who had difficulty finding suitable and affordable housing through a federal rental housing program or on their own. For example, DOD and installation officials described instances where installations had modified military-owned or privatized family housing units (e.g., merging two units). However, such units would not necessarily be readily available because of the generally limited amount of military- owned and privatized housing. [30] We used 2003 and 2004 data from HUD's Public and Indian Housing Information Center to estimate the median income of voucher recipients at the beginning of 2006. We assessed these data and concluded that they were sufficiently reliable for our purposes. [31] Each year, at least 75 percent of new Housing Choice Voucher recipients and at least 40 percent of new public housing and project- based Section 8 residents must have extremely low incomes. [32] We excluded HUD's multifamily mortgage insurance programs from our scope because they do not impose income eligibility requirements on tenants (although insured properties may also participate in other federal programs that do impose such requirements). We also excluded HUD's Section 202 Supportive Housing for the Elderly and Section 811 Supportive Housing for the Disabled programs because they generally would not be relevant to active-duty servicemembers. We also excluded USDA's small Section 538 program, which guarantees mortgages on multifamily properties in rural areas. [33] HUD publishes the 50 percent of area median income limits, which vary by location and family size. Consistent with IRS guidance for the LIHTC and tax-exempt multifamily housing bond programs, we also used these published limits to calculate the 60 percent of area median income limits. We then used zip codes and family sizes as of December 2005 to determine the applicable income limit for each servicemember. [34] DOD's Defense Manpower Data Center provided us with the personnel and pay files. [35] In addition to the military pay elements that we considered, some servicemembers also received additional military allowances, special types of pay, or bonuses. However, we excluded such payments from our analysis because we lacked a reliable method to estimate the total amount of other military pay received in 2005 using data from a single month. [36] With BAH included and excluded when determining income, we calculated the difference between servicemembers' annual incomes and both the 50 percent and 60 percent of area median income limits for all servicemembers. [37] A correction to an overpayment in a previous month can result in a negative value for an element of monthly pay. We excluded negative values because they would tend to understate servicemembers' actual annual income and, in turn, overstate their eligibility to apply for federal rental housing programs. [38] We used data on the expected net gains in military personnel from the following 2005 report that was approved by the President and Congress: Defense Base Closure and Realignment Commission, 2005 Defense Base Closure and Realignment Commission Report to the President (Washington, D.C.: Sept. 8, 2005). Forts Benning, Bliss, and Riley are among the installations gaining the most military personnel through the BRAC process. In addition, Fort Drum is gaining military personnel as a result of the Army's initiative to change its force structure. GAO's Mission: The Government Accountability Office, the investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through the Internet. GAO's Web site ( www.gao.gov ) contains abstracts and full-text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as "Today's Reports," on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select "Subscribe to e-mail alerts" under the "Order GAO Products" heading. Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability Office 441 G Street NW, Room LM Washington, D.C. 20548: To order by Phone: Voice: (202) 512-6000: TDD: (202) 512-2537: Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov Automated answering system: (800) 424-5454 or (202) 512-7470: Public Affairs: Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, D.C. 20548: