This is the accessible text file for GAO report number GAO-11-329 entitled 'Rural Housing Service: Opportunities Exist to Strengthen Farm Labor Housing Program Management and Oversight' which was released on March 30, 2011. 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. United States Government Accountability Office: GAO: Report to Congressional Committees: March 2011: Rural Housing Service: Opportunities Exist to Strengthen Farm Labor Housing Program Management and Oversight: GAO-11-329: GAO Highlights: Highlights of GAO-11-329, a report to congressional committees. Why GAO Did This Study: Congress created the Farm Labor Housing (FLH) Loan and Grant Program in the early 1960s to support the development of affordable housing for farm workers. In 2010, Congress appropriated $19.7 million for this program, which is administered by the Rural Housing Service (RHS) in the U.S. Department of Agriculture (USDA). GAO was asked to examine (1) demand for the FLH program; (2) RHS’s processes for ensuring that the program is providing decent housing for eligible farmworkers; and (3) the financial status and financial management of FLH properties. To do this work, GAO analyzed agency data, regulations, and FLH program documentation; convened a group of experts with assistance of The National Academies; selected and inspected 20 properties in five states; and interviewed RHS staff and various stakeholders. What GAO Found: Available RHS occupancy data indicate that overall demand for FLH units has remained stable in recent years—with the vacancy rate ranging from 12 to 16 percent from 2007 through 2010. But the data showed, and stakeholders GAO interviewed and other experts agreed, that rates varied significantly across states. For example, RHS occupancy data show that the vacancy rates ranged from 0 percent in South Carolina to 64 percent in Wisconsin in 2010. However, stakeholders and experts offered divergent perspectives on trends in the demand for FLH units, with some citing instances of declining demand and others suggesting that demand was still high. The experts frequently cited housing costs and eligibility requirements among the factors having the greatest impact on demand. RHS management processes have hindered the agency’s ability to assure farmworkers access to decent and safe housing and compliance with program requirements. For example, RHS cannot readily determine the severity of occurrences of noncompliance among FLH borrowers because the program information it uses to track borrower performance lacks specificity. Moreover, the enforcement mechanisms RHS uses may not be effective in bringing borrowers back into compliance in a timely manner, because some are too mild (servicing letters), and others too severe (acceleration of the loan payments) to have the intended effect. Additionally, the processes RHS has used for verifying tenant eligibility were inconsistent among states. For example, some states used third-party income and residency verification systems, while other states did not have access to or were unaware of these verification tools. Further, RHS has not analyzed all available program data to best target program funds to areas of greatest need. For example, information from program applicants has not been summarized to assess demand in a local area or state. Most FLH program borrowers were able to make timely loan payments; however, more could be done to ensure that FLH funds are used efficiently. For example, according to GAO’s analysis, RHS overestimated its credit subsidy costs for fiscal year 2010 by $3 million, and another $11.8 million in low-interest financing could have been available to loan applicants. An investigation of unusual fluctuations in the credit subsidy cost components and a greater degree of coordination by budget and program staff could have helped ensure that key assumptions, namely the predicted default rates, used in the credit subsidy model more closely reflected portfolio performance and would have allowed RHS to optimize funding use. In addition, more than $184 million in loans and grant obligations were unliquidated, or unpaid, as of September 2010 and the balance of unliquidated obligations has not changed significantly over the past 6 years. However, RHS had no guidelines in place on when to recapture these funds, making it difficult to ensure that limited program funds are used effectively by being made available to other projects in a timely manner. What GAO Recommends: GAO recommends that the Secretary of Agriculture take steps to strengthen oversight and management of the FLH program by, among other things, improving performance and financial information, increasing borrower compliance, and ensuring the efficient use of resources for the FLH program. The agency generally agreed with GAO’s recommendations. View [hyperlink, http://www.gao.gov/products/GAO-11-329] or key components. For more information, contact A. Nicole Clowers at (202) 512-8678 or clowersa@gao.gov. [End of section] Contents: Letter: Background: Limited Available Data Suggest Varying Levels of Demand for FLH Units and Funds, and Experts and Stakeholders Identified Options to Improve Program: Improvements in RHS Processes Needed to Better Manage FLH Program and Enforce Requirements: Most FLH Program Borrowers Are Not Delinquent or in Default on Their Payments, but Additional Management Attention Needed to Help Ensure Efficient Use of Funds: Conclusions: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Objectives, Scope, and Methodology: Appendix II: Experts Convened by GAO with the Assistance of The National Academies on Demand for Farm Labor Housing: Appendix III: Federal Data on Farmworkers: Appendix IV: Age of the FLH Property Portfolio and Condition of FLH Properties We Visited: Appendix V: FLH Credit Subsidy Rate Calculation: Appendix VI: Comments from the U.S. Department of Agriculture: Appendix VII: GAO Contact and Staff Acknowledgments: Tables: Table 1: RHS Performance Classification System for FLH Properties: Table 2: Age of Low Grades for All Properties in Site Visit States as of September 30, 2010: Table 3: 2010 FLH Program Characteristics for Site Visit States: Table 4: Estimated Credit Subsidy Rate for FLH Program: Figures: Figure 1: Structure of the Farm Labor Housing Loan and Grant Program: Figure 2: Number of Properties and Units by State and Housing Type, as of September 2010: Figure 3: Factors That Very Greatly Influence Demand for FLH Units, Based on Questionnaire Responses from Our Group of 11 Experts: Figure 4: Factors That Very Greatly Influence Demand for Funds to Develop Farm Labor Housing, Based on Questionnaire Responses from Our Group of 11 Experts: Figure 5: Performance Grades for FLH Borrowers from Fiscal Years 2006 through 2010: Figure 6: Percentage and Number of Properties with Delinquent Borrowers by State and Housing Type, as of September 2010: Figure 7: Open Financial Findings by Type, as of September 2010: Figure 8: Loans and Grants Obligated but Unliquidated, by Age, as of September 30, 2010: Figure 9: Age of Farm Labor Housing Program Portfolio, as of September 30, 2010: Figure 10: FLH Property in California with Reflective Roof Materials, Energy-efficient Appliances, On-demand Water Heater, and Artificial Turf: Figure 11: Newly Developed Florida Properties with Low Levels of Disrepair: Figure 12: Windows Replaced with Wooden Boards and a Kitchen in Need of Repair at an Older Florida FLH Property: Figure 13: FLH Unit in Michigan with Water Damage to the Exterior: Figure 14: Window Covered by a Board in an FLH Unit in New York: Figure 15: FLH Unit Undergoing Rehabilitation in Texas: Abbreviations: AMAS: Automated Multi-Family Housing Accounting System: CPS: Current Population Survey: FASAB: Federal Accounting Standards Advisory Board: FCRA: Federal Credit Reform Act of 1990: FLH: Farm Labor Housing: FLS: Farm Labor Survey: HUD: Department of Housing and Urban Development: MFIS: Multi-Family Housing Information System: NASS: National Agricultural Statistics Service: NAWS: National Agricultural Workers Survey: NOFA: Notice of Funding Availability: OMB: Office of Management and Budget: PHAS: Public Housing Assessment System: RD: Rural Development: RHS: Rural Housing Service: SAVE: Systematic Alien Verification for Entitlements Program: USDA: U.S. Department of Agriculture: [End of section] United States Government Accountability Office: Washington, DC 20548: March 30, 2011: The Honorable Herb Kohl Chairman: The Honorable Roy Blunt: Ranking Member: Subcommittee on Agriculture, Rural Development, Food and Drug Administration, and Related Agencies: Committee on Appropriations: United States Senate: The Honorable Jack Kingston: Chairman: The Honorable Samuel Farr: Ranking Member: Subcommittee on Agriculture, Rural Development, Food and Drug Administration, and Related Agencies: Committee on Appropriations: House of Representatives: Farmworkers play a critical role in the nation's agricultural sector. However, according to the U.S. Department of Agriculture (USDA), farmworkers frequently are among the most poorly housed people in the United States, sometimes living in tents, shacks without running water, or crowded, poorly built dormitories. To support the development of adequate, affordable housing for farmworkers, Congress enacted the Farm Labor Housing (FLH) Loan and Grant Program in the early 1960s--the loan program was enacted in 1961 and the grant program in 1964--which the Rural Housing Service (RHS) in USDA administers. This program provides capital financing to buy, develop, improve, or repair housing for domestic farmworkers employed on farms or in agricultural or processing industries off-farm. The FLH program is the only federally assisted source of housing dedicated to farm labor, which is defined as services associated with the spectrum of farming activities, from cultivating the soil to delivering commodities to market. In 2010, the FLH program had about 731 active properties with more than 16,800 units in 40 states, and a loan portfolio balance of around $300 million.[Footnote 1] Thirty-seven percent of properties and 93 percent of units were located off-farm; 61 percent of properties and 4 percent of units were located on-farm.[Footnote 2] Off-farm properties generally are developed by nonprofit organizations and public entities that assist farmworkers at off-farm locations with no requirements that workers be employed on a particular farm. On-farm properties are located on the farm where laborers work or in the nearby community. In general, on-farm housing occupants do not pay rent and utilities unless such charges are approved by the USDA. About half of all on- farm properties (51 percent) are more than 21 years old, and about half of all off-farm properties (53 percent) are from 11 to 30 years old. Thirteen states (Arizona, Arkansas, California, Colorado, Florida, Idaho, Michigan, New Mexico, New York, North Carolina, Oregon, Texas, and Washington) had more than 100 housing units. Arkansas, Michigan, and Vermont had the largest number of on-farm properties; while California, Florida, Oregon, and Washington had the largest number of off-farm properties. However, with an aging portfolio, changing labor needs in the agriculture industry, and changes in the demographics of farmworkers, the continued demand for these properties is unclear. In the conference report to the Omnibus Appropriations Act, 2009 (Public Law 111-8), Congress directed us to conduct an assessment of the properties financed under the FLH program, provide information on the physical condition, occupancy status, and financial status of the units, and discuss other management and compliance issues confronting FLH management entities. To respond to this mandate and your additional interests, we examined (1) how demand for the FLH program has changed over time, key factors that influence demand for such housing, and whether the program model addresses demand; (2) the extent to which RHS management processes assure farmworkers access to decent and safe housing and compliance with program requirements; and (3) the financial status of properties in the FLH portfolio and the extent to which RHS processes ensure the sound financial management of the program. To address these objectives, we reviewed literature and reports on farm labor housing demand; held a 1-day discussion with a group of experts with the assistance of The National Academies; and conducted interviews with academics, government entities, nonprofit organizations, and farm labor housing developers to gather perspectives on trends in farm labor housing and factors that contribute to demand.[Footnote 3] We completed site visits to five states that included interviews with USDA state and local office staff, walkthroughs of 20 properties (4 properties in each of the five states) that included inspecting the interior and exterior of the properties and speaking to the borrower or property manager, and reviews of tenant files, the results of which cannot be generalized across the portfolio.[Footnote 4] The five states were California, Florida, Michigan, New York, and Texas, which were selected to obtain regional diversity and a range in type (on-farm, off-farm and seasonal housing) and number of properties/units per state. We selected properties to include both property types (on-farm and off-farm properties) and a range of property sizes and performance grades. We present information about the age and condition of the properties we visited in appendix IV. We also obtained and analyzed electronic program data from RHS's Automated Multi-Family Housing Accounting System (AMAS) and its Multi-Family Housing Information System (MFIS). RHS uses AMAS to identify delinquencies and financially delinquent borrowers, and MFIS to identify program compliance and assess overall program needs using information on budgets, operating costs, nonfinancial defaults, insurance, reserve account funding, management plans, supervisory visits, taxes, and tenant changes. We assessed the reliability of these data by (1) performing electronic testing, (2) reviewing existing information about the data and the system that produced them, and (3) interviewing agency officials knowledgeable about the data and related management controls. Based on this assessment, we determined the data to be sufficiently reliable for the purposes of this report. We reviewed USDA and RHS handbooks, reports, and documentation of national FLH stakeholder meetings convened by USDA in 2008 and 2009. We interviewed headquarters, state, and local RHS staff knowledgeable about financial underwriting, the oversight of FLH loans and the program's credit subsidy model. Finally, we examined federal budget documentation for the program, our Standards for Internal Control in the Federal Government, and support for the program's credit subsidy calculation.[Footnote 5] We conducted this performance audit from March 2010 to March 2011 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Background: RHS administers the FLH Loan and Grant program under Section 514 and 516 of the Housing Act of 1949, as amended, which provides direct loans and grants for the development of on-farm and off-farm housing. [Footnote 6] Through the FLH program, the agency distributes capital financing annually to buy, develop, improve, or repair housing for laborers employed on farms or in associated handling or processing industries off-farm.[Footnote 7] In 2010, Congress appropriated $19,746,000 for this program. Grants for up to 90 percent of the development cost of the properties are made to farmworker associations, nonprofit organizations, Indian tribes, and public agencies. Direct loans are made for 33 years at 1 percent interest to these entities, individual farmers, associations of farmers, family farm corporations, or partnerships. Loan and grant recipients may manage the properties or contract with management agents. The RHS national office reviews state office funding requests, implements, and monitors performance measures to ensure program objectives are met, and provides authority and direction to field offices on customer service and program delivery. Although state office responsibilities may vary according to state, these offices typically accept, review, and service loans; monitor budgets; conduct fiscal and physical inspections; and engage in limited policy-making and oversight of local field offices.[Footnote 8] The state office also ranks, scores, and forwards eligible applications it receives for funding to the national office. Regional and local field offices provide day-to-day loan oversight and conduct reviews of FLH properties to ensure compliance with program rules. (See figure 1 for the organizational structure of the program.) Figure 1: Structure of the Farm Labor Housing Loan and Grant Program: [Refer to PDF for image: illustration] Top level: USDA RHS National Office: Program oversight, policy-making. Second level, reporting to USDA RHS National Office: USDA RHS State Offices: Oversight of local offices and loan servicing. Third level, reporting to USDA RHS State Offices: USDA RHS Local Offices: Day-to-day loan servicing and grant oversight. Fourth level, reporting to: USDA RHS Local Offices: On-farm properties (Section 514 loans): Eligible borrowers: Individual farmers, associations of farmers, family farm corporations or partnerships; * Seasonal units: Occupied 8 months or less per year; * Year-round units: Occupied more than 8 months per year. Off-farm properties (Section 514 loans and Section 516 grants): Eligible borrowers and grantees: Nonprofit organizations; farm labor associations; state, local or public agencies; Indian tribes; * Seasonal units: Occupied 8 months or less per year; * Year-round units: Occupied more than 8 months per year. Tenants: * Farm laborers and their families; * Must receive a substantial portion of income from primary production, processing, and transport of agricultural or aquacultural commodities; * Must be a citizen of the United States or a person legally admitted for permanent residence. Source: GAO analysis; Art Explosion (images). [End of figure] The number and type of FLH properties and units vary across states (see figure 2). States with many off-farm properties--such as California, Florida, and Texas--are often referred to as "home base states" where farmworkers live and work throughout the year. States with more on-farm housing--such as Michigan and Arkansas--tend to house workers seasonally and are often called "stream" states. Figure 2: Number of Properties and Units by State and Housing Type, as of September 2010: [Refer to PDF for image: 2 illustrated U.S. maps] Properties on-farm/properties off-farm: Total: 449 on-farm; 272 off-farm. Alabama: 3/0; Alaska: 1/0; Arizona: 5/5; Arkansas: 155/0; California: 4/91; Colorado: 0/12; Connecticut: none; Delaware: 0/2; Florida: 0/40; Georgia: 1/2; Hawaii: 1/2; Idaho: 0/10; Illinois: 3/1; Indiana: none; Iowa: 4/1; Kansas: none; Kentucky: 0/2; Louisiana: 9/1; Maine: 4/0; Maryland: 0/2; Massachusetts: 3/2; Michigan: 82/2; Minnesota: 0/3; Mississippi: 28/0; Missouri: none; Montana: 1/0; Nebraska: 1/2; Nevada: 2/0; New Hampshire: 3/0; New Jersey: 18/1; New Mexico: 0/8; New York: 17/2; North Carolina: 6/4; North Dakota: none; Ohio: 2/1; Oklahoma: 0/2; Oregon: 1/23; Pennsylvania: 0/2; Puerto Rico: 0/1; Rhode Island: none; South Carolina: 9/0; South Dakota: none; Tennessee: 9/1; Texas: 0/19; Utah: 0/2; Vermont: 71/0; Virginia: 0/1; Washington: 2/23; West Virginia: none; Wisconsin: 4/4 Wyoming: none. Units on-farm/units off-farm: Total: 1,081 on-farm; 15,772 off-farm. Alabama: 7/0; Alaska: 1/0; Arizona: 49/141; Arkansas: 233/0; California: 20/5,559; Colorado: 0/626; Connecticut: none; Delaware: 0/50; Florida: 0/4,647; Georgia: 20/48; Hawaii: 1/44; Idaho: 0/572; Illinois: 27/36; Indiana: none; Iowa: 7/4; Kansas: none; Kentucky: 0/42; Louisiana: 12/40; Maine: 4/0; Maryland: 0/90; Massachusetts: 3/48; Michigan: 320/44; Minnesota: 0/78; Mississippi: 75/0; Missouri: none; Montana: 1/0; Nebraska: 2/24; Nevada: 2/0; New Hampshire: 3/0; New Jersey: 23/24; New Mexico: 0/241; New York: 86/24; North Carolina: 32/109; North Dakota: none; Ohio: 8/24; Oklahoma: 0/42; Oregon: 1/800; Pennsylvania: 0/12; Puerto Rico: 0/24; Rhode Island: none; South Carolina: 9/0; South Dakota: none; Tennessee: 11/24; Texas: 0/1,426; Utah: 0/25; Vermont: 79/0; Virginia: 0/34; Washington: 41/799; West Virginia: none; Wisconsin: 4/63; Wyoming: none. Sources: GAO analysis of MFIS and AMAS data; map (MapInfo). Note: On-farm borrowers are not required to annually report the number of units on their properties, which may result in an underestimation of on-farm units shown above. However, RHS is required to conduct on- site supervisory reviews every 3 years of properties and units. Total includes 10 properties and 14 units that were of an unknown classification (e.g., on-farm or off-farm) or nonlabor housing properties. [End of figure] Under the FLH program, eligible tenants must receive a substantial portion of their income through the primary production of agricultural or aquacultural commodities, or those involved in off-farm handling or processing of such commodities. In addition, eligible tenants must be U.S. citizens or noncitizens with permanent residency status and program-eligible employment. According to USDA officials, applicants in off-farm properties must show documentation of their legal residency status or declare U.S. citizenship.[Footnote 9] In addition, off-farm FLH tenants must qualify as a very low-, low-, or moderate income household based on the Department of Housing and Urban Development's (HUD) income eligibility standards and provide borrowers documentation to verify their income eligibility.[Footnote 10] Although estimates of the domestic farm labor population have varied widely, according to the USDA's National Agricultural Statistics Service, the United States had more than 1 million hired farmworkers in 2010.[Footnote 11] In addition, the most recent Census of Agriculture in 2007, which provides state specific data on farmworkers, shows that six states--California, Florida, North Carolina, Oregon, Texas, and Washington--account for about 43 percent of all hired farmworkers. These data show that in the aggregate the number of hired farmworkers has remained relatively stable over the last 10 years and the geographic distribution of farmworkers has not changed significantly in the past decade. The labor market for farmwork typically includes a large population of relatively poor workers, a portion of whom migrate to, and within, the United States. Hired farmworkers also are typically young, more likely to be foreign- born, less likely to speak English, have lower levels of education, and are less likely to be U.S. citizens or to have a legally authorized work permit. According to National Agricultural Workers Survey performed by the U.S. Department of Labor, about half of all hired crop farmworkers lack legal authorization to work in the United States. Application process for off-farm and on-farm housing: Each year nonprofit organizations, state and local entities, community organizations, and federally recognized Indian tribes may submit proposals to develop off-farm labor housing. RHS state offices rank, score and forward all eligible applications to the national office for selection for funding, in accordance with national requirements outlined in an annual Notice of Funding Availability (NOFA). Off-Farm Labor Housing applications are ranked on a national basis according to the scoring awarded to each application on the state and local level. Funds are then awarded to the top-scoring applications. Once a proposal is selected for funding, applicants must submit a final application. Every year the national office notifies state offices of the deadline for submitting applications for consideration in the national funding selection process. Application deadlines, the type of funding available (such as loans versus grants) and the selection and scoring processes may vary from year to year as outlined in the NOFA. For example, in 2007, 2008, and 2010 applications received additional scoring points for participation in sustainable development and energy efficiency programs; whereas, such incentives did not previously exist. In addition, the deadline for applications has varied over the past 10 years from early May to mid-August. The NOFA may vary on whether or not funds are available that year for construction of new units only, or both for new construction and rehabilitation. Farm owners and associations of farmers who wish to provide housing to the farmworkers they employ may apply for loans for on-farm housing. Each year RHS establishes a specific allocation of funding for the development of on-farm labor housing. Interested applicants may submit relevant information to their local RHS field office to determine their likelihood for funding. Local and state offices then forward applications to the national office, which processes them on a first- come, first-served basis. Underwriting and Oversight of FLH Loans: RHS underwriting and loan oversight processes include financial analyses of applicants, annual budget reviews, the setting of reserve fund requirements, and other loan servicing activities. An RHS loan originator in a state or local office first assesses the financial feasibility of a proposed project and the financial condition of the applicant during the initial application stage.[Footnote 12] As part of these reviews, the loan originator reviews current credit reports, analyzes projected cash flows, and also confirms that the applicant has the ability to provide the required financial resources to the project--either 3 or 5 percent of the loan as equity and, if a for- profit entity, up to 2 percent initial operating capital. Finally, though RHS does not pre-determine the amount of loan versus grant funding it will obligate each year, when property cash flows are negative, the agency will consider various methods to help address the financial distress, including loan restructuring and consolidation. [Footnote 13] Once a project is approved by an RHS state director or loan approving official designated by a state director, borrowers must establish a replacement reserve account with funding levels sufficient to meet the major capital needs of a project over its life, such as replacing the roof or windows, doing major exterior work, and adding new kitchen fixtures. The aggregate, fully funded reserve amount must equal at least 10 percent of the greater of the total development cost or appraised value, and annual contributions must be a minimum of 1 percent of the total development cost. The agency requires that borrowers submit annual property budgets for approval, identify major maintenance and replacement needs during the annual budget cycle, and develop a schedule for making withdrawals from the reserve account. In the case of larger properties that have 24 units or more, borrowers must submit annual audited financial statements. RHS's Performance Management System: RHS's performance management system involves multiple monitoring activities. While local RHS loan servicers monitor FLH properties for program compliance, RHS state offices are responsible for oversight of these efforts. Local office loan servicers conduct a variety of off- site monitoring activities, or desk reviews and on-site supervisory reviews to assess whether the property is managed in accordance with FLH program objectives, the housing is decent, safe, sanitary, and affordable, and occupancy requirements are being met. The state office uses MFIS and AMAS to monitor the performance of FLH properties reported by local loan servicers. Based on these monitoring activities, RHS assigns all properties a performance grade in MFIS from A to D. See table 1 for more information on the classification system. Table 1: RHS Performance Classification System for FLH Properties: Classification designation: Class A; Description of classification: Properties have no unresolved findings or program violations. Classification designation: Class B; Description of classification: Properties for which RHS has taken servicing steps and the borrower is cooperating and has a plan to resolve identified findings or violations. Classification designation: Class C; Description of classification: Properties with identified findings or violations for which no plan has been developed to resolve the problem. Classification designation: Class D; Description of classification: Properties in monetary or nonmonetary default of the program. Properties in monetary default have mortgage payments that are 60 days overdue. Properties in nonmonetary default are those for which a loan servicer has notified the borrower of a program violation using at least three servicing letters and the borrower has not addressed the violation. Source: USDA. [End of table] Subsidy Cost of FLH Program: The Federal Credit Reform Act of 1990 (FCRA), enacted as part of the Omnibus Budget Reconciliation Act of 1990, reformed budgeting methods for federal credit programs, including RHS's farm labor housing direct loan program. As a result of FCRA, the Office of Management and Budget (OMB) requires federal agencies with credit programs to report the actual cost and estimated lifetime cost to the government of their programs in their annual budgets. Similarly, federal accounting standards require agencies to recognize the estimated lifetime cost of their programs in their financial statements. To determine the expected cost of credit programs, agencies predict or estimate the future performance of the programs on a cohort basis.[Footnote 14] This cost, known as the credit subsidy cost, is the net present value of estimated payments the government makes less estimated amounts it receives over the life of the direct loan or loan guarantee, excluding administrative costs.[Footnote 15] OMB also requires federal agencies with credit programs to reestimate subsidy costs annually to reflect actual loan performance and expected changes in estimates of future loan performance. Annual estimates of a program's expected lifetime subsidy cost changes from year to year. Each additional year provides more historical data on loan performance that may influence estimates of the amount and timing of future claims and prepayments, as well as changes in estimation methodology may cause changes in subsidy cost. Limited Available Data Suggest Varying Levels of Demand for FLH Units and Funds, and Experts and Stakeholders Identified Options to Improve Program: Available RHS Data and Stakeholder Perspectives Suggest Demand for Farm Labor Housing Has Varied: Although RHS maintains some program data through AMAS and MFIS--the agency's accounting system and program management information system-- it does not maintain or collect comprehensive data that could be used to better assess demand for farm labor housing. Demand can be measured from two perspectives: (1) demand from prospective tenants to occupy farm labor housing units and (2) demand for the funds to develop the units. Information needed to fully understand demand for FLH units could include occupancy rates, tenant applications, and waitlists for units. However, RHS's national office does not retain electronic data on tenant applications for units or waitlists for the units. Demand for funds can be measured by the number of applications submitted by potential borrowers, but RHS's national office does not retain electronic data on borrowers' applications after a funding round is completed. Therefore, it is difficult to assess demand for program funds on a state, regional, or nationwide basis.[Footnote 16] Given these limitations, we reviewed RHS's data on occupancy and anecdotal information from stakeholders and experts to describe demand for FLH units and funds. Available data on occupancy suggest that demand from tenants to occupy units has remained stable in recent years but varies significantly among states. RHS collects data on occupancy for each property in the portfolio and uses vacancy rates to track occupancy levels of properties. RHS occupancy data show that vacancy rates ranged from 0 percent in South Carolina to 64 percent in Wisconsin in 2010. These data also show that from 2007 to 2010, vacancy rates have remained relatively stable, with the average vacancy rate ranging from 12 to 16 percent during this time. In 2010, among the 13 states with more than 100 units, 8 states had average vacancy rates of under 10 percent and 11 of the 12 had rates under 20 percent. Although overall demand appears to have remained stable in recent years, the occupancy data also suggest that demand varies regionally and by state. Vacancy rates in states with the most FLH units, such as California, Florida, and Texas, ranged from 14 to 18 percent in 2010. Most states to the south, such as Arkansas, Arizona, Georgia, Louisiana, Mississippi, North Carolina, New Mexico, South Carolina, and Virginia, also had low vacancy rates--below 10 percent in 2010. In contrast, some states with shorter growing seasons such as Colorado, Minnesota, and Wisconsin had FLH vacancy rates above 50 percent in 2010. However, some states to the north, such as Massachusetts, Michigan, New York, Oregon, and Washington, had vacancy rates under 10 percent in 2010. RHS officials noted that seasonal units often have higher average vacancy rates, because these units are unoccupied for part of the year, compared to year-round units, which may impact a state's overall vacancy rate. Stakeholders we interviewed in the course of our audit work and the group of experts who participated in our 1-day group discussion, convened with the assistance of The National Academies in October 2010 also indicated that demand for units varies significantly across the country. They noted that farmworker populations have heavier concentrations in the home base states of California, Florida, and Texas, which have longer growing seasons and thus have a more consistent need for housing.[Footnote 17] However, they also offered contrasting examples of demand even among states with heavy concentrations of farmworkers. For example, stakeholders we interviewed and an expert indicated that tenant demand for housing among farmworkers has remained high, with the expert noting that farmworkers in California increasingly have been living in informal dwellings such as garages, sheds, and trailers because they lack other options. In contrast, a nonprofit organization, property managers, and RHS officials cited declining demand for units in Florida. For example, 7 of the 40 FLH properties in Florida have obtained waivers to rent to otherwise ineligible tenants (for example, tenants who are not employed in domestic farm labor) due to diminished demand. Stakeholders we interviewed and the experts selected for our 1-day group discussion offered divergent perspectives on trends in the demand for FLH units. RHS officials and experts said they have witnessed a decline in demand, anecdotally reporting that waitlists for FLH units have declined in recent years in several states. For example, during our expert group discussion, one expert stated that over the past 10 years waitlists at his FLH properties had declined, whereas in the past his FLH properties were consistently fully occupied and regularly had waitlists that were 100 households long. He also noted that waitlists continue to exist for non-FLH housing. Another expert stated that it often takes up to 6 months to fully rent FLH housing in his state, while housing that is not financed through the FLH program can be fully occupied within 1 day. In addition, one FLH property manager noted that six FLH properties have requested that RHS allow otherwise ineligible tenants who do not meet FLH requirements to rent the units due to a high level of vacancies in the FLH housing. In contrast, other experts in our group and stakeholders we interviewed suggested that there is a high demand for tenants to occupy FLH units and that the program does not fully meet this demand. In response to a questionnaire at our 1-day group discussion, all of the experts responded that overall the program did not meet demand among current and prospective tenants. For example, one expert in our group noted that in her state, changes in agriculture have had no significant impact on demand since the unmet need for housing among farmworkers far outstrips supply. Such differences in perspectives on trends in demand for FLH units may be explained by the experts' comments that sustained funding now and in the future is critical to meet the housing needs in the agriculture sector. Participants at a nationwide FLH stakeholder meeting convened by USDA in November 2009 also cited a critical need for housing to support farmwork. Overall demand for funds to develop or remodel FLH housing also varies. According to experts in our group, stakeholders we interviewed, and a study by the Housing Assistance Council, demand for funds to develop on-farm properties had decreased, while demand for funds to develop off-farm properties had increased.[Footnote 18] Experts attributed the lower demand for funds for on-farm properties to reluctance among individual growers, particularly those with small operations, to provide housing for workers because of uncertain economic conditions and being subject to program requirements and restrictions. In addition, demand for funds may vary according to geographic region. For example, stakeholders from nonprofits and developers and an expert reported that demand for FLH funds in California has increased as indicated by a high number of applications. In contrast, RHS officials, a nonprofit organization, and a developer in Florida indicated demand for FLH funds has decreased in certain areas, with fewer applications submitted for new property development. Furthermore, demand for the type of funds--off- farm versus on-farm--vary according to region or state. For example, almost all FLH properties in Michigan are on-farm, while states such as Florida and Texas currently have no on-farm properties. Housing Costs, Eligibility Requirements, and Availability of Rental Assistance Funding Frequently Cited among Other Factors Influencing Demand for Units and Funds: Experts in our group cited a number of factors that influence demand for FLH units. Among the various factors, experts identified the cost of the housing, program requirements for legal residency, and the availability of employment opportunities as having the greatest impact on tenant demand (see figure 3). Figure 3: Factors That Very Greatly Influence Demand for FLH Units, Based on Questionnaire Responses from Our Group of 11 Experts: [Refer to PDF for image: illustrated table] Factor: Cost of housing (e.g., monthly rent, weekly rent, and/or utility cost); Number of very great influence responses: 10. Factor: FLH program eligibility requirements for citizenship or permanent residency status; Number of very great influence responses: 9. Factor: Availability of employment opportunities in agriculture and processing; Number of very great influence responses: 6. Factor: Availability of affordable housing among farm worker populations; Number of very great influence responses: 4. Factor: Changes in agriculture (including increasing mechanization, changes in technology, or changes in types of crops/commodities produced); Number of very great influence responses: 2. Factor: Physical condition of housing available (i.e., property maintenance, sanitation, age of properties, amenities); Number of very great influence responses: 2. Factor: Factors associated with geographic region (including length of growing seasons, migration patterns, or weather patterns); Number of very great influence responses: 1. Factor: Type of housing available (seasonal or year-round); Number of very great influence responses: 1. Factor: FLH program eligibility requirements for income derived from farm labor; Number of very great influence responses: 1. Factor: Proximity of housing to work site; Number of very great influence responses: 0. Factor: Proximity of housing to area services (such as schools, shopping, childcare, etc.); Number of very great influence responses: 0. Factor: Availability of transportation from housing to work site; Number of very great influence responses: 0. Source: GAO analysis of experts’ questionnaire responses. [End of figure] * Cost of housing: Ten of the 11 experts in our group identified the cost of housing--monthly or weekly rents and utility costs--as very greatly influencing the demand for FLH units among prospective tenants. According to a USDA report, farmworkers tend to earn very low wages; therefore, prospective tenants may choose cheaper options, such as sharing a room with other workers, over FLH units.[Footnote 19] * Requirements for legal residency: According to nine experts in our group the program requirement that prospective tenants be U.S. citizens or document their permanent residency status limits the number of applicants for FLH units in properties around the country. Stakeholders and experts also cited the FLH residency requirement as a cause of declining waitlists to occupy FLH units. According to one borrower, farmworkers may not feel safe or comfortable living in FLH properties, even when the leaseholders can show legal residency status, because other members of the household or extended family members may not have documented legal residency. During our 1-day expert group discussion, many experts agreed that FLH demand was lower than it could be due to the program's residency requirements. Stakeholders we interviewed noted that other federally assisted housing programs--including RHS's Section 515 Rural Rental Housing program and low-income housing tax credits--do not require documentation of legal residency status. * Availability of employment opportunities: In addition, six experts in our group indicated that the availability of employment opportunities in agriculture and food processing very greatly influences the demand for FLH units among prospective tenants. Since many farmworkers travel to find employment, housing needs often are determined by the prevalence and length of available work. Changing patterns of agricultural production or the amount of work available in a given location may change the patterns of demand among farmworkers to occupy FLH units. For example, RHS officials noted that FLH properties around Orlando, Florida, at one time were located near orange groves. But due to increased housing development and urbanization, agriculture now plays a diminished role in the area and demand for farm labor housing decreased. Experts in our group discussion also cited a number of factors that influence demand for funds to develop FLH units. The factors they most frequently identified as having the greatest impact include the availability of Section 521 Rental Assistance, the level of community support for properties, and opportunities to leverage other sources of funds (see figure 4). Figure 4: Factors That Very Greatly Influence Demand for Funds to Develop Farm Labor Housing, Based on Questionnaire Responses from Our Group of 11 Experts: [Refer to PDF for image: illustrated table] Factor: Availability of Section 521 Rural Rental Assistance funding to subsidize rent costs in FLH program units; Number of very great influence responses: 8. Factor: Level of community support for developing FLH program units, including NIMBY issues[A]; Number of very great influence responses: 6. Factor: Availability of other funds or ability to leverage other funding with FLH program funds; Number of very great influence responses: 5. Factor: Availability of FLH program funding (i.e., amount of funds available and level of competition for funds during an award year); Number of very great influence responses: 3. Factor: Financial challenges associated with maintaining seasonal migrant properties year-round (financial challenges may include maintaining needed cash flow to operate the property); Number of very great influence responses: 3. Factor: Level of responsiveness, efficiency, and ease of working with USDA Rural Development staff; Number of very great influence responses: 3. Factor: Preference among nonprofits and borrowers/developers to devote resources to other affordable housing programs instead of the FLH program (e.g., tax credits, HUD’s Home Ownership Investment Partnership Program (HOME), Community Development Block Grants (CDBG), Section 515 Rural Rental Housing loans); Number of very great influence responses: 3. Factor: FLH program management challenges (including complexity of program requirements, regulations or application process, or requirements of managing FLH program properties in compliance with program requirements); Number of very great influence responses: 3. Factor: Management challenges associated with maintaining seasonal migrant properties year-round (management challenges may include maintenance, annual rental process sometimes called “rent up”, or other management issues associated with seasonal housing management); Number of very great influence responses: 2. Factor: Level of eligible tenant demand–size of farmworker population in need of housing and eligible to live in FLH program units; Number of very great influence responses: 2. Factor: Level of USDA RHS outreach and availability of education on FLH program to developers and growers; Number of very great influence responses: 2. Factor: Availability of support, technical expertise, or capacity among non-profit organizations to assist with the development of properties; Number of very great influence responses: 2. Factor: Cost to assemble an application, including pre-application development costs; Number of very great influence responses: 1. Factor: General economic conditions or concerns; Number of very great influence responses: 1. Factor: Changes in agriculture (including increasing mechanization, changes in technology, or changes in types of crops/commodities produced); Number of very great influence responses: 0. Factor: Housing site proximity to area services (such as schools, shopping, childcare, etc.); Number of very great influence responses: 0. Factor: Housing site proximity to work site(s); Number of very great influence responses: 0. Source: GAO analysis of experts’ questionnaire responses. [A] NIMBY means not in my backyard and is a term commonly used to describe community objections to projects such as low-income housing. [End of figure] * Availability of Section 521 Rental Assistance: Eight of 11 experts in our group identified the availability of Section 521 Rental Assistance funding to subsidize the cost of rent as a factor that very greatly influences the demand for FLH funds.[Footnote 20] USDA generally has limited funds for rental assistance that can be allocated to new or existing FLH properties. An expert in our group noted that rental assistance has not always been available, and it was not available during fiscal year 2008 for newly constructed FLH units. In addition, while approximately 64 percent of off-farm properties receive this subsidy from RHS, seasonal properties for migrant workers may not use rental assistance as an operating subsidy to help fund units when they are not occupied by the workers. However, RHS officials noted that two demonstration programs are currently underway that allow a select number of FLH seasonal properties to use rental assistance as an operating subsidy. Participants at a nationwide stakeholder meeting convened by USDA in 2009 and several experts in our group suggested that RHS commit to using rental assistance for operating assistance in seasonally operated housing facilities for migrant farmworkers. In addition, if a tenant who receives rental assistance becomes ineligible through an increase in income, that rental assistance subsidy may be recaptured. Since rental assistance subsidies are sometimes tied to individual units, future tenants may no longer receive rental assistance in that unit and property owners report difficulties renting out units without rental assistance. One USDA official who participated in the 2009 stakeholders meeting noted that, when rental assistance is taken from a property, it can cause long-term financial hardship. * Level of community support: Six experts in our group identified the level of community support for developing FLH units as very greatly influencing the demand for funds to develop farm labor housing. Experts noted the presence of organized community objections to FLH developments, often referred to as "not in my back yard" or NIMBY. Experts from our group and stakeholders we interviewed noted that community bias toward migrant farmworkers can pose major roadblocks to the success of development properties. One expert actively encouraged USDA to proactively take a stance against NIMBY and consider making its other agency services or funding contingent upon its ability to meet housing needs locally. * Opportunities to leverage other sources of funds: Five experts indicated that the ability to leverage other sources of funding very greatly influences demand for funds to develop FLH housing. Because housing developers often leverage multiple sources of funding for property development, such as combining federal low-income housing tax credits with FLH funds, the financial viability of a project may rest on the ability to obtain multiple funding sources. During a 2009 FLH nationwide stakeholder meeting convened by USDA, staff from nonprofit organizations and farm labor housing developers emphasized the need for increased assistance from RHS and more RHS staff experienced with leveraging other sources of funding with FLH program monies. Experts and Stakeholders Noted Existing Program Was Useful and Necessary but Provided Opinions on Changes to the Program: Stakeholders we interviewed and experts in our group noted that they believe the existing program is useful and should not be eliminated; however, many provided opinions on possible changes to the program that would better meet demand. When asked whether to eliminate, radically redesign, or relocate the program's oversight to another agency such as HUD, almost all members of our group of experts agreed that the program should be preserved in its current form as a loan and grant mechanism run out of USDA but undergo significant reform to better meet demand. However, one expert disagreed with the rest of the group, arguing that the program was incapable of meeting demand for farm labor housing and should be replaced with an alternative model. During the expert group discussion, experts suggested many program changes, which included reforming RHS's overall management approach to the program; conducting a comprehensive needs assessment of the current program to inform policy and regulatory changes; better targeting program funds to areas of greatest need; and considering innovative housing designs to lower the cost of housing units. For example, many experts discussed the importance of increasing RHS management capacity, such as the number of staff and years of experience, at the national and state level, as well as prioritizing and emphasizing the program within USDA by increasing commitment to and awareness of the program among staff. During the expert group discussion, several experts noted that the FLH program does not receive appropriate care or attention from the national office. Experts noted shortcomings in leadership from headquarters, staff capacity, and training. Currently, the national RHS office has five specialists, who help track servicing efforts for the entire multifamily housing portfolio, of which FLH properties account for approximately 5 percent. Each specialist is responsible for oversight and guidance of 9 to 10 state offices and each state's local offices. In addition, one financial and loan analyst in the national office works on FLH loan and grant making activities, and each of the three team leaders reviews the underwriting of all multifamily housing loans and grants in 14 to 21 states, which may include FLH loans and grants. Experts in our group also discussed the importance of conducting a comprehensive needs assessment of the current program to inform policy and regulatory changes. Specifically, experts noted that a lack of information on farmworkers, housing needs of farmworkers, and agricultural patterns of production inhibits RHS's ability to effectively assess demand for the program and target resources. Stakeholders we interviewed and studies from our literature review also described a lack of available data on farm labor housing demand. Similarly, during stakeholder meetings for organizations and individuals involved with the FLH program that USDA convened in 2008 and 2009, participants recommended that USDA conduct a needs assessment of the year-round, seasonal, and migrant workforce and travel patterns, which would help USDA set policy by utilizing information on trends in agricultural production needs and concentrations of the agricultural workforce. During the 2009 stakeholders meeting, officials underscored the importance of better understanding the target population of the FLH program to set policy by obtaining data on the industry and utilizing state RHS directors as a resource for information. Stakeholders we interviewed and experts in our group also noted the importance of targeting program funds to areas of greatest need. Further, several experts in our group discussed the potential benefits of considering innovative housing designs to lower housing costs. For example, experts noted that modular and manufactured housing may expand production and decrease costs per unit. Others suggested that USDA make available innovative pre-approved housing designs plans to borrowers. However, two experts expressed skepticism about manufactured housing and pre-approved housing designs, noting that these methods have not lowered costs in other programs. Participants at a national stakeholder meeting convened by USDA in 2008 had a similar discussion on alternative housing models to better serve farmworkers and lower costs. For example, participants at this conference discussed the costs and benefits of using temporary housing standards, dormitory housing, and temporary emergency housing. Participants noted that providing housing for the migrant agricultural workforce that is employed from 2 to 3 months of the year in a given location may require a different approach. Such housing has limited use during the off season and may not be cost-effective to maintain or develop. Improvements in RHS Processes Needed to Better Manage FLH Program and Enforce Requirements: Deficiencies in RHS management processes have limited the extent to which the agency can determine whether farmworkers have access to decent and safe housing and ensure compliance with program requirements. Numerous occurrences of noncompliance among FLH borrowers can be identified in RHS's performance management system, MFIS, and the proportion of compliance problems has grown over the last 5 years. However, RHS cannot readily determine the severity of noncompliance among FLH borrowers because the program information it uses to track borrower performance lacks specificity. Moreover, RHS has used limited enforcement mechanisms to address the full range of performance problems among its borrowers because its penalties either are too mild or too severe to be applied to many types of noncompliance. Additionally, the processes RHS uses for verifying tenant eligibility are inconsistent across states, as some states have access to third-party income and residency verification while others do not or are unaware of the verification tools. Finally, RHS does not analyze all available program data to best target program funds to areas of greatest need. For example, information from program applicants is not summarized to assess demand in a local area or state. Low Performance Grades Increased in the Past 5 Years, but RHS Did Not Always Resolve Problems in a Timely Manner and Could Not Readily Assess Their Severity: In the past 5 years, the proportion of FLH borrowers with low performance grades has grown to half of the overall borrower portfolio. RHS uses multiple methods to measure program performance including assigning a performance grade (grade of A to D) to each borrower based on documented physical, financial or management problems with the FLH property. From 2006 through 2009, RHS had more class A and B grades associated with FLH properties than C and D grades. However, the proportion of class C and D properties grew from 40 percent in 2009 to 50 percent of the portfolio in 2010 (see figure 5).[Footnote 21] Figure 5: Performance Grades for FLH Borrowers from Fiscal Years 2006 through 2010: [Refer to PDF for image: vertical bar graph] Year: 2006; Grade A/B: 377; Grade C/D: 260; Total projects: 637. Year: 2007; Grade A/B: 383; Grade C/D: 280; Total projects: 663. Year: 2008; Grade A/B: 419; Grade C/D: 267; Total projects: 686. Year: 2009; Grade A/B: 427; Grade C/D: 281; Total projects: 708. Year: 2010; Grade A/B: 365; Grade C/D: 366; Total projects: 731. Source: GAO analysis of MFIS data. [End of figure] From 2006 through 2010, performance grades varied among 13 states with larger FLH portfolios--states with more than 100 units.[Footnote 22] For example, three states (California, New Mexico, and North Carolina) reflected the national average, with about half of FLH borrower or grantees in those states receiving Cs. In contrast, the proportion of Michigan and Texas borrowers to receive a C or D grade was above 80 percent in 2010. Furthermore, in Colorado, the proportion of borrowers with C and D grades was 58 percent in 2006 but grew dramatically to 100 percent with a C grade in 2010. Within the FLH program, a number of properties maintained C or D grades with unresolved findings and violations for multiple years (see table 2). For example, in Michigan, 19 out of 85 properties have been graded as a C or D for more than 4 years. In one case, a Michigan borrower received a C grade 10 years ago, which has yet to be resolved. According to loan servicers in two states, borrowers (particularly for on-farm properties) have disregarded notices from RHS that they are out of compliance for multiple years. Table 2: Age of Low Grades for All Properties in Site Visit States as of September 30, 2010: Site visit state: California (95 total properties); Properties by number of years of C or D performance classification: 2 years or less: 34; 3 or 4 years: 5; 5 to 8 years: 4; 9 years or more: 0. Site visit state: Florida (40 total properties); Properties by number of years of C or D performance classification: 2 years or less: 22; 3 or 4 years: 1; 5 to 8 years: 0; 9 years or more: 1. Site visit state: Michigan (85 total properties); Properties by number of years of C or D performance classification: 2 years or less: 48; 3 or 4 years: 6; 5 to 8 years: 18; 9 years or more: 1. Site visit state: New York (20 total properties); Properties by number of years of C or D performance classification: 2 years or less: 9; 3 or 4 years: 1; 5 to 8 years: 2; 9 years or more: 0. Site visit state: Texas (19 total properties); Properties by number of years of C or D performance classification: 2 years or less: 12; 3 or 4 years: 4; 5 to 8 years: 1; 9 years or more: 0. Site visit state: Total; Properties by number of years of C or D performance classification: 2 years or less: 125; 3 or 4 years: 17; 5 to 8 years: 25; 9 years or more: 2. Source: GAO analysis of MFIS data. Note: Total row does not refer to the total number of properties or units we visited. Properties include multiple units. [End of table] According to our standards for internal control, an important aspect of a program's internal control includes monitoring the results of reviews.[Footnote 23] Monitoring of internal control should include policies and procedures for ensuring that the findings of audits and other reviews are promptly resolved.[Footnote 24] However, the Multi- Family Housing Asset Management Handbook does not designate a length of time to resolve underlying findings to improve a C or D grade. The handbook states that the loan servicer, state office, and national office should be available to provide further oversight of a borrower with a D grade and that loan servicers should be concerned when findings or violations resulting in a C grade continue for an extended period of time with no indication of resolution efforts. Furthermore, although RHS assigns a performance grade to each property, the agency's performance classification and information systems do not readily provide agency officials with information necessary to assess the causes of low performance by borrowers or the severity of findings. Specifically, RHS's classification system does not provide specific information on performance problems, such as severity or type, associated with FLH properties. Performance classifications are listed in the RHS system as a grade, but the grade does not specifically indicate its cause. For example, a D grade could be due to a long-standing failure of the borrower to submit required paperwork or a more severe finding or violation such as a health and safety problem with the physical condition of the property. An RHS official noted that the causes of C grades, in particular, can vary widely in severity. According to this official, RHS is considering methods to better specify of the severity of the grade for properties graded C. In 2010, 46 percent of FLH properties received a C grade. During site visits, local office staff explained that they submit servicing information into MFIS--such as the results of RHS visits to the property or a finding that required documentation from the borrower is missing. Although they document physical, financial, or management problems with the property as supervisory findings, it is not readily clear which specific findings yielded poor performance grades and the extent to which each finding affected the final performance grade. Moreover, performance grades may not always match the most current information gathered by local office loan servicers. According to loan servicers in two states we visited, documentation of monitoring activities was not always entered into MFIS in a timely manner due to reportedly large workloads. For example, according to one local office the completion and results of supervisory visits often were not entered into MFIS for 6 or more months following the review. Furthermore, in 2009, RHS determined through an internal review that MFIS had not been maintained or updated in a timely manner.[Footnote 25] Therefore, current information is not always available on individual property or overall portfolio performance. According to our standards for internal control, program information should be recorded and communicated to management in a form and within a time frame that enables them to carry out their internal control and other responsibilities.[Footnote 26] Relevant, reliable, and timely information should be available for management decision-making and to identify risks and problem areas in a program. To supplement the performance grades, RHS officials told us that they use performance information that local staff enter into MFIS. Specifically, RHS national office staff told us that they use the findings report in MFIS to track existing and ongoing findings related to physical or management problems within the portfolio and coordinate with state and local offices to resolve existing findings. As of the end of fiscal year 2010, RHS reported 306 out of 1,460 total open findings in the FLH portfolio, or 21 percent of all open findings, as physical condition findings. However, the findings report generated from MFIS, often does not detail the type and severity of these problems. The report lists the property and borrower identification numbers, type of housing, RHS loan servicer in charge of the property, and finding type, among other details. It also includes a category that provides a brief description of the finding titled "Finding Name." In the finding name field, short descriptions such as "lighting," "flooring," "foundation," "smoke alarms," and "windows, doors, and external structure" are listed. For some findings, loan servicers have used the comment field to better document the specific physical condition problem and its severity. For other physical findings, the comment fields do not contain detailed descriptions of the finding and, therefore, limit the extent to which management can understand of the severity of the problem. Thirty-eight percent of all open physical findings as of July 2010 had no additional comments to expand on the limited description of the physical condition problem at the property. Our site visits to 20 FLH properties in five states that included on- farm, off-farm, and year-round and seasonally occupied properties illustrate the limitations of using performance data from RHS's classification and information systems. We selected properties, in part, based on their performance grades--with the goal of selecting properties with varying performance, including some that were in poor physical condition. However, most of these properties selected and visited generally met the standards of decent, safe, and sanitary housing--even some properties classified as Cs or Ds--with low levels of disrepair often observed on interiors and exteriors of the units. Using RHS's regulations--which outline physical condition standards FLH properties--as the baseline, we evaluated the physical condition of each property visited. We found differences in the quality of the units but such differences were often related to the age of the units, age of renovation, or maintenance of the unit by the tenant. For example, in two newly renovated properties in Texas the units had central air conditioning, while one property that had not been renovated lacked this amenity. For 17 of the 20 properties we visited, we noted low or medium levels of disrepair or deficiencies; for the other remaining 3 properties, we noted high levels of disrepair. [Footnote 27] RHS's performance classification system differs from HUD's Public Housing Assessment System (PHAS), which applies performance scores to housing agencies that own and manage public housing properties. According to the regulation that governs PHAS, the purpose of the system is to score four aspects of a housing agency's performance: the physical condition of public housing properties, the financial condition of the housing agency, the management operations of the housing agency, and the housing agency's performance with respect to the obligation and expenditure of Capital Fund program grants. [Footnote 28] To measure management performance, HUD assigns a score for each of these four aspects of a housing agency's operations.[Footnote 29] In contrast to RHS's performance classification system, HUD's PHAS and performance score, on its face, specifies performance areas of concern. Because the methods RHS uses to measure and oversee FHL borrower performance do not provide sufficiently specific information, RHS cannot readily determine the severity of occurrences of noncompliance among borrowers using its classification system and findings report and it has been limited in the extent to which it can understand overall portfolio performance. FLH Program Uses Only Mild or Severe Enforcement Mechanisms to Address a Range of Noncompliance Issues: Throughout the FLH program, RHS loan servicers have used limited enforcement mechanisms to compel noncompliant borrowers to address program findings or violations. Federal regulations and program guidance set forth specific enforcement actions available for use in the FLH program to address borrower noncompliance. These include the following: 1. issuing servicing letters to notify borrowers that they are out of program compliance, 2. accelerating a borrowers mortgage to make all outstanding borrower debts to the agency immediately payable, 3. transferring ownership of an FLH property from one borrower to another, 4. suspending a borrower from participation in federal programs, 5. debarring a borrower from participation in federal programs, 6. assessing civil money penalties, and: 7. civil or criminal sanctions. However, loan servicers told us that the primary enforcement mechanisms they used included sending servicing letters to the borrower to serve notice of noncompliance or accelerating the borrower's mortgage payments, which could lead to foreclosure. In the past 5 years, RHS has issued 145 servicing letters and accelerated 17 mortgages. The enforcement mechanisms used by loan servicers are either too mild or too severe responses to noncompliance. Civil money penalties, which RHS has the authority to use, can be tailored to the level of severity of noncompliance, but the agency does not use this enforcement mechanism. According to RHS officials, USDA's general counsel determined that there was not sufficient detail in the regulation to enforce civil money penalties, and recommended that the regulation section be revised to specify step-by-step procedures for enforcement and identify who would arbitrate cases involving civil money penalties. However, to date, the regulations have not been revised, and RHS does not currently use its authority to impose civil money penalties. Forms of borrower noncompliance vary in severity. More severe forms of noncompliance include mortgage default or health and safety violations on a property. However, other forms of borrower noncompliance are less severe. For example, a common form of noncompliance among borrowers is failure to submit annual budget documentation. Although local servicers we spoke to said that borrowers should submit annual budgets, as required, they did not believe failure to do so warranted a severe enforcement mechanism such as the acceleration of mortgage payments. On the other hand, loans servicers noted sending servicing letters does not often result in borrower compliance because there are no appropriately tailored penalties associated with such letters. Consequently, FLH properties often retain low performance classifications for multiple years because they are out of program compliance for an extended period of time. For example, loan servicers in two states stated borrowers in their portfolios have remained out of compliance for multiple years because of limited enforcement mechanisms. We have previously reported that penalties in federal award programs should correspond to performance.[Footnote 30] Specifically, penalties can sometimes be too mild to discourage violations or perceived as too severe to invoke, as is the case with the primary enforcement mechanisms used in the FLH program.[Footnote 31] In the past, we have also reported that penalties may lose their effectiveness and credibility over time if they are not executed consistently.[Footnote 32] Inconsistencies Exist among Methods for Verifying Legal Residency and Income: As part of the application process, borrowers must verify the legal residency of tenants, as only U.S. citizens or permanent residents are eligible for FLH units. The methods RHS uses for ensuring that borrowers (or their designated management agents) verify the legal residency status of tenants differ across states. In our review of 111 tenant files in five states, we found 63 files without legal permanent residency documentation. However, according to some loan servicers with whom we spoke, documentation of legal status is only required of tenants who are not U.S. citizens. Therefore, missing permanent residency documentation does not necessarily indicate that the tenant is ineligible to rent in FLH housing. For files with permanent residency documentation, we examined the documentation to determine if it appeared to be questionable or inconsistent with legally issued permanent resident cards. Five of the 43 permanent resident cards that we reviewed either were expired or appeared questionable.[Footnote 33] The U.S. Citizenship and Immigration Services division of the Department of Homeland Security provides an online process for verifying the legitimacy of legal residency documentation called the Systematic Alien Verification for Entitlements (SAVE) program. This system is available, upon request, to all federal, state, and local benefit-granting government agencies. One local RHS loan servicer in California described using this system when conducting FLH tenant file reviews in preparation for triennial supervisory visits. Staff from two other RHS offices also described using SAVE. However, when asked, officials from the other seven offices that service FLH loans we visited around the country stated that they either did not use the SAVE system or were not aware that the system existed. An RHS official in the national office stated that access to the SAVE system would help RHS verify that tenants have eligible residency status and be useful tool for local offices. According to this official, SAVE use has not yet become an FLH requirement because RHS is not certain that all offices have the technical capacity to access it. In addition to verifying the legal residency of prospective tenants, loan servicers must ensure that borrowers (or their management agents) verify tenants' income levels during the application process. Some loan servicers have access to and use state wage matching systems to verify income levels of FLH tenants, but loan servicers in other states do not have access to these systems. For example, the loan servicers with whom we spoke in California described using a wage matching system administered by the California Employment Development Department when completing tenant file reviews. In New York, loan servicers stated that they did not have access to a state wage matching tool but would like to have such a method to verify the income amounts tenants report. In 2004, we recommended that Congress consider giving RHS access to the Department of Health and Human Services' national wage matching system, or the National Directory of New Hires.[Footnote 34] RHS has stated that access to a national wage matching database would help the agency with tenant income verification. According to RHS's asset management handbook, loan servicers in field offices are to provide consistent, effective oversight of properties financed by RHS to ensure that they are operated in accordance with applicable regulatory and administrative requirements. Because some loans servicers either do not have access or are unaware of resources to verify the legal residency and income information provided by tenants, the level of oversight applied across the FLH portfolio varies. In cases where verification systems are not used, RHS cannot be assured that borrowers and management companies are properly implementing tenant legal residency and income requirements, and ineligible tenants may reside in FLH units. RHS Does Not Take Advantage of Existing Data Sources to Assess Program Demand, Making It Difficult to Target Available Funding: RHS has several sources of information collected within the FLH program that would help estimate trends in demand to occupy FLH units and demand for funds but it does not analyze the information for these purposes. These data include the number, geographic location, and type (that is, for on-farm or off-farm properties) of applications submitted by potential borrowers for FLH funding each year; local market studies submitted as part of an FLH application that document supply and demand for FLH in a given area; and data on properties' vacancy rates that are stored in MFIS.[Footnote 35] RHS uses application information as part of its FLH loan and grant award process. However, it does not analyze borrower applications or data associated with application submission to assess trends in demand in a local area or state. The number of applications or the information in the application package could serve as indicators of demand for funds in particular states or regions. For example, market studies are submitted with each application to assess the need for the project. These studies are used to identify the supply and demand for farm labor housing in a specific market area as part of the loan and grant award process. The market analysis must include information on the annual income level of farmworker families in the area; an estimate of the number of farmworkers who remain in the area where they work and the number of workers who migrate into the region; general information concerning the types of labor-intensive crops in the region and prospects for continued demand for farmworkers; and the condition, number and adequacy of housing currently available to farmworkers in that area, among other information. However, according to the RHS national office, because applicants that were not successful do not become program participants, the national office does not retain data associated with their applications, including the market studies of the applicants' area. According to the RHS national office, state offices do retain application data but do not systematically analyze these data in later funding cycles to estimate trends in demand in a given local area or state. RHS also does not analyze vacancy data to assess trends in demand in a local area or state. RHS routinely collects vacancy data on each property in the portfolio and calculates 3-year average vacancy rates to track the occupancy level of individual properties. While RHS uses information on unit vacancies to identify issues with individual properties, according to RHS, partly due to limited resources, it does not systematically assess overall demand or trends in demand for the FLH program on a local, statewide, or national level. According to RHS, program directors also meet at least on an annual basis to discuss program needs with RHS's Administrator. However, according to RHS, it does not conduct a regional comparison or national review of available data to help manage their program and determine demand for units or funds to develop and rehabilitate units. Some in our group of experts noted that the lack of detailed information on demand for the program hinders RHS's ability to target funds to areas of greatest need. Furthermore, our standards for internal control state that relevant, reliable, and timely information should be available for management decision-making.[Footnote 36] Without analyzing available information--such as application data, market studies including the population of eligible farmworkers, and occupancy data--it is difficult for RHS to estimate demand for FLH units and changes in the needs of its target population, and best allocate FLH resources during its loan and grant award process. However, in February 2011, RHS's Administrator told us that the agency is committed to modifying the program, as necessary, to meet changes in demand trends. For example, she noted that her office had recently discussed conducting a systematic review of the need for the FLH program with USDA's Economic Research Service. Most FLH Program Borrowers Are Not Delinquent or in Default on Their Payments, but Additional Management Attention Needed to Help Ensure Efficient Use of Funds: Most Borrowers in the FLH Portfolio Were Able to Make Timely Payments on Their Properties: According to RHS data and interviews with RHS officials, most borrowers were able to meet the financial needs of their properties. About 6 percent, or 37 out of 731, of farm labor housing borrowers were delinquent on their loans in September 2010. In terms of the age of delinquencies, an estimated 4 percent of properties had delinquent borrowers who were from 91 to 365 days delinquent, and 2 percent were a year or more delinquent.[Footnote 37] About half, or 19, of delinquent loans were associated with properties having fewer than 5 units, and total delinquent properties accounted for 718 units, or 4 percent, of the total FLH portfolio of 16,032 units. Of the 40 states with farm labor housing properties, 15 had borrowers with delinquent FLH accounts. Delinquency rates by state ranged from 0 to 50 percent, though many of the higher delinquency rates were in states with few FLH properties, as shown in figure 6. In interviews with agency officials, staff reported that delinquencies and defaults generally were not a problem with farm labor housing properties and that few properties in the portfolio were in poor financial condition. Figure 6: Percentage and Number of Properties with Delinquent Borrowers by State and Housing Type, as of September 2010: [Refer to PDF for image: illustrated table] Total[A]: On-farm percentage of properties: 5.12%; On-farm number of properties: 23; Off-farm percentage of properties: 4.78%; Off-farm number of properties: 13; All percentage of properties: 5.06%; All number of properties: 37. Arkansas: On-farm percentage of properties: 4.52%; On-farm number of properties: 7; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 4.32%; All number of properties: 7. California: On-farm percentage of properties: 25.00%; On-farm number of properties: 1; Off-farm percentage of properties: 1.10%; Off-farm number of properties: 1; All percentage of properties: 2.11%; All number of properties: 2. Colorado: On-farm percentage of properties: 0%; On-farm number of properties: 0; Off-farm percentage of properties: 8.33%; Off-farm number of properties: 1; All percentage of properties: 8.33%; All number of properties: 1. Florida: On-farm percentage of properties: 0%; On-farm number of properties: 0; Off-farm percentage of properties: 10.00%; Off-farm number of properties: 4; All percentage of properties: 10.00%; All number of properties: 4. Idaho: On-farm percentage of properties: 0%; On-farm number of properties: 0; Off-farm percentage of properties: 50.00%; Off-farm number of properties: 5; All percentage of properties: 50.00%; All number of properties: 5. Maine: On-farm percentage of properties: 50.00%; On-farm number of properties: 2; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 50.00%; All number of properties: 2. Massachusetts: On-farm percentage of properties: 33.33%; On-farm number of properties: 1; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 20.00%; All number of properties: 1. Michigan: On-farm percentage of properties: 3.66%; On-farm number of properties: 3; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 3.45%; All number of properties: 1. Mississippi: On-farm percentage of properties: 3.57%; On-farm number of properties: 1; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 3.45%; All number of properties: 1. New Jersey: On-farm percentage of properties: 11.11%; On-farm number of properties: 2; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 10.53%; All number of properties: 2. New York: On-farm percentage of properties: 0%; On-farm number of properties: 0; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 5.00%; All number of properties: 1. Pennsylvania: On-farm percentage of properties: 0%; On-farm number of properties: 0; Off-farm percentage of properties: 50.00%; Off-farm number of properties: 1; All percentage of properties: 50.00%; All number of properties: 1. Texas: On-farm percentage of properties: 0%; On-farm number of properties: 0; Off-farm percentage of properties: 5.26%; Off-farm number of properties: 1; All percentage of properties: 5.26%; All number of properties: 1 Vermont: On-farm percentage of properties: 7.04%; On-farm number of properties: 5; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 7.04%; All number of properties: 5. Remaining states: On-farm percentage of properties: 0%; On-farm number of properties: 0; Off-farm percentage of properties: 0%; Off-farm number of properties: 0; All percentage of properties: 0%; All number of properties: 0. Source: GAO analysis of MFIS data. [A] Total includes one property in New York that is of an unknown classification (e.g., on-farm or off-farm) or is a nonlabor housing property. [End of figure] RHS data show that the majority of financial noncompliance among FLH properties was related to the late submission of required paperwork. Loan servicers track borrower compliance with financial requirements in the multifamily housing information system, or MFIS.[Footnote 38] We found that about 26 percent of all properties in the FLH portfolio had one or more unresolved financial findings associated with them. However, as shown in figure 7, most open financial findings--89 percent--were related to the late submission of financial documentation, which would include management certificates, budgets, and audit reports. Some loan servicers we interviewed suggested that properties that do not submit all the required financial documentation are not necessarily in poor financial condition, as measured by their ability to make loan repayments and fund their reserve account. For example, an RHS official in Michigan and New York stated that on-farm borrowers may be resistant to sending financial documentation because they consider it an unnecessary bureaucratic burden, particularly if they are making timely payments. Figure 7: Open Financial Findings by Type, as of September 2010: [Refer to PDF for image: pie-chart] Information not received from borrower: 89%; Delinquency or unpaid financial obligation: 9%; Unacceptable submissions or other: 2%. Source: GAO analysis of MFIS data. Note: Values in this figure represent the percentage of open financial findings among FLH properties that are delinquent, as opposed to the percentage of all FLH properties that are delinquent. [End of figure] The majority of properties were able to fund ongoing maintenance needs, as measured by data on reserve account balances and observations made during site visits. Specifically, about 85 percent of the 269 FLH properties with reserve requirements had reserve account balances that met or exceeded the requirement, and about 94 percent of properties had accounts that were at least 75 percent funded, according to RHS data.[Footnote 39] Although we analyzed the extent to which reserve account balances met RHS requirements, we did not assess the extent to which these requirements enabled borrowers to make substantial long-term capital investments in their properties. While the majority of the 20 properties we visited appeared to meet ongoing maintenance needs, we also noted that many of those that recently had engaged in major rehabilitation efforts had sought additional financing for these improvements. Certain characteristics of the FLH program and its borrowers may contribute to the relatively lower delinquency rates in the portfolio. Specifically, about 64 percent of all off-farm FLH units receive RHS Section 521 Rental Assistance funding, which subsidizes tenant rent payments, and according to RHS staff newer properties generally have an even higher proportion of units receiving the subsidy.[Footnote 40] Some RHS officials cited the number of FLH units receiving rental assistance as one likely reason for timely payments from borrowers, since full rental assistance may cover a borrower's loan payment. In addition, the loan program provides a fixed 1 percent loan for a term of 33 years. Therefore, the borrowers' monthly interest expenses are relatively small. High Default Rates Used in Credit Subsidy Calculations Led to Overestimation of Program Costs and Less Low Interest Financing Available for Applicants: Rural Development (RD) overestimated its credit subsidy costs for the fiscal year 2010 FLH loan cohort (that is, the year the loans were obligated), based on supporting documentation the agency provided us and our analysis, which resulted in $11.8 million less available for low-interest financing for program applicants.[Footnote 41] As required by OMB budget guidance, RD officials prepare a credit subsidy estimate during the annual budget formulation process. They estimate the lifetime costs of any direct loans obligated through the FLH program in the applicable budget year using the latest OMB-approved cash flow model and cost components.[Footnote 42] This budget formulation credit subsidy estimate is used to determine the maximum amount of loans that can be obligated under the budget authority available. Then, when RD assembles end-of-year financial statements and subsequent budget submissions, it annually reestimates the credit subsidy cost of each cohort year to include actual performance of individual loan cohorts, expected changes in future loan performance, and changes to the estimation methodology. Four cost components comprise the credit subsidy estimate for the FLH program: defaults, net of recoveries; interest; fees; and a component labeled "all other," which includes prepayments. Appendix V provides additional information on the requirement for agencies to develop credit subsidy rate estimates and reestimates, and information on RD's processes for doing so. During the fiscal year 2010 budget formulation process, in January 2009, RD estimated the credit subsidy rate for the 2010 cohort of the FLH program at 36.14 percent, comprising primarily of interest and default cost components. With a credit subsidy rate of 36.14 percent, for every $100 of direct loans obligated RD estimated that it would incur a cost of $36.14. However, in October 2010, when RD reestimated the credit subsidy rate for fiscal year 2010, the original estimate was shown to be too high and the agency decreased the rate to 25.83 percentage points--or more than 10 percentage points lower than the original credit subsidy rate. This credit subsidy reestimate indicated that the cost of loans obligated in that year decreased by about $3 million.[Footnote 43] According to RD, when it developed the 2010 credit subsidy reestimate it used actual 2010 program data and updated economic assumptions, as required by OMB. It also changed its estimation methodology. RD noted that these changes to the credit subsidy estimation model may also have affected its reestimated rate. However, we found that the primary driver of the change from the fiscal year 2010 credit subsidy estimate to the reestimate was the default cost component and, more specifically, how this cost component was calculated. Specifically, when the fiscal year 2010 budget formulation credit subsidy estimate was calculated, the estimated default cost component was inflated by a prepayment estimate. That is, RD overstated the estimated default cost component to reflect the effect of prepayment. RD, includes the impact of prepayment estimates in the all other cost component. However, RD also included its prepayment estimate to calculate its overall default cost component in 2010. To determine the impact of including the prepayment estimate in both cost components, we recalculated the subsidy rate by removing the prepayment estimate from the default cost component. We found that the inclusion of the prepayment estimate in the default assumptions overstated the cost of the program by about $3 million.[Footnote 44] When we removed the prepayment estimate from the default cost component, the credit subsidy rate was 25.25 percent. According to our analysis, at a credit subsidy rate of 25.25 percent another $11.8 million dollars would have been available in low-interest financing for program applicants for the fiscal year 2010 FLH loan program cohort. RD's supporting documentation for the fiscal year 2010 credit subsidy rate calculations showed that the prepayment estimate was included in the default cost component, but the supporting documentation did not describe the rationale for the inclusion. In response to our questions about the inclusion, an USDA official said that the inclusion of the prepayment estimate in the default cost component was suggested by a USDA Office of Inspector General official to help adjust future cash flows for the impact of partial prepayments. However, this does not explain why partial prepayments would significantly increase default costs. For the 2011 budget formulation process, RD changed its methodology for calculating the credit subsidy rate to consider only the prepayment estimate in the "all other" component as opposed to also including this factor in the default cost component. A more thorough review comparing the key assumptions used in the credit subsidy rate calculations with actual program characteristics may have helped identify the overstated costs earlier, because the default assumptions used in the cash flow model did not reflect actual program performance. For example, including the prepayment estimate in the default cost component resulted in a predicted borrower default rate of 59 percent before recoveries for the 2010 cohort.[Footnote 45] However, this rate is inconsistent with historical actual default performance data. Specifically, for the fiscal years 1992 through 2009 cohorts, actual defaults have been less than 1 percent. According to Federal Accounting Standards Advisory Board (FASAB) guidance, preparing reliable and timely direct loan subsidy estimates must be a joint effort between the budget, accounting, and program offices at each agency, and these offices should coordinate all key assumptions used.[Footnote 46] The FASAB guidance also directs agencies to perform a trend analysis of the credit subsidy cost components, including interest, defaults, and fees, and investigate or explain any unusual fluctuations that are identified. RD documentation shows that the default cost component decreased from $6 million to $72 thousand in the 2010 credit subsidy rate reestimate but there is no explanation of this change. Agency officials' responses to our questioning about the high default rate suggests that RD had not have been closely monitoring unusual fluctuations in credit subsidy cost components, which in 2010 resulted in less money being available for low-interest financing for program applicants. More Than $184 Million in FLH Loan and Grant Obligations Were Unliquidated in 2010 and No Guidelines for De-obligation Were in Force: RHS had more than $184 million in loans and grant obligations for the FLH program that were unliquidated--that is unused--as of September 2010 and the balance of unliquidated obligations has been over $125 million for the past 6 years.[Footnote 47] As shown in figure 8, about $184 million in loans and grants obligated were unliquidated at the end of fiscal year 2010 (about $71 million of this was in grant funding and $112 million was in direct loans). About $24 million of the loans and grants were obligated at least 5 years prior and the oldest unliquidated obligations dated to fiscal year 2001.[Footnote 48] Figure 8: Loans and Grants Obligated but Unliquidated, by Age, as of September 30, 2010: [Refer to PDF for image: stacked vertical bar graph] Obligations in millions: Years: 0 to 0.9; Grants: $13 million; Direct loans: $10 million. Years: 1 to 1.9; Grants: $29 million; Direct loans: $14 million. Years: 2 to 2.9; Grants: $27 million; Direct loans: $13 million. Years: 3 to 3.9; Grants: $19 million; Direct loans: $11 million. Years: 4 to 4.9; Grants: $8 million; Direct loans: $16 million. Years: 5 to 5.9; Grants: $1 million; Direct loans: $5 million. Years: 6 to 6.9; Grants: $1 million; Direct loans: $1 million. Years: 7 to 7.9; Grants: $ million; Direct loans: $0 million. Years: 8 to 8.9; Grants: $4 million; Direct loans: $2.9 million. Years: 9 or more; Grants: $0; Direct loans: $1 million. Total obligations 5 or more years old: $24,175,785. Source: GAO analysis of MFIS data. [End of figure] Reasons may exist for loans and grants obligated when the FLH award was made to remain unliquidated for extended periods until construction or rehabilitation begins. RHS officials described several reasons why a property could experience long periods of inactivity between loan or grant obligation and liquidation. Officials explained that developers can experience difficulties securing funding from multiple federal and state sources. For example, RHS officials in California told us that many developers have been financing FLH properties with multiple funding sources--a mix of federal, state, and private loans, credits, and grants. Many developers also obtain federal low-income housing tax credits to complement FLH program funding during each funding cycle, according to multiple developers with whom we spoke. Some developers with whom we spoke noted that if they do not qualify for tax credits in the first year of applying, they may re-apply in a subsequent year, thus extending the time between agency obligation and liquidation of the funds. According to one agency official, the unliquidated balances also are attributable to existing FLH properties intended for rehabilitation. Funds may remain unliquidated when a property is unable to secure the full amount of financing necessary to complete all needed improvements. The official cited the example of a property in Florida that received about $9 million in loan and grant obligations over 3 years for rehabilitation. According to the official, the property had about $60 million in rehabilitation needs. The official noted that the entire $9 million remained unliquidated as of June 2010. RHS plans to end the practice of using the FLH program as an ongoing source for rehabilitation funding, and anticipated that the balance of unliquidated funds should decline as funding from the agency's Multi- Family Housing Revitalization Demonstration Program was made available for rehabilitation work on older properties.[Footnote 49] Although about $24 million in FLH funds have remained unliquidated for more than 5 years there are no guidelines on de-obligation time frames. RHS issued guidance in 2008 on setting obligation expiration dates after an internal review identified the liquidation of obligated FLH funds as a weakness. The review found that all states visited during the course of the review had loans or grants that were obligated and not closed after 2 to 5 years. The review also found that funds obligated and not closed after 5 years were likely to be insufficient to complete the property, due to increased construction costs, thereby increasing the likelihood that additional agency funds would be needed. The review recommended that RHS issue guidance requesting that all states with loans or grants obligated and not closed after 5 years de-obligate those funds to allow for more immediate program use.[Footnote 50] To address the weakness identified in the review, RHS issued program guidance in the form of an unnumbered letter in 2008 to establish de-obligation time frames for FLH loans and grants.[Footnote 51] In the unnumbered letter, RHS stipulated that obligations for off-farm housing should expire 5 years from the date of obligation and on-farm housing 2 years from the date of obligation.[Footnote 52] The unnumbered letter expired on March 31, 2009, and officials from state and local offices that we visited did not appear to be familiar with it. In particular, none of the RHS staff in five state offices we visited referenced the unnumbered letter or guidance for obligation expiration dates when we asked about their processes for ensuring that obligations do not remain outstanding for more than 5 years. Although no current guidance or requirement setting forth timelines to de-obligate unliquidated loans and grants exists, RHS officials stated that the guidance set forth in the expired unnumbered letter are still warranted. Conclusions: As the only federally assisted source of housing for farmworkers, the FLH program plays an important role in constructing and rehabilitating housing for residents that support the national agricultural sector. However, in several areas RHS could strengthen its management processes to more effectively implement and oversee the FLH program. For instance, RHS performance information indicates a decrease in performance grades among borrowers in recent years. However, low performance grades can stem not only from serious safety and soundness concerns but also from late paperwork. The grade alone does not indicate the severity or type of the problems and on a findings report more than a third of open MFIS entries on the physical condition of properties do not contain additional, descriptive information that could do so. Agency managers require readily usable information. RHS could improve both the functionality and content of its information systems and reporting by considering methods to improve the specificity its performance grades and comments related to performance findings in MFIS. By undertaking such actions, RHS managers could more readily use performance information to plan and conduct its oversight. RHS not only faces some constraints in effectively monitoring FLH performance, but also in enforcing compliance. Due, in part, RHS's use of only mild or severe penalties for noncompliance, some findings and violations leveled against borrowers remain unresolved for extended periods. That is, the enforcement actions RHS uses often may not be applicable or effective against the range of noncompliance that occurs because they are either too mild to be effective or too severe to be invoked. We previously have reported that penalties in federal award programs should correspond to performance. By putting in place more tailored enforcement actions, such as the civil money penalty provided for in program regulations for which RHS has not developed procedures to use, RHS could appropriately and more effectively ensure that FLH program requirements are met. RHS must ensure that borrowers (or their management agents) verify that tenants meet eligibility requirements. However, RHS did not consistently do so because its staff could not access or were unaware of electronic third-party verification systems for tenant legal residency or income documentation. For example, some local offices used the SAVE program to verify residency, but others were unaware of it. To verify income, some RHS offices have access to state wage matching systems, while others do not. We previously recommended that Congress consider giving RHS access to the National Directory of New Hires. RHS has stated that such access would help with income verification and that access to SAVE would help verify residency status. By consistently applying oversight methods--and being able to leverage the information in third-party verification systems--RHS can help assure that only eligible tenants reside in FLH units. RHS's financial management and cost estimation of the FLH program also needs attention because weaknesses could impede achievement of a key program goal--to increase housing for farmworkers. For example, RHS must prepare reliable estimates of program costs to ensure the efficient use of appropriated funds. However, for fiscal year 2010, we found that the agency overestimated the cost of the FLH program by $3 million. As a result, according to our analysis, another $11.8 million could have been available to loan applicants. Reasons for the overestimate include a change to the credit subsidy model and apparent inattention to unusual fluctuations in credit subsidy cost components. A more thorough investigation of unusual fluctuations in key assumptions, namely the predicted default rates, used in the credit subsidy model could help ensure that these assumptions more closely reflect portfolio performance and would allow RHS to optimize funding use. Additionally, RHS had more than $184 million in unliquidated obligations for the FLH program as of September 2010. RHS state and local offices must report and certify the ongoing need for unliquidated obligations semiannually, but no agency guidance to state and local offices on when to recapture these funds is currently in place. Although there may be legitimate reasons why it could take multiple years to liquidate FLH obligations, the lack of agency guidance makes it difficult for management to ensure that limited program funds are timely and efficiently used. Issuing guidance to all RHS staff in the state and local offices about how and when to recapture program funds would help ensure greater utilization of these limited funds for the development and rehabilitation of farm labor housing. Finally, RHS has an opportunity to leverage existing data to strengthen program management. RHS uses application data and market studies to manage individual applications and properties, but it does not analyze these data sources to identify trends or patterns in demand over time in local areas or states. By utilizing existing data sources for these purposes, RHS could better estimate the extent of demand for farm labor housing and funding and more effectively target funds to areas of greatest need. Recommendations for Executive Action: We recommend that the Secretary of Agriculture direct the Administrator of RHS to take the following seven actions: * To better determine and track compliance across the portfolio, RHS should implement mechanisms to improve the specificity and timely reporting of its compliance review information--such as findings data and performance grade data in MFIS. * To help resolve identified borrower noncompliance in a timely manner, RHS should implement enforcement mechanisms that can be tailored to the severity of the borrower noncompliance, such as the civil money penalty enforcement provision in its program regulations. * To better ensure that requirements for tenant eligibility are met across the FLH portfolio, RHS should (1) require its loan servicers to use the Systematic Alien Verification and Entitlements (SAVE) program administered by the Department of Homeland Security to verify tenant's residency status during supervisory reviews; and (2) seek legislative authority to gain access to the Department of Health and Human Services' National Directory of New Hires and make this information available to RHS so that they can assess the accuracy of tenant income documentation during supervisory reviews and other oversight activities. * To help ensure that reliable program costs are estimated in future years, program officials should, on an annual basis, work with budget staff to investigate key assumptions, including comparing these assumptions to actual program performance, in order to explain unusual fluctuations impacting the credit subsidy rate used in budget formulation. * To better ensure that FLH funds obligated but unliquidated are efficiently used to provide farm labor housing, RHS should issue guidance on obligation expiration dates and make all RHS staff in the state and local offices aware of the guidance and how to implement it. * RHS should also better utilize available data on demand for the FLH program--such as systematically reviewing local market analyses, further analyzing occupancy data on a statewide, regional, or national level, and retaining and analyzing application information--to help target available funding to areas of greatest need. Agency Comments and Our Evaluation: We provided a draft of this report to USDA for review and comment. USDA's Under Secretary for Rural Development provided written comments that are discussed below and presented in appendix VI. USDA generally agreed with all of our recommendations, noting that the recommendations will help make the FLH program better. In its letter, however, the agency provided some additional information and disagreed with certain statements in the report. For example, USDA stated it disagreed with a comment from "an expert" who noted that the FLH program does not receive appropriate care or attention from the national office. However, as noted in the report, this statement reflected the opinions of multiple experts who participated in our 1- day discussion on demand for farm labor housing and the extent to which the FLH program is positioned to meet demand. In addition, as USDA commented in its letter, we noted in the report that the national RHS office has specialists, team leaders, and a financial and loan analyst who work on multifamily housing loans and grants, including FLH loans and grants, across multiple states. Although USDA also generally agreed with our recommendation to improve the specificity of its compliance review information, the agency offered additional explanation for observed decreases in performance scores (grades) in the FLH portfolio. The agency noted that in 2008 it automated performance scoring, a change that, according to USDA, increased the number of low grades by identifying previously unaddressed open findings. We also reported this change to the performance classification system as a potential cause for the observed increase in low grades. However, other deficiencies in FLH program management processes drove our recommendation to improve the specificity and timely reporting of compliance review information. For instance, USDA notes that the performance classification system monitors the quality of its FLH properties as determined by a property's physical, financial, and management operations. However, a grade in the classification system of A through D does not yield information on whether the problem is related to physical, financial, or management operations. In the report, we also discuss the FLH performance classification system and a similar system that the HUD uses and state that the FLH system differs from HUD's in that it does not specify the performance problem causing the grade. USDA commented that further examination of its full electronic data system--the MFIS-- would yield specific information on these performance areas. We agree that additional information is available in MFIS. But, the classification system itself lacks specificity and does not readily provide agency officials with information necessary to assess the causes of low grades. Furthermore, we noted deficiencies in MFIS findings information--as reflected in the finding reports generated from MFIS, which often do not detail the type and the severity of the findings. In the report, we also discuss reported deficiencies in the timely submission of compliance information into the database that underlies the performance classification system, which could impact the accuracy of the information available to FLH management. Therefore, we believe our findings and recommendation on improving the program's compliance review information remain valid--and USDA also commented that more detail on specific physical condition issues from its performance management processes would be beneficial to all users. In its letter, USDA acknowledged that the default cost component in the 2010 estimate of the credit subsidy rate was overstated, as described in this report. However, the agency notes that subsidy estimates are routinely revised and identifies several factors that may have contributed to the downward reestimate. For example, according to USDA, the original estimate was based on an "interim data solution" until a new model could be developed. According to USDA, when it developed the reestimate, the agency changed its estimation methodology, used actual 2010 program data, and updated economic assumptions, as required by OMB. We discussed these changes with the Deputy Director of Rural Development's Budget Division while the draft was with USDA for comment and made technical changes to the report as a result. However, during this meeting, the Deputy Director also noted that, while the additional factors may have affected the downward credit subsidy reestimate rate, these factors likely had a lesser influence on the overall reestimate than did the corrected default cost component--which is what our analysis of the credit subsidy model and supporting materials indicated was the primary cause for downward adjustment. In the letter, the Undersecretary further describes recent efforts to review and revise credit subsidy rate assumption data and calculations, and agrees that increased cooperation among program, financial, and budget staff would improve the FLH program. We are sending copies of this report to interested congressional committees and the Secretary of Agriculture. In addition, the report will be available at no charge on GAO's Web site at [hyperlink, http://www.gao.gov]. If you or your staffs have any questions about this report, please contact me at (202) 512-8678 or clowersa@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs are listed on the last page of this report. GAO staff who made major contributions to this report are listed in appendix VII. Signed by: A. Nicole Clowers, Acting Director: Financial Markets and Community Investment: [End of section] Appendix I: Objectives, Scope, and Methodology: The objectives of this report were to examine: (1) how demand for the Farm Labor Housing (FLH) program has changed over time, key factors that influence demand for such housing, and whether the program model addresses demand; (2) the extent to which Rural Housing Service (RHS) management processes assure farmworkers access to decent and safe housing and compliance with program requirements; and (3) the financial status of properties in the FLH portfolio and the extent to which RHS processes ensure the sound financial management of the program. To address the first objective, we contracted with The National Academies to convene a diverse group of experts, to discuss trends in demand for farm labor housing, factors that influence demand, and the extent to which the FLH program is positioned to meet demand. To select the experts, we and The National Academies identified 12 individuals for the group through interviews on the basis of their extensive knowledge of the FLH program and trends in demand for farm labor housing and to obtain regional diversity and a range of types of organizations.[Footnote 53] While we attempted to select experts who provide a range of experience and views, the group of experts selected may not represent all perspectives on demand for FLH, including that of RHS, as no RHS staff were invited to the group discussion in order to encourage openness among other participants who use FLH program funds. The final group of 11 experts who convened at The National Academies in Washington, D.C., on October 13, 2010, represented housing developers; borrowers of FLH funds; researchers who conduct research or are involved in the study of issues related to farmworker housing; staff of nonprofit organizations who are knowledgeable about and advocate for issues related to farmworker housing; and USDA contractors who provide technical assistance to FLH developers. A contractor recorded and transcribed the meeting to ensure that we had accurately captured the group's statements. The day was divided into three discussion sessions which were structured to focus on the aspects of demand noted above. A moderator and an assistant moderator helped guide the discussions in each session. To help elicit additional information relevant to our three-part focus on demand, we administered a questionnaire to the experts to collect their responses on factors that most influence demand for units and for FLH funds, the extent to which the FLH program meets demand for units and for FLH funds, and the extent to which the FLH program could be changed to better meet demand. In addition, to systematically analyze information experts suggested for changes to the FLH program, we conducted a content coding review of the transcripts by coding relevant quotes and grouping them into categories. An analyst identified, coded, and entered relevant text into a spreadsheet, while another verified these entries. To balance and augment the perspectives of our group of experts, we also reviewed relevant studies and reports to identify research studies that examined how demand for farm labor housing in general and the FLH program in particular had changed over time, as well as key factors that influenced these changes and influence demand for such housing. We used various Internet search databases to identify studies, including ProQuest, ABI Inform, SIRS Researcher, and Agricola. We sought to identify additional studies by consulting with government officials, researchers, and staff from nonprofit organizations throughout the course of research and by reviewing the bibliographies of the previously identified studies. As part of this effort, we also reviewed documentation of national FLH stakeholder meetings convened by USDA in 2008 and 2009. The studies and reports we reviewed primarily focused on specific states or regions, and also indicated that demand may vary across states and agricultural regions. Many of the studies noted limitations in the data available on farmworkers. We present additional information about federal data sources on farmworkers in appendix III. To address the second objective, we conducted site visits to RHS local and state offices, and FLH properties to determine the extent to which RHS management processes assure farmworkers access to decent and safe housing and assure compliance with program requirements. To address this objective we also conducted tenant file reviews, analyzed electronic program data, reviewed Multifamily Housing program regulations and handbooks, and interviewed program staff at all levels as well as program borrowers. To obtain more in-depth information about the oversight of the FLH program in individual states, and the servicing of FLH properties, we completed multi-day site visits to five states including California, Florida, Michigan, New York, and Texas that each included interviews with the state office and two local offices with the exception of Michigan. In Michigan we met with only one local office because the state's on-farm labor housing program was serviced by the state office, which was not the case in the other four states. The five site visit states were selected to obtain regional diversity and a range in type (on-farm and off-farm) and number of properties and units per state (see table 3). We completed walkthroughs of 20 properties (4 properties in each of the five states), which included a tour and inspection of the interior and exterior of the properties and conversations with the borrower or property manager, and a tenant file review for each off-farm property.[Footnote 54] We selected properties to include both property types (on-farm and off-farm) and a range of property sizes and performance grades. An expert in facilities and construction management accompanied us to Texas. On-farm borrowers are not required to maintain tenant files for each unit. Prior to each site visit we received a list of tenants from each off-farm property scheduled for a walkthrough, randomly selected 10 tenant files per property for 10 of the 13 off-farm properties visited, and requested copies of the files, which were sent to us in hard copy or electronically prior to the site visit.[Footnote 55] We developed a data collection instrument to review and summarize contents of each tenant file, including documents to assess income and residency eligibility. However, the contents of these tenant files are not necessarily representative of the contents of all other FLH tenant files. Table 3: 2010 FLH Program Characteristics for Site Visit States: State: California; U.S. region: West; Total properties: 95; On-farm properties: 4; Off-farm properties: 91; Total units: 5,490; On-farm units: 16; Off-farm units: 5,474. State: Florida; U.S. region: South; Total properties: 40; On-farm properties: 0; Off-farm properties: 40; Total units: 4,547; On-farm units: 0; Off-farm units: 4,547. State: Michigan; U.S. region: Midwest; Total properties: 85; On-farm properties: 82; Off-farm properties: 2; Total units: 353; On-farm units: 309; Off-farm units: 44. State: New York; U.S. region: Northeast; Total properties: 20; On-farm properties: 17; Off-farm properties: 2; Total units: 91; On-farm units: 67; Off-farm units: 24. State: Texas; U.S. region: South; Total properties: 19; On-farm properties: 0; Off-farm properties: 19; Total units: 1,320; On-farm units: 0; Off-farm units: 1,320. Source: GAO analysis of MFIS data. Note: According to MFIS data, two properties, one in Michigan and one in New York, are of an unknown type and are not listed under on-farm or off-farm properties in those states. [End of table] To analyze portfolio-wide data on compliance with FLH program requirements, we obtained extracts from the agency's Multi-Family Housing Information System (MFIS). To assess the performance of FLH properties over time, we reviewed performance classification data in MFIS from fiscal year-end 2006 through fiscal year-end 2010. To assess the types of findings assigned to FLH properties over time, we reviewed the number of open and resolved findings by type, specifically financial and physical finding. We assessed the reliability of these data by (1) performing electronic testing, (2) reviewing existing information about the data and the system that produced them, and (3) interviewing agency officials knowledgeable about the data and related management controls. Based on this assessment, we determined the data to be sufficiently reliable for the purposes of this report. We also consulted our Standards for Internal Control in the Federal Government to review control activities that apply to RHS's performance management and servicing activities. We interviewed the RHS national office to determine its role in monitoring the FLH program and state and local offices with FLH oversight and servicing responsibilities. Lastly, we interviewed nonprofit organizations in each state we visited that were current or past FLH borrowers, had received funds from the FLH program to provide technical assistance to other developers, or provided services to farmworkers to help them find safe and decent housing. To address the third objective, we analyzed delinquency, reserve account, and financial findings data from extracts of RHS's MFIS to assess the financial status of properties in the FLH portfolio. To identify program compliance and assess overall program needs, MFIS contains information on budgets, operating costs, non-financial defaults, insurance, reserve account funding, management plans, supervisory visits, taxes, and tenant changes. We also obtained and analyzed electronic program data from RHS's Automated Multi-Family Housing Accounting System (AMAS), which contains accounting information that is used to identify delinquencies and financially delinquent borrowers. For both AMAS and MFIS, we received data that were current as of the end of fiscal 2010. According to RHS's officials these data systems contain only the last 3 years of data for each property. RHS underwriting and servicing processes include financial analyses of applicants, annual budget reviews, and the setting of reserve fund requirements.[Footnote 56] We assessed the reliability of these data by (1) performing electronic testing, (2) reviewing existing information about the data and the system that produced them, and (3) interviewing agency officials knowledgeable about the data. Based on this assessment, we determined the data to be sufficiently reliable for the purposes of this report. We examined documents and reports, such as financial statement audits, which RHS officials use to monitor the performance of the loan portfolio. We also reviewed agency handbooks that contained guidance on asset management and project servicing, and interviewed headquarters, state, and local staff knowledgeable about financial underwriting and servicing efforts. To specifically assess the extent to which RHS processes ensure the sound financial management of the program, we studied the credit subsidy estimation process and RHS's management of its balance of unliquidated loan and grant obligations. For our credit subsidy work, we examined the fiscal year 2010 and 2011 credit subsidy cash flow models for the FLH program, reestimate data, and supporting documentation. To verify the validity of the fiscal year 2010 model, we entered data that RHS provided into OMB's Credit Subsidy Calculator 2 and confirmed that the resulting estimates matched the figures provided in federal budget documents. Based on these results, we determined that the information was sufficiently reliable for our analysis. We also interviewed program and budget staff about the default assumptions used in and recent changes to the model. For our examination of program obligations, we examined end-of-fiscal-year unliquidated obligations reports for 2003 through 2010 and obligation data from AMAS. We compared agency documents with obligation data in federal budget appendixes and confirmed that these figures were sufficiently reliable for our analysis. We also interviewed RHS staff from the national, state, and local offices about their management of unliquidated loan and grant obligations and reasons for extended obligation periods. We conducted this performance audit from March 2010 to March 2011 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. [End of section] Appendix II: Experts Convened by GAO with the Assistance of The National Academies on Demand for Farm Labor Housing: This appendix provides the names and affiliation of individuals who participated in our 1-day expert group discussion convened by GAO, with the assistance of The National Academies on October 13, 2010. The following experts discussed topics related to the demand for farm labor housing: * Gideon Anders, Senior Attorney, National Housing Law Project, San Francisco, Calif. * Pamela Borton, President, Southwind Management Services, Inc., Clearwater, Fla. * Peter Carey, President and Chief Executive Officer, Self-Help Enterprises, Visalia, Calif. * Dennis Harris, Housing Director, Telamon Corporation, Raleigh, N.C. * Moises Loza, Executive Director, Housing Assistance Council, Washington, D.C. * Joe Myer, Executive Director, National Council on Agricultural Life and Labor Research Fund, Inc., Dover, Del. * Brien Thane, Executive Director, Washington State Farmworker Housing, Seattle, Wash. * Kathy Tyler, Director of Housing, Motivation Education and Training, Inc., New Caney, Tex. * Don Villarejo, Founder and Director Emeritus, California Institute for Rural Studies, Davis, Calif. * Rob Williams, Director, Florida Legal Services, Inc., Tallahassee, Fla. * John Wiltse, Senior Operations Director, PathStone Corporation, Rochester, N.Y. [End of section] Appendix III: Federal Data on Farmworkers: This appendix provides information about nationwide, federal data available on farmworker populations. Both the terms farmworker and farm laborer are used by researchers, government entities, and nonprofit organization in reference to individuals who work in agriculture, aquaculture, and processing activities. The U.S. Census Bureau considers migrant and seasonal farmworkers to be a "hard to count" population for reasons such as language barriers, mobility, unconventional housing arrangements (such as dormitories, cabins, or trailers in labor camps), and mistrust of formal government efforts to collect data. Available sources of federal data on farmworker populations include, but are not limited to: * The U.S. Department of Agriculture's (USDA) National Agricultural Statistics Service (NASS) conducts the Farm Labor Survey (FLS), which provides quarterly estimates of the number of hired farmworkers, the percentage of workers who are migrants, and average weekly hours worked. Four times a year, USDA surveys about 14,500 farms in all states except Alaska and provides total numbers of farmworkers obtained from farm establishments. The FLS also provides average wage rates for hired workers by type (field, livestock, supervisor, and other) for 16 states and 15 regions. Its data on hired farmworkers refer to all types of workers on the farm, including bookkeepers, secretaries, and mechanics, as well as persons who pay themselves regular salaries, such as partners or corporate shareholders. * NASS also conducts the Census of Agriculture on nationwide farmworker employment data every 5 years with the last survey conducted in 2007. The Census of Agriculture offers comprehensive geographic coverage of hired and contract farm labor use as measured by labor expenditures, and currently is the only national level data source that offers consistent farm labor information at the county and state level. The Census of Agriculture also reports the number of hired workers, separated by whether they worked less than 150 days or 150 days or more. As with the FLS, the data refer to all hired workers on the farm, including those not generally considered farmworkers. * The Department of Labor (Labor) sponsors the National Agricultural Workers Survey (NAWS), which is an employment-based, random sample survey that collects detailed information on individual farmworkers, including their legal residency status. NAWS data are limited to hired crop farmworkers and excludes hired livestock farmworkers and processing workers. NAWS collects data from personal interviews of between 1,518 and 3,600 randomly selected crop field workers. According to a 2008 Economic Research Service report, NAWS data are collected at the worksite and, therefore, are more likely to capture persons who have less stable living arrangements and who tend to avoid participation in more formal data collection efforts. * The Bureau of the Census for the Bureau of Labor Statistics conducts the Current Population Survey (CPS), which provides employment and demographic information on the entire U.S. workforce. It is conducted each month using a probability sample of households over 16 months and is designed to represent the U.S. civilian non-institutional population. Since the survey is conducted for the same households over an extended period, it may undercount unauthorized and foreign-born persons who migrate frequently and are reluctant to participate in formal government questionnaires. Although estimates of the domestic farm labor population have varied widely depending on the survey, according to USDA's NASS, the United States had an average of 1,041,250 hired farmworkers in 2010.[Footnote 57] These data show that the total number of farmworkers has remained relatively stable over the past decade. However, available nationwide data sources have limitations, especially for determining characteristics related to tenant eligibility in USDA's Farm Labor Housing (FLH) program, such as residency status and type of farmworker. No single source of data is available to provide all the necessary detail for understanding farm labor supply, demand, and characteristics that relate to eligibility criteria for the FLH program. The data sets mentioned above provide information for different subgroups within the entire population of persons employed in agriculture and (1) may exclude a portion of FLH- eligible program participants such as processing workers, (2) may include a population not eligible for the FLH program, or (3) may not collect information on characteristics that determine program eligibility such as residency status. For example, the Census of Agriculture and FLS provide numbers of farmworkers nationwide; however, they lack information on residency or housing status, and the data do not include processing workers. NAWS collects information on residency status, but excludes farmworkers who work on ranches. The FLS defines hired workers on farms to include bookkeepers, secretaries, and mechanics, as well as persons who pay themselves regular salaries, such as partners or corporate shareholders. This population is not eligible to reside in FLH program units. [End of section] Appendix IV: Age of the FLH Property Portfolio and Condition of FLH Properties We Visited: The overall FLH program portfolio is aging, with 46 percent of the properties more than 20 years old according to U.S. Department of Agriculture (USDA) data (see figure 9). Nearly three quarters, 73 percent, of properties were more than 10 years old. States with the highest number of units also have high proportions of aging properties. In Texas, with the third highest number of units, 79 percent of FLH program properties are more than 20 years old. In Florida, 35 percent of the properties is more than 20 years old. In California, 39 percent of properties are more than 20 years old. Properties in California and Florida have received revitalization funds in recent years. In 2009, revitalization funds became available for FLH properties through Multi-Family Housing Revitalization Demonstration Program administered by the RHS. In 2010, RHS obligated $2.4 million in funds to repair and rehabilitate three FLH properties. Figure 9: Age of Farm Labor Housing Program Portfolio, as of September 30, 2010: [Refer to PDF for image: vertical bar graph] Age of properties (in years): 0 to 5; On farm (449): 71; Off farm (272): 54; Total projects(731): 125 Age of properties (in years): 6 to 10; On farm (449): 30; Off farm (272): 45; Total projects(731): 75 Age of properties (in years): 11 to 20; On farm (449): 120; Off farm (272): 72; Total projects(731): 192 Age of properties (in years): 21 to 30; On farm (449): 209; Off farm (272): 73; Total projects(731): 291 Age of properties (in years): More than 30; On farm (449): 19; Off farm (272): 28; Total projects(731): 48 Source: GAO analysis of MFIS data. [End of figure] We completed site visits to five states that included walkthroughs of four properties in each state. A brief description of some of the findings from our site visits to assess FLH properties in five states are as follows: * California: All four properties visited had low levels of disrepair with few visible minor deficiencies and no major deficiencies. We noticed some vermin infestation in an unoccupied, seasonal property. Some recently developed properties met energy-efficient standards. For example, one property in California exceeds California Title 24 energy standards according to the borrower. This property includes energy- efficient appliances, solar reflective roof materials that decrease heat absorption, on-demand water heaters, and artificial turf (see figure 10). Figure 10: FLH Property in California with Reflective Roof Materials, Energy-efficient Appliances, On-demand Water Heater, and Artificial Turf: [Refer to PDF for image: 3 photographs] Source: GAO. [End of figure] * Florida: Two properties that we visited had been newly developed and had low levels of disrepair. (See figure 11.) However, one large property with more than 700 units had not undergone rehabilitation since 1968 and exhibited deficiencies on the exterior and interior of the units visited. For example, windows in some units were blocked or replaced with wooden boards, and, in some cases, kitchen appliances, including ovens and refrigerators, were not provided by the landlord and had to be provided by farmworkers. None of the units had central air conditioning and some of the kitchen appliances were in need of repair (see figure 12). Figure 11: Newly Developed Florida Properties with Low Levels of Disrepair: [Refer to PDF for image: 2 photographs] Source: GAO. [End of figure] Figure 12: Windows Replaced with Wooden Boards and a Kitchen in Need of Repair at an Older Florida FLH Property: [Refer to PDF for image: 2 photographs] Source: GAO. [End of figure] * Michigan: Two on-farm properties that we visited had a number of deficiencies such as rotting and unstable porch steps, water damage to the exterior, and an open crawl space (see figure 13). Two other properties we visited in Michigan were well maintained, with few visible minor deficiencies. Figure 13: FLH Unit in Michigan with Water Damage to the Exterior: [Refer to PDF for image: photograph] Photograph highlights water damage and moss. Source: GAO. [End of figure] * New York: The properties we visited in New York exhibited both high and low levels of disrepair. Tenant living standards partly contributed to the observed deficiencies. For example, grease covered the surfaces in one kitchen we observed. However, we also observed deficiencies, such a window covered by a board and severely damaged carpeting, which the owner is required to address (see figure 14). Figure 14: Window Covered by a Board in an FLH Unit in New York: [Refer to PDF for image: photograph] Source: GAO. [End of figure] * Texas: The FLH properties we visited in Texas exhibited low to medium levels of disrepair. Some units had newer sinks, countertops, and ovens, while some units had kitchen appliances in need of repair. Three of the properties have received recent or ongoing rehabilitation (see figure 15). Figure 15: FLH Unit Undergoing Rehabilitation in Texas: [Refer to PDF for image: 2 photographs] Source: GAO. [End of figure] [End of section] Appendix V: FLH Credit Subsidy Rate Calculation: Under the Federal Credit Reform Act of 1990 (FCRA), USDA and other federal agencies must estimate the net lifetime cost--known as credit subsidy cost--of their direct loan programs and include the costs to the government in their annual budgets. Credit subsidy cost represents the net present value of expected lifetime cash flows, excluding administrative costs. Generally, agencies must produce annual updates of their credit subsidy cost estimates--known as reestimates--for each cohort on the basis of information on actual performance and estimated changes in future loan performance. Agencies may makes changes in their estimation methodology, which can effect reestimates, and each additional year provides more historical data on loan performance that may influence future year estimates. Economic assumptions (such as interest rates) also can change from year to year. The credit subsidy cost is frequently presented as a credit subsidy rate. For example, RD estimated that the loans obligated during 2010, would have a credit subsidy rate of 36.14 percent meaning that for every $100 of direct loans obligated, RD estimated that it would incur a cost of $36.14. [Footnote 58] Agencies estimate four cost components that account for total program costs: defaults, net of recoveries; interest; fees; and all other, which includes an estimate of prepayments, both during the budget formulation process and again when assembling year-end financial statements. RD's fiscal year 2006 to 2011 estimated credit subsidy rates and estimated subsidy rate components are shown in table 16. Table 4: Estimated Credit Subsidy Rate for FLH Program: Cohort year: 2006; Subsidy rate components: Defaults, net of recoveries: 0; Subsidy rate components: Interest: 44.9; Subsidy rate components: All other: -0.3; Total subsidy rate: 44.6. Cohort year: 2007; Subsidy rate components: Defaults, net of recoveries: 0.2; Subsidy rate components: Interest: 45.5; Subsidy rate components: All other: 2.2; Total subsidy rate: 48.0. Cohort year: 2008; Subsidy rate components: Defaults, net of recoveries: 8.9; Subsidy rate components: Interest: 44.5; Subsidy rate components: All other: -10.1; Total subsidy rate: 43.3. Cohort year: 2009; Subsidy rate components: Defaults, net of recoveries: 9.5; Subsidy rate components: Interest: 41.0; Subsidy rate components: All other: -8.4; Total subsidy rate: 42.1. Cohort year: 2010; Subsidy rate components: Defaults, net of recoveries: 11.5; Subsidy rate components: Interest: 25.5; Subsidy rate components: All other: -0.8; Total subsidy rate: 36.1. Cohort year: 2011; Subsidy rate components: Defaults, net of recoveries: 0.1; Subsidy rate components: Interest: 39.1; Subsidy rate components: All other: -0.8; Total subsidy rate: 38.4. Source: Federal budget credit supplements. Note: The subsidy rate component of defaults net of recoveries includes the estimated cost of defaults less recoveries of defaults. The interest component reflects the cost associated with the interest payments from the borrower based on the borrower interest rate of 1 percent and the interest cost to the government to provide the loans, which for the 2010 cohort was estimated to be 2.92 percent. "All other" includes the effect of prepayments, losses other than defaults, and any forecasted subsidy reduction due to program fees. [End of table] [End of section] Appendix VI: Comments from the U.S. Department of Agriculture: USDA: United States Department of Agriculture: Rural Development: Office of the Under Secretary: 1400 Independence Ave, SW: Washington, DC 20250-0700: [hyperlink, http://www.rurdev.usda.gov] Committed to the future of rural communities. "USDA is an equal opportunity provider, employer and lender." To file a complaint of discrimination write: USDA, Director, Office of Civil Rights, 1400 Independence Avenue, S.W., Washington, DC 20250-9410 or call (800) 795-3272 (voice) or (202) 720- 6382 (TDD). March 15, 2011: A. Nicole Clowers: Acting Director: Financial Markets and Community Investment: United States Government Accountability Office: 441 G Street, NW: Washington, DC 20548: Dear Ms. Clowers: Thank you for providing the Department of Agriculture (USDA) Rural Development and the Rural Housing Service (RHS) with your Government Accountability Office (GAO) draft report entitled, "Opportunities Exist to Strengthen Farm Labor Housing Program Management and Oversight," Report Number GA0-11-329. We appreciate the opportunity to respond to GAO's comprehensive study of the Farm Labor Housing (FLH) program, and the agency generally agrees with the report's recommendations. For your consideration, USDA offers the following comments to the draft report and requests that a copy of these comments be included in your final report. USDA's FLH program is vital to providing safe, decent, and affordable housing for farm workers throughout America. Currently more than 13,000 seasonal and non-seasonal farm workers and their families live in more than 730 on-farm and off-farm housing developments made possible through USDA's Section 514 and Section 516 FLH loan and grant programs. USDA takes its management and oversight of the FLH program very seriously, although it recognizes that there are opportunities for improvement, and the recommendations provided by GAO will help make the FLEA program better. However, the agency disagrees with the comment "does of an expert cited by GAO that the program not receive appropriate care or attention from the national office." As GAO notes in the report, several members of the national office staff are involved in FLH loan making, servicing, and oversight activities. At least one-third of the national office multi-family housing staff work on FLH projects and issues, even though FLH accounts for only 5 percent of the overall multi-family housing portfolio. The report also notes that there was a substantial increase in the rate of low performance grades in the agency's FLH portfolio. This was a result of improved reporting capabilities and does not necessarily indicate deterioration in the condition of USDA's FLH stock. The agency uses a performance classification system that monitors the quality of its FLH as determined by the property's physical, financial, and management operations. In 2008, the agency found that its existing classification scoring standards were inadequately accounting for problems with borrower plans to work out certain open findings. To address the issues, changes were made to automate project classification scoring to identify all open findings addressed in the borrower work out plans. As a result of the automation process, unaddressed open findings were identified, which caused projects' classification designations to be changed from B to C. Although agency loan servicers had already been working with owners of B classified properties, the automation enhancement provided servicers with more comprehensive information to monitor the progress of the borrower work out plan. GAO compares RHS' Multi-Family Information System (MFIS) to HUD's Public Housing Assessment System (PHAS), and states that MFIS does not specify performance areas of concern, unlike PHAS. RES disagrees. MFIS does include information on specific deficiencies in each of its three categories of operations, so that a user of the system can quickly identify the type of performance concern. A user proficient in MFIS, and familiar with the specific project, would be able to understand the areas of concern based on the information in the system. However, we agree that more detail on specific physical conditions issues would be beneficial for all users, such as those not familiar with the specific project or new to the system. We appreciate GAO's recommendation that RHS obtain access to the Department of Health and Human Services' (HHS) National Directory of New Hires, because we agree that access to the system would provide significant benefits, and savings, in the FLH program. RHS has met with HHS about system access for both RHS multi-family and single-family staff, and the legislative authority to gain access to the New Hires database is under review with the Administration. In the report, GAO indicates that the need for further analysis of available data on local, state, and national markets could help RHS better determine markets most in need of FLH assistance, and allow the agency to better target its funding to areas of greatest need. RHS is interested in using its FLH application data to better understand seasonal and non-seasonal migrant patterns and improve our understanding of current migrant housing needs. Although the agency still needs to determine whether the quantity of available FLH application data is sufficient to completely meet its needs, we are optimistic that it will enable RHS to better target its FLH awards to the markets that can benefit the most. RHS agrees with GAO that the market studies provided with each FLH loan or grant application may be a valuable tool in better targeting loans and grants toward projects with the greatest chance of success. The market studies, which are submitted with every application package, must support the need for the FLH housing in that market area. Clearly, there are cases where the demand as demonstrated in the market study does not materialize, causing high vacancy rates and potentially a change in the use of the project to accept tenants eligible under the Section 515 program. RHS believes that a review of the market studies of such projects may help us identify patterns indicative of future problems, thereby improving our analysis of future FLH loan and grant applications. RHS also agrees that it needs to be more aggressive in pursuing the de- obligation of transactions that were never closed. However, it should be recognized that many of the transactions involve funding from multiple sources, which increases the complexity of project transactions and may cause delays in the timing of the transactions as all parties work to finalize a complete funding package. The delays in funding may have been exacerbated by the recent weakness in financial markets, including the Low Income Housing Tax Credit market as well as funding from state sources. Therefore, while RIIS will review its unliquidated FLH transactions for potential de-obligation, it anticipates that its guidance to the field would include allowing reasonable time frames for funding packages to be assembled before requiring the funding de-obligation. Finally, GAO recommends that RD staff review key assumptions that impact the FLU credit subsidy rate. During reestimates execution, agencies routinely revise their original subsidy estimates using the latest available information. The purpose of reestimates is to make the original estimate more accurate by including actual data, updating assumptions, and improving estimation methodology. At the time of 2010 budget formulation Rural Development (RD) was aware of the old model's limitations. An interim data solution was implemented to support assumptions until a new model could be developed to properly forecast out-years assumptions. With this interim solution and a minor model revision, there was a misallocation of cash flows into the wrong component which had nothing to do with the actual defaults. However, the fact that the default cost component was overstated does not automatically imply that the overall subsidy rate was overstated by that amount. Historically, formulation subsidy rates for the Farm Labor Direct Loan Program range from 34.15% (FY 2012) to as high as 56.80% (FY 1996). The formulation rate for FY 2010 was 36.14% which was comparable with the historical rate pattern for this model even though the default rate was overstated. A new model, described below, was used to formulate subsidy rates for FY 2011 and FY 2012. The agency has learned much and improved many Credit Reform processes since the development of this program's original model. RD initiated a process of developing a new model to accurately calculate the subsidy rate for this program. In the course of this process, RD re-evaluated the old model methodology, reviewed the program's historical performance, built a new cash flow model, and updated the assumption calculation methodology. As a result, the new cash flow model used in 2010 reestimates produced a subsidy rate that estimated the program's cost more accurately and correctly allocated cash flows among various cost components. The assumption curves for this program have been completely revised to be consistent with the new model format and calculations. The prepayments and default curves have been separated and the calculations for these curves are independent of one another. Additionally, surviving principal is no longer used as the basis for the curve calculations and the computation of these curves are now based on the obligation. Actual program data are used to calculate the prepayment and default curves. The actual data is verified to the account and/or financial data. Additionally, significant enhancements have been made to the review process of the assumption data to ensure its accuracy. The Office of the Deputy Chief Financial Officer (DCFO) and Budget Division (BD) have implemented a process in which both offices now review the assumption data and calculations. Finally, the new assumption data and curve for this model has been thoroughly tested and audited, and no inconsistencies between the model and input assumption calculations have been noted. Even though DCFO, BD and mission program have been working together during the budget formulation process, we agree that increased coordination among the three groups would benefit the program. Once again, we appreciate the opportunity to respond to GAO's report on FLH, and we hope that our comments will help GAO in the preparation of its final report. If you have any questions, please contact John Purcell, Director, Financial Management Division, at (202) 692-0328. Sincerely, Signed by: Cheryl L. Cook, for: Dallas Tonsager: Under Secretary: Rural Development: [End of section] Appendix VII: GAO Contact and Staff Acknowledgments: GAO Contact: A. Nicole Clowers (202) 512-8678 or clowersa@gao.gov: Staff Acknowledgments: In addition to the individual named above, Andy Finkel, Assistant Director; Michael Armes; Marcia Carlsen; Kimberly Cutright; Terence Lam; John McGrail; John Mingus, Jr.; Marc Molino; Luann Moy; Amy Radovich; Barbara Roesmann; Julie Trinder; and Michelle Wong made major contributions to this report. [End of section] Footnotes: [1] Throughout the report, we refer to FLH "properties" and "units." An FLH property is a piece of real estate (land and buildings) with one or more rental units and related facilities operated under one management plan and financed with FLH funds. "Units" are rented individual dwellings on a property, such as an apartment. [2] Percentages do not add up to 100 because RHS data included 10 properties that were of an unknown classification (e.g., on-farm or off-farm) or nonlabor housing properties. [3] Experts in our group consisted of FLH borrowers and property managers, FLH property developers, staff from nonprofit organizations, researchers, and USDA contractors that provide technical assistance to developers. For more information on the selection of experts, please see appendix I. [4] RHS may award both loan and grant funds in the FLH program, and it may award both types of funds to one recipient. Therefore, we refer to recipients as borrowers throughout this report, as RHS does in its management handbooks and FLH regulation. [5] GAO, Standards for Internal Control in the Federal Government, [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: November 1999). [6] 42 U.S.C. §§ 1484, 1486. RHS is a mission area in USDA's Office of Rural Development and administers most federal rural housing programs. The FLH program is the only program in RHS that does not have to meet rural eligibility criteria--that is, it funds properties in both urban and rural areas. [7] For the FLH program, farm labor is defined as a service or services in connection with cultivating the soil or raising or harvesting any agriculture or aquaculture commodity; or in catching, netting, handling, planting, drying, packing, grading, storing, or preserving in the unprocessed stage, without respect to the source of employment (but not self-employed), any agriculture or aquaculture commodity; or delivering to storage, market, or a carrier for transportation to market or to processing any agricultural or aquaculture commodity in its unprocessed stage. [8] RHS loan servicers, who generally operate in local offices, conduct a variety of off-site monitoring activities, or desk reviews, and on-site supervisory reviews to assess whether a property is managed in accordance with the goals and objectives of the FLH program. [9] Occupancy requirements and income restrictions for off-farm properties do not apply to on-farm properties, as they are owned by farmers with the purpose of providing housing for their specific employees only. On-farm FLH borrowers are expected to manage their own properties and are required to maintain a lease or employment contract with each tenant specifying employment with the borrower as a condition for continued occupancy. [10] Three different income limits are used to establish eligibility for the FLH program: (1) the very low-income limit is established at approximately 50 percent of the median income for the area, adjusted for household size; (2) the low-income limit is established at approximately 80 percent of the median income for the area, adjusted for household size; and (3) the moderate-income limit is established by adding $5,500 to the low-income limit for each household size. [11] This number does not include hired laborers employed in agricultural processing activities, such as canning fruits and vegetables. For more information on data that relate to farmworkers, see appendix III. [12] RHS requirements and procedures for originating FLH loans are often similar to those of the Section 515 loan program. [13] In general, an obligation is a definite commitment that creates a legal liability of the government for the payment of goods and services ordered or received. An agency makes an obligation, for example, when it places an order, signs a contract, awards a grant, purchases a service, or takes other actions that require the government to make payments to the public or from one government account to another. An unliquidated obligation is the amount of outstanding liability for goods and services ordered and obligated but not yet received. [14] A cohort is defined as all direct loans or loan guarantees of a program for which a subsidy appropriation is provided for a given fiscal year. For direct loans for which multi-year or no-year appropriations are provided, such as the Section 514 FLH Loan Program, the cohort is defined by the year of obligations. [15] Present value is the worth of the future stream of cash inflows and outflows, as if they had occurred immediately. In calculating present value, prevailing interest rates provide the basis for converting future amounts into their "money now" equivalents. Net present value is the present value of estimated future cash inflows minus the present value of estimated future cash outflows. [16] We discuss RHS management of program information, (including information related to demand for housing and funding, monitoring of property condition and program compliance, and tenant eligibility) in greater detail in the next section of this report. [17] According to USDA data, approximately 35 percent of farm workers hired directly by farm operators lived in California, Florida, Oklahoma, and Texas throughout the year in 2010. [18] Housing Assistance Council, USDA Section 514/516 Farmworker Housing: Existing Stock and Changing Needs (Washington, D.C.: October 2006). [19] William Kandel, Profile of Hired Farmworkers, a 2008 Update, Economic Research Report No. 60, Economic Research Service, U.S. Department of Agriculture (June 2008). [20] Rental subsidies, which are funded through the Section 521 Rental Assistance program and provided to property owners through multiyear contracts, are intended to limit rent payments to 30 percent of the household's adjusted monthly income. Only off-farm FLH properties are eligible for rental assistance subsidies. [21] According to RHS, C grades increased as the result of a recent change to its classification system. RHS added a process whereby certain instances of noncompliance, such as a missed loan payment, would trigger an automatic C grade in the classification system. [22] The states with more than 100 units were Arizona, Arkansas, California, Colorado, Florida, Idaho, Michigan, New Mexico, New York, North Carolina, Oregon, Texas, and Washington. [23] See [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. [24] The resolution process begins when audit or other review results are reported to management and is completed only after action has been taken that corrects identified deficiencies, produces improvements, or demonstrates the findings and recommendations do not warrant management action. See [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. [25] In response to the results of its internal review, in January 2010 RHS's national office issued guidance to state and local offices to evaluate findings in MFIS to ensure up-to-date, accurate information, and review a missing data report on each property. RHS also held four Web-based trainings in 2009 to improve the integrity of MFIS. [26] [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. [27] During our inspections of FLH properties, we rated aspects of property exteriors and interiors by the level of disrepair we observed (that is, no, low, medium, or high levels of apparent disrepair). We assigned a rating of zero (no disrepair) when no physical condition problems were noted with an aspect of a property's interior or exterior. We assigned a rating of one (low) to properties with minor physical condition problems such as worn or older aspects of a unit's interior or exterior. We assigned a rating of two (medium) to properties with more advanced physical condition problems that appeared to have deteriorated over time. We assigned a rating of three (high) to properties with physical condition problems that could affect the health and safety of residents. [28] Housing agencies are the type of entity eligible for federal public housing funds administered by HUD, whereas FLH borrowers may be public agencies, nonprofit entities, individual farmers, or other types of organizations. The Capital Fund provides funds to housing agencies for the development, financing, and modernization of public housing developments and for management improvements. On February 23, 2011, HUD released an interim rule to make changes to PHAS. These changes became effective on March 25, 2011. 24 CFR Parts 902 and 907. [29] For more information on PHAS, see GAO, Public Housing: New Assessment System Holds Potential for Evaluating Performance, [hyperlink, http://www.gao.gov/products/GAO-02-282] (Washington, D.C.: Mar. 15, 2002); and Public Housing: HUD's Oversight of Housing Agencies Should Focus More on Inappropriate Use of Program Funds, [hyperlink, http://www.gao.gov/products/GAO-09-33] (Washington, D.C.: June 11, 2009). [30] GAO, Grants Management: Enhancing Performance Accountability Provisions Could Lead to Better Results, [hyperlink, http://www.gao.gov/products/GAO-06-1046] (Washington, D.C.: Sept. 29, 2006). [31] For more information on graduated penalties, see GAO, Federal User Fees: Key Aspects of International Air Passenger Inspection Fees Should Be Addressed Regardless of Whether Fees Are Consolidated, [hyperlink, http://www.gao.gov/products/GAO-07-1131] (Washington, D.C.: Sept. 24, 2007). [32] [hyperlink, http://www.gao.gov/products/GAO-06-1046]. [33] We considered permanent residency cards that were expired or were deemed invalid through third party verification questionable. [34] GAO, Rural Housing Service: Updated Guidance and Additional Monitoring Needed for Rental Assistance Distribution Process, [hyperlink, http://www.gao.gov/products/GAO-04-937] (Washington, D.C.: Sept. 13, 2004). [35] There may be limitations with these data sources and these data may not fully capture the extent and dynamics of tenant demand. For example, in one state we visited, according to the local RHS office, a market study submitted with an application documented a sufficient number of eligible residents; however, once the property opened, there were few farm workers to fill the units. [36] [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. [37] No properties were delinquent for fewer than 90 days. In comparison, about 8.5 percent of Federal Housing Administration single family loans were seriously delinquent, or 90 or more days delinquent as of September 2010, according to a report published by HUD. The overall delinquency rate for properties in RHS's Section 515 Rural Rental Housing Program as of the end of fiscal year 2010 was about 3 percent. [38] A finding is recorded in MFIS when the agency finds that a borrower is not operating in accordance with the loan or grant agreement, with agency regulations, or with applicable local, state, or federal laws. [39] The reserve account requirement excludes on-farm properties with fewer than 12 units. [40] Rental assistance owed to the borrower can either offset the payment owed on the loan, or even exceed the loan payment amount, resulting in RHS remitting payment to the borrower. [41] RHS is located within USDA's RD mission area. The FLH credit subsidy estimates and reestimates are prepared by RD's budget division. The $11.8 million was calculated, in part, using the FLH loan program allocation amount as described in the conference report for the fiscal year 2010 Department of Agriculture appropriation. [42] Federal agencies use OMB's credit subsidy calculator to calculate the subsidy cost of direct loan and loan guarantee programs for budget and financial reporting purposes. The subsidy cost is the net present value of estimated payments the government makes less estimated amounts it receives over the life of the direct loan or loan guarantee, excluding administrative costs, as described in the background section of this report. [43] When the reestimate is reflected in the financial statements and budget, the reestimate amount will be adjusted for the obligations that were not disbursed by the end of fiscal year 2010. As a result, the recorded reestimate will be less than $3 million. The remaining impact of the reestimate will be recorded as the remaining obligations are disbursed in the future. [44] Congress may place specific limits on the total obligations that can be made by a program. The appropriated subsidy level and the estimated subsidy rate combine to produce the loan level. Specifically, subsidy budget authority divided by subsidy rate equals supportable loan level. For example, $9,873,000 in budget authority for FLH loan obligations was described in the conference report for the fiscal year 2010 Department of Agriculture appropriation. Based on the estimated credit subsidy rate of 36.14 percent, RD would be allowed to obligate $27.3 million of direct loans ($9,873,000/0.3614). Given the same amount of budget authority for subsidy costs, an agency would be able to obligate more funding for direct loans when the credit subsidy rate is lower. As a result, had RD correctly considered defaults in its credit subsidy estimate for the loans obligated in 2010, it would have used $3 million less of its budget authority for subsidy costs and could have obligated an additional $11.8 million of direct loans. [45] The fiscal year 2010 predicted recovery rate was 67 percent. The fiscal year 2008 and 2009 cohorts both had predicted default rates of 56 percent and predicted recovery rates of 61 and 59 percent, respectively, which contributed to relatively high predicted default costs for these years (see appendix V). Our audit scope did not include an assessment of the fiscal year 2008 and 2009 credit subsidy models, and since these cohorts report a negative "all other" expense that offsets the default costs, we are not reporting on the accuracy of the estimates in these years. [46] FASAB Federal Financial Accounting and Auditing Technical Release No. 6 (January 2004). [47] Unliquidated obligations are outstanding obligations or liabilities that have not yet been paid. [48] According to agency obligation reports, between the fiscal years 2004 and 2010, the balance of unliquidated loans and grants ranged from a low of $127 million in 2004 to a high of $184 million in 2010. [49] The Multi-Family Housing Revitalization Demonstration Program is intended to restructure selected existing RHS FLH and Rural Rental Housing loans and grants to ensure that sufficient resources are available to revitalize these properties. [50] Fund appropriated in 1994 or later were "no-year" funds and do not expire after 1 or multiple years. For funds appropriated to the FLH program in 1993 or earlier, the appropriations for the costs of the Section 514 direct loan program were 1-year appropriations. Because the appropriation was available only for that fixed period under the terms of 31 U.S.C. 1552(a), RHS had 5 years from the end of the 1-year period of availability to liquidate the obligations. [51] Generally unnumbered letters issued by RHS's national office only clarify existing rules or regulations and do not set new guidelines regarding policies and procedures. [52] Although the previously established de-obligation time frames are not currently in force, FLH obligations are still monitored by program staff semiannually, as part of an agencywide requirement. USDA directs its chief financial officer to have program staff review and certify unliquidated obligations quarterly in order to properly document obligation balances and deobligate any unliquidated obligations found to be either unnecessary, or for which a bona fide purpose for the obligation and justification for the period of inactivity do not exist. According to RHS, to comply with the regulation, on a semiannual basis finance staff furnish a list of all unliquidated obligations more than 6 months old to RHS state offices and state directors then review the list and certify that the listed obligations are valid, including those obligations associated with the FLH program. See USDA Departmental Regulation 2230-001 (Apr. 21, 2009). According to RHS, it obtained a waiver to the regulation in 2009 to perform the reviews of unliquidated obligations semiannually instead of quarterly. [53] We invited 12 experts, but 1 invitee was unable to attend. [54] Findings from site visits and tenant file reviews cannot be generalized across the FLH portfolio. [55] RHS allowed one off-property in Texas to rent to tenants who are normally ineligible under the FLH program. Therefore, no tenant files were selected from this property. A second property in Texas had only five FLH tenants; therefore, we reviewed all five tenant files. Finally, a third property in New York had only six FLH tenants and all six tenant files were reviewed. [56] Once an FLH project is approved, borrowers must establish a replacement reserve account with funding levels sufficient to meet the major capital needs of a property over its life, such as replacing the roof or windows, doing major exterior work, and adding new kitchen fixtures. The aggregate, fully funded reserve amount must equal at least 10 percent of the greater of the total development cost or appraised value, and annual contributions must be a minimum of 1 percent of the total development cost. RHS requires that borrowers submit annual property budgets to the agency for approval, identify major maintenance and replacement needs during the annual budget cycle, and develop a schedule for making withdrawals from the reserve account, and, in the case of larger properties, submit annual audited financial statements. [57] This figure is a rounded average of the four quarterly FLS report figures for hired farm workers in 2010. [58] A total of $19,746,000 was appropriated to the program for the 2010 fiscal year, and according the Conference Report for the 2010 Appropriations Act, about $9.9 million of which was available for Section 514 loan subsidies and $9.9 million of which was available for Section 516 grants. Subject to the availability of funding, RHS has the ability to adjust loan and grant levels. [End of section] GAO's Mission: The Government Accountability Office, the audit, evaluation and 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 GAO's Web site [hyperlink, http://www.gao.gov]. Each weekday, GAO posts newly released reports, testimony, and correspondence on its Web site. To have GAO e-mail you a list of newly posted products every afternoon, go to [hyperlink, http://www.gao.gov] and select "E-mail Updates." Order by Phone: The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s Web site, [hyperlink, http://www.gao.gov/ordering.htm]. Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537. Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information. To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: E-mail: fraudnet@gao.gov: Automated answering system: (800) 424-5454 or (202) 512-7470: Congressional Relations: Ralph Dawn, Managing Director, dawnr@gao.gov: (202) 512-4400: U.S. Government Accountability Office: 441 G Street NW, Room 7125: Washington, D.C. 20548: Public Affairs: Chuck Young, Managing Director, youngc1@gao.gov: (202) 512-4800: U.S. Government Accountability Office: 441 G Street NW, Room 7149: Washington, D.C. 20548: