This is the accessible text file for GAO report number GAO-13-542 entitled 'Federal Housing Administration: Improving Disposition and Oversight Practices May Increase Returns on Foreclosed Property Sales' which was released on July 22, 2013. 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: AGAO: Report to Congressional Requesters: June 2013: Federal Housing Administration: Improving Disposition and Oversight Practices May Increase Returns on Foreclosed Property Sales: GAO-13-542: GAO Highlights: Highlights of GAO-13-542, a report to congressional requesters. Why GAO Did This Study: With mortgage foreclosures at historic levels in recent years, FHA is faced with disposing of a high volume of REO properties. The enterprises, other federal agencies, and private sector mortgage servicers also dispose of REO properties from their foreclosures. To assess the relative effectiveness of FHA’s REO dispositions, GAO examined (1) FHA’s disposition goals, strategies, practices, and effectiveness in disposing of properties compared with those of the enterprises and private servicers; and (2) FHA’s oversight of the contractors that maintained and marketed its REO properties. GAO analyzed REO disposition data from FHA and the enterprises, including modeling to control for property differences across the entities. GAO also collected information from program requirements, policies, and interviews on each entity’s oversight of its REO dispositions. What GAO Found: The Federal Housing Administration’s (FHA) performance in selling its foreclosed properties—known as real estate-owned (REO) properties— lagged the performance of both of the government-sponsored enterprises (enterprises), Fannie Mae and Freddie Mac. FHA disposed of more than 400,000 properties from January 2007 through June 2012. Its combined 2007-2012 returns, measured by the net execution rate (net sales proceeds divided by independently assessed property values) were about 4 to 6 percentage points below the enterprises’ returns (see figure). After controlling for certain differences in their properties’ characteristics (e.g., value, location, and local market conditions), differences in combined returns between FHA and the enterprises persisted at an estimated 2 to 5 percentage points. Figure: FHA’s Aggregate Net Execution Rate Based on Independently Assessed Property Value Relative to the Enterprises’ Rates for REO Dispositions (Results Not Controlled for Property Characteristics), January 2007 through June 2012: [Refer to PDF for image: vertical bar graph] Percentage points difference: Year: 2007; FHA performance relative to Enterprise 2: 1.52%; FHA performance relative to Enterprise 1: 1.66%. Year: 2008; FHA performance relative to Enterprise 2: -5.12%; FHA performance relative to Enterprise 1: -3.15%. Year: 2009; FHA performance relative to Enterprise 2: -8.94%; FHA performance relative to Enterprise 1: -8.58%. Year: 2010; FHA performance relative to Enterprise 2: -5.64%; FHA performance relative to Enterprise 1: -2.35%. Year: 2011; FHA performance relative to Enterprise 2: -8.67%; FHA performance relative to Enterprise 1: -4.95%. Year: 2012; FHA performance relative to Enterprise 2: -6.37%; FHA performance relative to Enterprise 1: -1.67%. Year: 2007-2012 combined; FHA performance relative to Enterprise 2: -6.49%; FHA performance relative to Enterprise 1: -3.83%. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. [End of figure] Further, while the enterprises took an average of around 200 days after foreclosure to dispose of REO properties, FHA took about 340 days—more than 60 percent longer (see figure). A similar pattern persisted even after controlling for certain property differences. FHA also took longer than the Department of Veterans Affairs (VA). For FHA, unlike the others, a significant part of the time between the foreclosure sale and REO sale is taken by loan servicers who must complete certain activities before conveying title to FHA. In the first half of 2012, FHA’s disposition returns and timelines generally improved relative to the enterprises’. All three entities use similar strategies to dispose of their REO properties, but FHA does not use some practices that the enterprises and private mortgage servicers use that may have the potential to improve its sales performance. For example, FHA does not repair its properties to increase their marketability, something both enterprises do. And unlike the enterprises, FHA does not incorporate information from multiple sources in setting list prices or consistently take into account market conditions when reducing prices. Instead, it relies on one appraisal in setting initial prices and often reduces them by set amounts. GAO found that if FHA’s execution rate and disposition time frame had equaled those of the enterprises in 2011, it could have increased its proceeds by as much as $400 million and decreased its holding costs—which can include items such as taxes, homeowners’ association fees, and maintenance costs—by up to $600 million for the year. Figure: Average Disposition Timelines by Entity for REO Dispositions (Results Not Controlled for Property Characteristics), January 2008 through June 2012 (All Years Combined): [Refer to PDF for image: 4 timelines] FHA: Foreclosure to REO: 340 days; REO Acquisition: 161 days; Initial valuation: 184 days. Enterprise 2: Foreclosure to REO: 208 days; Initial valuation: 66 days. Enterprise 1: Foreclosure to REO: 201 days; Initial valuation: 69 days. VA[A]: Foreclosure to REO: 268 days. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. [End of figure] In addition, FHA’s oversight of the contractors that it uses to maintain and dispose of REO properties has weaknesses, and it does not use some of the oversight tools other entities use that might prove effective. First, government internal control standards require complete, updated policies and procedures to guide program oversight. But FHA has not updated its REO disposition handbook since 1994, even though the agency implemented a different program and contractor structure in 2010. In the absence of a central source of updated guidance, GAO and FHA internal auditors found inconsistencies in both contractor activities and staff oversight across FHA’s four regional homeownership centers. Second, FHA has not implemented a uniform system for evaluating contractor performance. For instance, FHA has yet to implement a proposed version of the type of scorecard that the enterprises use to assess differences in contractor performance. Also, its planned incentive structure for contractors has been found not to comply with federal contracting rules. These two shortcomings have prevented FHA from assigning work according to contractors’ performance—a key quality control in its new REO program structure. Further, FHA aims to inspect 2 to 6 percent of its REO properties annually, although other entities with REO properties report inspecting between 25 and 35 percent monthly, or between 7 and 40 percent annually. Finally, FHA has not taken steps to ensure that the listing brokers marketing its REO properties are located close enough to the properties to have adequate knowledge of local markets. Without implementing more effective activities to evaluate contractor performance and ensure compliance with program requirements, FHA’s REO properties may continue to remain on the market longer and sell for lower prices than properties held by the enterprises. What GAO Recommends: GAO makes 10 recommendations intended to increase FHA’s returns on the disposition of REO properties, including considering repairs that increase net proceeds, requiring the use of additional information for setting initial and subsequent listing prices; and improving its oversight of its contractors, including updating and maintaining comprehensive guidance, implementing a performance scorecard, increasing in-person inspections, and ensuring that listing brokers are appropriately located. FHA reviewed a draft of this report and agreed with GAO’s recommendations. View [hyperlink, http://www.gao.gov/products/GAO-13-542]. For more information, contact Mathew J. Scirè at (202) 512-8678 or sciremj@gao.gov. [End of section] Contents: Letter: Background: Changes to Its Disposition Practices Could Help FHA Improve Results: FHA Could Further Improve Its Monitoring and Evaluation of REO Contractors: Conclusions: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Objectives, Scope, and Methodology: Appendix II: Comments from the Department of Housing and Urban Development: Appendix III: GAO Contact and Staff Acknowledgments: Table: Table 1: Number of REO Dispositions for FHA and Other Entities from January 1, 2007 through June 30, 2012: Figures: Figure 1: Typical Foreclosure Processes: Figure 2: FHA Homeownership Center Operating Regions: Figure 3: REO Property Dispositions by Various Housing Program Entities, First Quarter 2007 through Second Quarter 2012: Figure 4: FHA's Aggregate Net Execution Rate Based on Independently Assessed Property Value Relative to the Enterprises' Rates for REO Dispositions, January 2007 through June 2012: Figure 5: FHA's Aggregate Net Execution Rate Based on Initial List Price Relative to the Enterprises' Rates for REO Dispositions, January 2007 through June 2012: Figure 6: Average Disposition Timelines by Entity for REO Dispositions, January 2008 through June 2012 (All Years Combined): Figure 7: Average Disposition Timelines by Entity for REO Dispositions, January 2008 through June 2012 (by Year): Figure 8: Sales of REO Properties by Type of Sales Method, January 2007 through June 2012: Figure 9: Aggregate Share of REO Dispositions by Type of Buyer for All Sales, January 2007 through June 2012: Figure 10: Differences in FHA Homeownership Centers' and the Enterprises' Aggregate Net Execution Rates Based on Independently Assessed Value for All REO Dispositions, January 2007 through June 2012: Figure 11: Difference between FHA Homeownership Centers' Average Disposition Times and the Enterprises' Times for All REO Dispositions, January 2008 through June 2012: Abbreviations: AVM: Automated Valuation Model: BPO: Broker's Price Opinion: FAR: Federal Acquisition Regulation: FHA: Federal Housing Administration: HUD: Department of Housing and Urban Development: M&M: Management and Marketing: OGC: Office of General Counsel: OIG: Office of Inspector General: REO: Real estate-owned: RHS: Rural Housing Service: VA: Department of Veterans Affairs: [End of section] GAO: United States Government Accountability Office: 441 G St. N.W. Washington, DC 20548: June 20, 2013: The Honorable Randy Neugebauer: Chairman: Subcommittee on Housing and Insurance: Committee on Financial Services: House of Representatives: The Honorable Spencer Bachus: House of Representatives: The Federal Housing Administration (FHA) within the Department of Housing and Urban Development (HUD) strives to assist low-and moderate- income borrowers obtain home loans by providing guarantees to private lenders. In 2012, FHA insured about 1.2 million single-family mortgages, providing about $227 billion in mortgage insurance. With the high rate of mortgage loan defaults since 2006, FHA has had to take possession of a large number of foreclosed properties.[Footnote 1] FHA's inventory of these real estate-owned (REO) properties grew from over 25,000 in 2007 to a peak of more than 65,000 by the end of 2011, an increase of about 160 percent. Other government agencies that operate housing finance programs, including the Department of Veterans Affairs (VA) Loan Guaranty Service and the Department of Agriculture's Rural Housing Service (RHS), as well as the two government-sponsored enterprises (the enterprises), Fannie Mae and Freddie Mac, also have experienced increasing numbers of REO properties. All of these entities seek to sell or otherwise dispose of their REO properties in order to recover some portion of the loss on the original failed loans.[Footnote 2] REO properties must be maintained, repaired if necessary, evaluated to determine an appropriate selling price, and marketed for sale. Conducting these activities effectively can ensure that FHA maximizes returns and minimizes maintenance and other holding costs. You requested that we examine various aspects of FHA's REO activities. This report examines (1) how FHA's disposition goals, strategies, and practices and effectiveness in disposing of properties compares with those of other entities; and (2) how FHA oversees the contractors that maintain and market its REO properties. To identify the goals and strategies for REO programs at FHA and other entities, we reviewed program regulations, requirements, and policies. To assess the effectiveness of FHA's REO property dispositions, we analyzed disposition data from FHA and the other federally related entities, including all REO properties disposed of from January 2007 through June 2012.[Footnote 3] We did not include RHS in our analysis of REO property disposition performance because it only obtains and manages REO properties through its direct loan program and it had a comparatively small number of such dispositions. During the period of our analysis RHS's direct loan property dispositions were less than 1 percent of FHA's dispositions. RHS also did not have property-level data available for many of the data elements that we included in our analyses. Similarly, we were not able to obtain some information on VA's REO dispositions because it was in the process of transitioning to a new REO property management contractor. We created regression models to assess the extent of differences between FHA's performance and that of the two enterprises after accounting for some differences in the characteristics of the properties they acquired. We also used these models to identify factors that could explain differences in performance. We determined that the REO disposition data were sufficiently reliable for the purposes of our analysis. To determine how FHA and the other entities oversaw the contractors that managed and disposed of their REO properties, we reviewed program regulations, requirements, and policies and interviewed staff from these entities as well as from some private sector mortgage servicers that also acquired and disposed of REO properties.[Footnote 4] We conducted this performance audit from February 2012 to June 2013 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. For more information on the methodologies used in this report, see appendix I. Background: Established by the National Housing Act of 1934, FHA's single-family mortgage insurance program helps home buyers obtain home mortgages by providing insurance on single-family mortgage loans. The mortgage insurance allows FHA-approved private lenders to provide qualified borrowers with mortgages on properties with one to four housing units and generally compensates lenders for nearly all of the losses incurred on such loans. FHA insures mortgages on properties that meet its criteria, providing guarantees for initial purchases, construction and rehabilitation, and refinancing. To support the program, FHA imposes up-front and annual mortgage insurance premiums on home buyers.[Footnote 5] The agency has played a particularly large role among minority, low-income, and first-time home buyers. In 2012, about 78 percent of FHA-insured home purchase loans went to first-time home buyers, about 32 percent of whom were minorities. Other Mortgage Market Participants: A number of other federal and private sector entities participate in the mortgage market. Along with FHA, the VA Loan Guaranty Service and RHS administer federal government programs that insure or guarantee single-family mortgages made by private lenders. In addition to these government agencies, private companies insure lenders against losses on home mortgages, and private lenders make loans without mortgage insurance. The enterprises also participate in the U.S. housing market by purchasing mortgages from lenders. VA's Loan Guaranty program is an entitlement program that provides eligible veterans, active duty military personnel, and certain other individuals with housing benefits. The VA guaranty program allows mortgage lenders to extend loans to eligible borrowers on favorable terms--for example, with no down payment--and provides lenders with financial protections against the losses associated with such mortgages. To help support the program, borrowers are required to pay a funding fee that equals a certain percentage of the loan amount, although service-connected disabled veterans are exempt from paying this fee. The program may also receive congressional appropriations if needed. RHS operates guaranteed and direct loan programs to help rural Americans with very low incomes, low incomes, and in some cases moderate incomes purchase single-family homes. The purpose is to provide financing with no or low down payments at favorable rates and terms. The loans are generally for the purchase, construction, rehabilitation, or relocation of a dwelling and related structures. RHS-guaranteed loans are made through approved local lenders, with RHS providing the lenders substantial financial protections against associated losses. The loans are available to qualifying borrowers who meet applicable household income limits and seek to buy properties in eligible rural areas. Under its direct loan program, RHS extends loans to qualified borrowers--who must have low incomes and be without adequate housing--for the purchase of properties that are modest in size, design, and cost. Congress established the enterprises as for-profit, shareholder-owned corporations. They share a primary mission to stabilize and assist the U.S. secondary mortgage market and facilitate the flow of mortgage credit. To accomplish this goal, the enterprises purchase conventional mortgages that meet their underwriting standards, obtaining their funds through borrowing or by issuing mortgage-backed securities, which are securities backed by pools of mortgages.[Footnote 6] The enterprises hold some of the purchased mortgages in their portfolios, but they package most of them for sale to investors in the secondary mortgage market. In exchange for a fee, the enterprises guarantee these investors the timely payment of principal and interest. Both enterprises are also required to purchase mortgages that serve low-and moderate-income families. On September 6, 2008, the Federal Housing Finance Agency placed the enterprises into conservatorship out of concern that the enterprises' deteriorating financial condition threatened the stability of financial markets. Institutions that originate home mortgages generally do not hold such loans as assets on their balance sheets but instead sell them to other financial institutions for the purpose of securitizing the mortgage. These securities pay interest and principal to their investors, which include financial institutions and pension funds, among others. In the past, institutions originating mortgage loans took care of all the activities associated with servicing them--including accepting payments, initiating collection actions for delinquent payments, and foreclosing if necessary. With the advent of securitization, entities known as mortgage servicers--which can be large mortgage finance companies or commercial banks--typically undertake such activities on behalf of the current owners of the loans. Foreclosure Process: If a borrower defaults on a mortgage loan secured by the property, the mortgage note holder is generally entitled to pursue foreclosure to obtain title to the property. The foreclosure process is governed by state laws and differs across states, but foreclosed properties are typically sold at auction, as shown in figure 1. Once the borrower is in default, the mortgage servicer--often in conjunction with the borrower and entities with an interest, such as mortgage guarantors and insurers--must decide whether to pursue a home retention workout or other foreclosure alternative or to initiate foreclosure.[Footnote 7] The mortgage owner or servicer generally initiates foreclosure once the loan becomes 90 days or more delinquent unless the borrower can resolve the loan's delinquency by paying the outstanding amount or some other resolution occurs, such as a borrower repayment plan or loan modification. If the foreclosure process is completed and no third party purchases the home at the foreclosure sale, the home usually becomes the property of the loan holder or servicer as part of an REO inventory. Figure 1: Typical Foreclosure Processes: [Refer to PDF for image: process illustration] 1. Borrower Misses payments. 2. Workout period: Servicer and borrower may pursue loss mitigation and home retention strategies. 3. Notice of default: Written notice to borrower that there has been a default and legal action is possible. Judicial foreclosure: Involves a judge or court official that presides over the case: 4A. Foreclosure initiation: Servicer initiates formal foreclosure action by filing a lawsuit through court. 5A. Judgment: Judge grants servicer right to dispose of property, and schedules the foreclosure sale. 6A. Foreclosure sale: Property is sold to a third party through an auction or conveyed to the servicer. Cancel/reschedule: Servicer may cancel or reschedule foreclosure sales. 7A. Possible redemption period. For specified period of time, the borrower may reclaim his/her property by matching the winning bid at the foreclosure sale. Statutory foreclosure: Without court action but in accordance with state law (also called “nonjudicial” or “power-of-sale”): 4B. Foreclosure initiation: Servicers are required to publish the notice of foreclosure or sale in accordance with state law. 5B. Foreclosure sale: Property is sold to a third party through an auction or conveyed to the servicer. Cancel/reschedule: Servicer may cancel or reschedule foreclosure sales. 6B. Possible redemption period. For specified period of time, the borrower may reclaim his/her property by matching the winning bid at the foreclosure sale. Sources: GAO (analysis); Art Explosion (images). [End of figure] However, certain states provide the previous owners of foreclosed properties with a right of redemption that allows them to pay amounts owed to the lender and reclaim ownership. During redemption periods, the previous borrower or current occupant is allowed to remain in the residence and the REO property owner or servicer generally is not permitted to pursue activities such as evicting property residents or securing properties. Some states may have a confirmation process to complete the foreclosure process and transfer title that may also delay possession and marketing of an REO property. Typically, the redemption or confirmation period begins after the foreclosure sale and lasts from around 1 to 6 months or more. However, if properties become vacant, some state laws may permit the shortening of redemption periods and allow the REO property owners or servicers to take control of foreclosed properties. The acquisition of REO properties differed across the entities that we reviewed. When a servicer forecloses on an FHA-insured property that is not sold to a third party, the foreclosure is held in the lender's or servicer's name, and the lender or servicer is responsible for the property until it is conveyed to FHA. FHA requires servicers to oversee properties during redemption periods, to evict residents if properties not in redemption periods are occupied, and to perform critical maintenance on properties. The servicer files a claim with FHA, and FHA conducts its inspections before accepting the title. The length of time between the foreclosure sale and entry of a property into FHA's REO inventory depends in part on state foreclosure laws as well as the actions of the loan's servicer. According to FHA rules, a servicer needs marketable title before it can convey a property to FHA, and the title is generally considered to be marketable only after the borrower has left the home or been evicted, any redemption period has expired, and other required actions have taken place. After conveyance to FHA, the property is assigned to FHA's contractors, which begin the process of preparing the REO property for sale. Other federally related entities that acquire REO properties take custody of and are responsible for them closer to the time of the foreclosure sale. For example, the enterprises require servicers to convey properties to them within 24 hours of foreclosure sales, while VA requires servicers to provide notice of their intent to convey properties within 15 days of foreclosure sales, although servicers have 60 days (or longer in certain jurisdictions) to provide evidence of acceptable title to conveyed properties. RHS's REO process varies, depending on whether the property has had a guaranteed or direct loan. With direct loans, RHS takes possession of the property after the foreclosure sale and manages the entire REO process. With guaranteed loans, the lender receives title to the property and maintains, markets, and disposes of the property, and RHS oversees the process. Because RHS acquires and disposes only of REO properties related to its direct loans, we generally considered only these types of properties when discussing RHS's REO disposition. REO Disposition Process: Disposing of REO properties can involve various activities, although the disposition process is generally similar across entities. While a property is held as part of an entity's REO inventory, the entity is responsible for maintenance, including cleaning, lawn care, snow removal, and security. If a property has a tenant or is otherwise occupied, eviction proceedings may need to occur before it is offered for sale.[Footnote 8] After any redemption periods expire and evictions take place, properties are usually assessed to determine their market value. The market value is used to determine the selling, or listing, price. A listing real estate broker is usually chosen to market the property publicly, generally through a multiple listing service system--a database set up by a group of real estate brokers to provide information about properties for sale. If a property does not attract interest at its initial listing price, the price can be reduced. Once a purchase offer is accepted, the sale closing process occurs, and ownership is transferred to the new owner. REO Program Structures and Use of Contractors: Since 1999, FHA has been outsourcing to private sector contractors the maintenance and disposition of its REO properties. Entities that dispose of REO properties typically use various types of contractors, including those that manage the marketing or maintenance activities for a large number of properties. These larger contractors often use subcontractors to provide specific services related to the marketing or maintenance of properties, such as listing the property for sale or cutting the lawn. In June 2010, FHA launched the third generation of its Management and Marketing (M&M) contractor program, known as M&M III. Under prior arrangements, FHA's M&M contractors were responsible for both the maintenance and marketing of FHA's REO properties. However, under M&M III these functions are performed by separate contractors, including maintenance contractors that are responsible for preserving properties and marketing contractors that are responsible for selling the homes.[Footnote 9] Under the M&M III structure, FHA also uses contractors called Mortgagee Compliance Managers that protect FHA's interests in foreclosed properties that lenders have not yet conveyed to FHA.[Footnote 10] The M&M III structure was also meant to include an additional contractor to serve as an oversight monitor to assist FHA in overseeing its REO program's performance, including the other M&M III contractors. FHA conducts its mortgage loan insurance programs and its REO disposition program through four regional operating locations called homeownership centers, or HOCs. The homeownership centers are located in: * Atlanta, Georgia; * Denver, Colorado; * Philadelphia, Pennsylvania; and: * Santa Ana, California. Each of these homeownership centers oversees FHA operations in the states within their region (see figure 2). Under M&M III, FHA contracts with entities to provide the necessary maintenance or marketing services in multiple areas that cover several states. Within each of these contract areas, multiple entities generally will receive either maintenance or marketing contracts as a way of fostering competition among the contractors to improve their responsiveness, reduce risk, and increase net returns to FHA. FHA staff--known as government technical representatives--in each of these locations oversee the maintenance and marketing contractors, including monitoring and evaluating their performance and providing technical guidance and assistance. Figure 2: FHA Homeownership Center Operating Regions: [Refer to PDF for image: illustrated U.S. map] Philadelphia region: Connecticut; Delaware; District of Columbia; Maine; Maryland; Massachusetts; Michigan; New Hampshire; New Jersey; New York; Ohio; Pennsylvania; Rhode Island; Vermont; Virginia; West Virginia. Atlanta region: Alabama; Caribbean; Florida; Georgia; Illinois; Indiana; Kentucky; Louisiana; Mississippi; North Carolina; South Carolina; Tennessee. Denver region: Arkansas; Colorado; Iowa; Kansas; Minnesota; Missouri; Montana; Nebraska; New Mexico; North Dakota; Oklahoma; South Dakota; Texas; Utah; Wisconsin; Wyoming. Santa Ana region: Alaska; Arizona; California; Hawaii; Idaho; Nevada; Oregon; Washington. Sources: HUD and Map Resources (map). [End of figure] Other entities also use contractors to manage and dispose of REO properties, but to varying degrees. For example, officials from one of the enterprises said that it used three nationwide contractors for property maintenance, with each operating in different states. According to these officials, contractors inspect the properties, but field-based quality control employees also do property inspections. For disposition, this enterprise has its own internal management group that markets and sells REO properties using its own network of listing brokers. The enterprise's officials said that to help adjust capacity based on REO inventories, the enterprise had also contracted with external asset management companies that did not use its broker network. The officials noted that the enterprise recently ended its use of these contractors, as its REO inventories had declined. The other enterprise has used contractors to manage and dispose of its REO properties since 2008. According to officials, this enterprise initially used one sales management contractor but added two more in March 2011. These officials explained that each of the three contractors operated in every geographic area, allowing the enterprise to more easily change the contractor managing a property if necessary. The enterprise's officials also said that it managed the network of service providers--such as listing brokers, property maintenance companies, and repair contractors--that its sales management contractors used to ensure a consistent process and minimize potentially adverse relationships among contractors. Officials said that the enterprise primarily maintained relationships with local service providers rather than nationwide providers because markets differed at the local level. The enterprise's oversight structure includes several different groups, including oversight monitors that review specific areas of contractor performance, individual business teams that also review performance across all contractors, and vendor oversight teams that serve as liaisons between the business units and contractors. VA and RHS use different approaches. VA staff told us that the agency used a single contractor to manage the maintenance and disposition of its entire REO inventory. This contractor uses numerous subcontractors to maintain and market properties and manage the subcontractors. According to its staff, VA wanted one company to manage everything so that it could have a single point of interaction and accountability. RHS uses both contractors and its own staff to dispose of REO properties, according to agency officials. For its direct loans, RHS staff manage the disposition of any foreclosed properties for which it takes possession in about half of the states, while a central RHS office oversees contractors that manage and dispose of properties in the remaining states. Officials said that for several years RHS had been transitioning from using its own staff to using contractors to dispose of properties. For its guaranteed loans, RHS does not use contractors but relies on lenders that take possession of the REO properties with some RHS oversight. If a lender completes foreclosure or a deed-in-lieu of foreclosure on a loan that RHS guarantees, the lender receives title to the property and maintains, markets, and disposes of the property. The Housing Crisis Increased the Volume of REO Properties: As a result of the large number of loan defaults arising during the housing crisis, FHA and the other entities have generally experienced significant growth in their REO property disposition activities. As shown in the figure below, each of the entities disposed of an increasing number of REO properties during that time. RHS was not included in the figure because it had no more than 218 direct loan REO property dispositions in any quarter during the time period. Figure 3: REO Property Dispositions by Various Housing Program Entities, First Quarter 2007 through Second Quarter 2012: [Refer to PDF for image: multiple line graph] 2007, Q1; Enterprise A dispositions: 8,860; Enterprise B dispositions: 3,253; FHA dispositions: 11,792; VA dispositions: 1,828. 2007, Q2; Enterprise A dispositions: 9,875; Enterprise B dispositions: 3,784; FHA dispositions: 13,180; VA dispositions: 1,662. 2007, Q3; Enterprise A dispositions: 9,509; Enterprise B dispositions: 3,740; FHA dispositions: 11,061; VA dispositions: 1,558. 2007, Q4; Enterprise A dispositions: 9,526; Enterprise B dispositions: 4,315; FHA dispositions: 10,694; VA dispositions: 1,665. 2008, Q1; Enterprise A dispositions: 10,134; Enterprise B dispositions: 5,341; FHA dispositions: 10,943; VA dispositions: 2,075. 2008, Q2; Enterprise A dispositions: 12,304; Enterprise B dispositions: 8,254; FHA dispositions: 11,925; VA dispositions: 2,241. 2008, Q3; Enterprise A dispositions: 15,519; Enterprise B dispositions: 8,975; FHA dispositions: 12,905; VA dispositions: 2,171. 2008, Q4; Enterprise A dispositions: 24,400; Enterprise B dispositions: 10,233; FHA dispositions: 14,058; VA dispositions: 334. 2009, Q1; Enterprise A dispositions: 25,966; Enterprise B dispositions: 13,095; FHA dispositions: 15,013; VA dispositions: 4,524. 2009, Q2; Enterprise A dispositions: 31,305; Enterprise B dispositions: 15,615; FHA dispositions: 19,978; VA dispositions: 3,505. 2009, Q3; Enterprise A dispositions: 30,872; Enterprise B dispositions: 16,713; FHA dispositions: 19,203; VA dispositions: 3,287. 2009, Q4; Enterprise A dispositions: 32,885; Enterprise B dispositions: 19,532; FHA dispositions: 19,522; VA dispositions: 4,462. 2010, Q1; Enterprise A dispositions: 37,431; Enterprise B dispositions: 19,978; FHA dispositions: 20,285; VA dispositions: 3,596. 2010, Q2; Enterprise A dispositions: 48,838; Enterprise B dispositions: 24,060; FHA dispositions: 25,590; VA dispositions: 3,902. 2010, Q3; Enterprise A dispositions: 47,078; Enterprise B dispositions: 23,524; FHA dispositions: 22,834; VA dispositions: 3,695. 2010, Q4; Enterprise A dispositions: 49,305; Enterprise B dispositions: 23,584; FHA dispositions: 15,009; VA dispositions: 3,784. 2011, Q1; Enterprise A dispositions: 61,827; Enterprise B dispositions: 29,384; FHA dispositions: 16,622; VA dispositions: 3,465. 2011, Q2; Enterprise A dispositions: 70,069; Enterprise B dispositions: 26,759; FHA dispositions: 37,565; VA dispositions: 5,342. 2011, Q3; Enterprise A dispositions: 56,797; Enterprise B dispositions: 23,416; FHA dispositions: 32,876; VA dispositions: 3,794. 2011, Q4; Enterprise A dispositions: 50,122; Enterprise B dispositions: 21,882; FHA dispositions: 27,246; VA dispositions: 2,848. 2012, Q1; Enterprise A dispositions: 51,154; Enterprise B dispositions: 23,309; FHA dispositions: 23,587; VA dispositions: 2,439. 2012, Q2; Enterprise A dispositions: 47,847; Enterprise B dispositions: 24,101; FHA dispositions: 25,414; VA dispositions: 3,094. Sources: GAO analysis of FHA, Fannie Mae, Freddie Mac, and VA data. Note: We do not identify the identity of the enterprises for certain analyses. We use "enterprise A" and "enterprise B" when referring to the entities' disposition volumes. [End of figure] Changes to Its Disposition Practices Could Help FHA Improve Results: Our review of FHA and other federally related housing entities found that they pursued similar goals and strategies for disposing of REO properties, including seeking to sell most properties to owner- occupants--individuals planning to occupy the homes as primary residences. But data for REO dispositions from January 2007 through June 2012 showed that FHA's returns from selling its REO properties generally trailed the returns earned by the enterprises. This difference declined when differences in property characteristics, such as location and value, were considered. In addition, we found that FHA, on average, took significantly longer to sell its properties than the enterprises and VA--more than 130 and 70 additional days, respectively. We also evaluated how the method used to sell properties and the type of purchaser affected FHA's performance. The large majority of FHA's dispositions were retail sales to owner-occupants or investors, and our analysis of FHA's performance indicated that the agency achieved higher returns on sales to owner-occupants than on sales to investors and other buyers. However, FHA had a smaller share of owner-occupant sales than the enterprises. We also found that, in making these sales, FHA did not follow several practices that other entities used in disposing of REO properties that could potentially increase the agency's returns, including repairing properties to increase their market value, using multiple inputs to set list prices, and using market-based information to make subsequent list price reductions. FHA and Other Federally Related Entities Have Similar Goals and Strategies for Disposing of REO Properties: FHA's goals for disposing of its REO properties were similar to those of the other entities we reviewed. According to staff from FHA, VA, and the enterprises, each entity aims to maximize the financial return of REO dispositions while minimizing each property's time in inventory. Specifically, FHA's regulatory goals for its REO program are to dispose of properties in a manner that expands home ownership opportunities, strengthens neighborhoods and communities, and ensures a maximum return to the mortgage insurance fund.[Footnote 11] Currently, the performance of FHA's REO program is assessed against three formal goals: * reducing the average number of days from acquisition to listing REO properties for sale by 2 percent from the prior fiscal year's average, * reducing the average number of days in inventory by 2 percent from the prior fiscal year's average, and: * conducting at least 12 REO workshops/meetings to promote acquisition and reuse of foreclosed properties in Neighborhood Stabilization Program areas.[Footnote 12] While not having a formal goal addressing the expected return on its REO sales, FHA staff told us that they attempted to improve the solvency of FHA's insurance fund by targeting a gross execution rate-- that is, a property's sale price as a share of its assessed value, or list price--of 100 percent although they said they use a specific target of at least 80 percent when evaluating homeownership center performance. VA staff said that it advises the contractors that market its REO properties that their execution rate target is 80 percent-- based on a property's net sales proceeds rather than gross sales price- -but offers additional compensation incentives for a rate of 88 percent or better. One of the enterprises has a procedural manual for REO sales that notes that its goal is to sell properties for "as close to 100 percent of the list price as possible." FHA and the other entities we reviewed pursued similar strategies to achieve their goals. Staff from these entities explained that they attempted to sell their REO properties primarily via retail sales to either owner-occupants or to investors, who would likely renovate them for resale or rentals. Retail REO sales usually are conducted by private real estate brokers that market the properties the same way they market other properties--for example, listing them in a multiple listing service. Officials from these entities explained that they pursued retail sales because such sales produced higher net returns than other methods of selling the properties, such as selling multiple properties at once--bulk sales--or selling individual or multiple properties at auctions. Officials of some of these entities said that they generally pursued these alternative disposition strategies only after retail sales had proven unsuccessful and that often such sales involved low-value properties or properties with problems that made sales through retail methods difficult. Officials from FHA and other federally related housing entities said that they preferred retail sales not only because of their higher returns, but also because such sales generated more purchases by owner- occupants. Further, some of the officials and representatives of some nonprofit organizations said that owner-occupant sales were better for stabilizing communities and protecting home values. Owner-occupants are assumed to have more incentive to maintain their properties than investor owners that may be absent or focused primarily on maximizing rental income. They also said that owner-occupant sales generally yielded a higher financial return than sales to investors. FHA staff noted that selling to owner-occupants helped to achieve one of HUD's overall agency goals--to expand home ownership opportunities--and said that FHA required sales records for REO properties to reflect the type of buyer. To further promote sales to owner-occupants, FHA and some of the other entities often use methods such as exclusive access periods. For example, FHA accepts offers on properties that qualify for FHA insurance only from owner-occupants for the first 30 days of the listing.[Footnote 13] In some cases, FHA also disposes of REO properties at a discount through certain programs intended to further its agency's mission goal of creating strong, sustainable, inclusive communities and quality affordable homes. These discount sale programs represented only a small fraction of FHA's total REO dispositions from January 2007 through June 2012. They include the Asset Control Area program, discount sales to nonprofit organizations and government entities, the Good Neighbor Next Door program, and the $1 Home Sale program. * Asset Control Area. Properties located in areas that HUD has designated for revitalization based on the area's household incomes, home ownership rates, and level of FHA-insured mortgage foreclosure activity can be offered to sale to municipal government entities or approved nonprofits through this program.[Footnote 14] These properties are offered at discounts of at least 50 percent of the appraised value of the property for properties valued at over $25,000. Properties valued at less than that amount may be sold for as little as $100. * Discount sales to nonprofit organizations and government entities. Qualified nonprofit organizations and government entities may also purchase FHA properties at a discount. Discounts range from 10 to 30 percent, depending on the property's FHA insurability status, location, and other factors. For example, one initiative provides grantees participating in the Neighborhood Stabilization Program exclusive access to newly conveyed REO properties that are located in their designated areas for 2 days after the grantee is notified that an appraisal has been obtained. * Good Neighbor Next Door. Under this program, properties in revitalization areas can also be offered to police officers, teachers, fire fighters, and emergency medical technicians at 50 percent off of the list price. * $1 Home Sale. Sales of "aged inventory" (properties listed for sale for more than 6 months) can also be made for $1 plus closing costs to local governments to support local housing and community development initiatives. FHA's Returns on REO Sales Slightly Lagged Those of the Enterprises and Dispositions Took Substantially Longer to Complete: FHA's returns from selling its REO properties generally have trailed the returns of the two enterprises but its performance has improved recently. FHA also took longer to dispose of properties than the enterprises but showed recent improvement in this area. Each of these entities achieved better results when conducting retail sales of individual properties than when using other disposition methods and when selling properties to owner-occupants rather than to investors and other buyers. All of the entities, including private REO servicers, that we interviewed assessed their performance in disposing of REO properties using a variety of metrics. For example, execution rates gauge success in maximizing a property's sale prices. They can be calculated by comparing either a property's gross sales price or net sales proceeds with some measure of the property's value, such as an assessed value from an independent appraiser or the price at which the property was listed for sale.[Footnote 15] Properties that sell for larger percentages of the assessed value or listed price generate higher returns for the seller. We analyzed data from REO properties disposed of from January 2007 through June 2012 to determine FHA's and the enterprises' net execution rates based on independently assessed value, which represents the entities' net sales proceeds as a percentage of those values.[Footnote 16] Our analysis showed that the aggregate net execution rate for FHA was 4 and 6 percentage points lower depending on the enterprise (see figure 4).[Footnote 17] Figure 4: FHA's Aggregate Net Execution Rate Based on Independently Assessed Property Value Relative to the Enterprises' Rates for REO Dispositions, January 2007 through June 2012: [Refer to PDF for image: vertical bar graph] Percentage points difference: Year: 2007; FHA performance relative to Enterprise 2: 1.52%; FHA performance relative to Enterprise 1: 1.66%. Year: 2008; FHA performance relative to Enterprise 2: -5.12%; FHA performance relative to Enterprise 1: -3.15%. Year: 2009; FHA performance relative to Enterprise 2: -8.94%; FHA performance relative to Enterprise 1: -8.58%. Year: 2010; FHA performance relative to Enterprise 2: -5.64%; FHA performance relative to Enterprise 1: -2.35%. Year: 2011; FHA performance relative to Enterprise 2: -8.67%; FHA performance relative to Enterprise 1: -4.95%. Year: 2012; FHA performance relative to Enterprise 2: -6.37%; FHA performance relative to Enterprise 1: -1.67%. Year: 2007-2012 combined; FHA performance relative to Enterprise 2: -6.49%; FHA performance relative to Enterprise 1: -3.83%. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. Note: The year 2012 represents data from January 1, 2012, through June 30, 2012. The figures that we report for execution rates are aggregate calculations for individual years and the entire period rather than averages of the execution rates for individual property dispositions. For example, we calculated the aggregate net execution rate based on independent value by dividing the sum of the net sales proceeds for properties disposed during a given period by the sum of the independently assessed values for those properties. This calculation included data from 416,929 (99.91 percent of total) dispositions for FHA, 352,490 (99.90 percent) for one enterprise, and 648,966 (87.51 percent) for the other enterprise. It did not include all dispositions for each entity because of missing values for certain data elements. VA did not have the property-level data available that are necessary to calculate net execution rates. [End of figure] To account for the possibility that the performance differences were attributable to differences in the property characteristics of each of these entities' REO inventories, we developed a regression model to control for the effects on the net execution rates of certain property characteristics--such as location, value, and local real estate market conditions--that were beyond the entities' control. Based on this analysis, we estimate that FHA's aggregate independently assessed value net execution rate still trailed that of the enterprises by 2 and 5 percentage points.[Footnote 18] Similarly, FHA's aggregate net execution rate based on initial list prices for the entire period was worse than enterprise 2's and enterprise 1's by 3 and 6 percentage points, respectively, as shown in figure 5. After controlling for the effects of certain property characteristics--as we did for the independently assessed value net execution rate--we estimate that FHA's net execution rate based on initial list prices was less than that of enterprise 2 and enterprise 1 by 2 and 4 percentage points, respectively. Figure 5: FHA's Aggregate Net Execution Rate Based on Initial List Price Relative to the Enterprises' Rates for REO Dispositions, January 2007 through June 2012: [Refer to PDF for image: vertical bar graph] Percentage points difference: Year: 2007; FHA performance relative to Enterprise 2: 3.91%; FHA performance relative to Enterprise 1: 0.29%. Year: 2008; FHA performance relative to Enterprise 2: -4.3%; FHA performance relative to Enterprise 1: -10.17%. Year: 2009; FHA performance relative to Enterprise 2: -7.07%; FHA performance relative to Enterprise 1: -11.15%. Year: 2010; FHA performance relative to Enterprise 2: -0.77%; FHA performance relative to Enterprise 1: -3.14%. Year: 2011; FHA performance relative to Enterprise 2: -4.35%; FHA performance relative to Enterprise 1: -7.36%. Year: 2012; FHA performance relative to Enterprise 2: -0.2%; FHA performance relative to Enterprise 1: -0.13%. Year: 2007-2012 combined; FHA performance relative to Enterprise 2: -2.74%; FHA performance relative to Enterprise 1: -5.76%. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. Note: The year 2012 represents data from January 1, 2012, through June 30, 2012. The figures that we report for execution rates are aggregate calculations for individual years and the entire period rather than averages of the execution rates for individual property dispositions. The calculation of initial list price net execution rates included data from 416,690 (99.85 percent of total) dispositions for FHA, 352,461 (99.89 percent) for one enterprise, and 649,040 (87.52 percent) for the other enterprise. It did not include all dispositions for each entity because of missing values for certain data elements. VA did not have the property-level data available that are necessary to calculate net execution rates. [End of figure] When comparing aggregate execution rate performance based on gross sales prices rather than net sales proceeds, the difference between FHA and the enterprises widened slightly for both independently assessed values and initial list prices.[Footnote 19] Both FHA's net and gross execution rates improved relative to the enterprises during the first half of 2012 and for the list price ratio, FHA's performance was equal to that of the enterprises. Although the difference in aggregate net sales returns between FHA and the enterprises over the entire period was 6 percent or less, a small improvement in performance can yield substantial amounts of additional revenue because FHA disposes of tens of thousands of properties each year. For example, if FHA had achieved a similar net execution rate based on independently assessed value as enterprise 2, it would have received over $400 million in additional revenue for 2011 alone. [Footnote 20] And if FHA's aggregate net execution rate for 2011 had been 1 percentage point higher, FHA would have received over $90 million in additional revenue. FHA Took Longer to Sell Its REO Properties Than the Enterprises: FHA took significantly longer to sell its properties than the enterprises and VA, but the differences decreased in the first half of 2012.[Footnote 21] For each entity, we analyzed the time in days that elapsed between the date of the foreclosure sale of the property to the closing date of the REO sale.[Footnote 22] Based on this analysis, FHA took on average about 340 days from the foreclosure sale to sell its REO properties based on dispositions from January 2008 through June 2012, compared with just over 200 days for the enterprises and about 270 for VA (figure 6).[Footnote 23] This difference between FHA and the enterprises in average REO timelines persisted after controlling for the average effects of certain property characteristics such as location, value, and local real estate market conditions.[Footnote 24] Figure 6: Average Disposition Timelines by Entity for REO Dispositions, January 2008 through June 2012 (All Years Combined): [Refer to PDF for image: 4 timelines] FHA: Foreclosure to REO: 340 days; REO Acquisition: 161 days; Initial valuation: 184 days. Enterprise 2: Foreclosure to REO: 208 days; Initial valuation: 66 days. Enterprise 1: Foreclosure to REO: 201 days; Initial valuation: 69 days. VA[A]: Foreclosure to REO: 268 days. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. Note: The enterprises generally acquire REO properties within 24 hours of foreclosure sales, and VA generally acquires properties within 15 days. The REO sale date represents the closing date of the REO sale. Each of the timelines included data from 337,755 (100.00 percent of total) dispositions for one enterprise and 703,853 (100.00 percent) for the other enterprise. For FHA and VA, the calculation of the foreclosure sale to REO sale timelines included data from 344,396 (92.94 percent) and 58,558 (100.00 percent) dispositions, respectively. For FHA, the foreclosure sale to REO acquisition timeline included data for 344,396 (92.94 percent) dispositions and the foreclosure sale to initial valuation timeline included data from 344,395 (92.94 percent) dispositions. The calculations did not include all dispositions for each entity because of missing values for certain data elements. [A] VA did not have data available for its REO properties' initial valuation dates. [End of figure] Longer disposition times can create additional costs for items such as taxes, insurance, home owners' association fees, maintenance costs, and other expenses (holding costs) and leave properties exposed to an increased risk of vandalism and other property damage. We found that if FHA's average number of days from foreclosure sale to REO disposition had equaled those of the enterprises, it could have avoided around $600 million in extra holding costs for 2011 alone. [Footnote 25] Furthermore, the negative impact of vacant properties on neighborhoods and property values has been identified in prior GAO reports and other sources, underscoring the importance of minimizing the amount of time required to dispose of properties after the foreclosure sale.[Footnote 26] The differences in the length of time FHA took to dispose of REO properties relative to the other entities was largely attributable to differences in one specific period--the time from the foreclosure sale until the date of the initial REO valuation. FHA and the other organizations obtain the initial value assessment once a property is ready to be listed for sale and marketed to potential buyers. The assessment is typically made after the expiration of any redemption period after the foreclosure sale and any necessary eviction action, as well as cleaning, maintenance, and repair of the property. FHA's average for this period was 184 days, while the enterprises' averages were 69 (enterprise 1) and 66 days (enterprise 2).[Footnote 27] As discussed previously, FHA acquires its REO properties from mortgage servicers at a later date after the foreclosure sale than the enterprises and other entities. Having multiple entities complete its postforeclosure sale activities, such as eviction, could be one reason that FHA's average time from foreclosure to initial valuation was longer than the times for the other entities.[Footnote 28] As we stated in our 2002 report on FHA's process for selling REO properties, the enterprises, VA, and RHS have one entity that is responsible for the custody, maintenance, and sale of foreclosed properties, but FHA divides these responsibilities among its mortgage servicers and REO contractors, all of which operate largely independently of one another.[Footnote 29] This divided approach to property custody could delay the initiation of critical steps necessary to sell REO properties quickly. As a result, in 2002 we recommended that HUD establish unified property custody as a priority for FHA and that it determine and implement the optimal method for establishing unified property custody. In their response at the time of the report, HUD said that it agreed that a unified custody approach may streamline processes and oversight, reduce holding time, and increase net return. HUD also said that the agency intended to continue research to determine the feasibility of unified custody within the framework of existing statutory requirements and explore statutory changes that would increase efficiencies in the property disposition program. However, HUD subsequently determined that it would not be advisable for the agency to establish unified property custody as an objective for the agency and it did not implement our recommendation. The analysis in this report once again highlights the need for FHA to consider whether the potential benefits from unified property custody, such as shorter REO disposition timelines and lower holding costs, outweigh any costs and challenges associated with acquiring REO properties from servicers closer to the foreclosure sale date. Once the initial value assessment was completed, it took FHA an average of 168 days to sell a property, which was more comparable to the enterprises' average timelines of 142 and 132 days. And in the first half of 2012, FHA had a shorter average time from initial valuation to completed REO sale than the enterprises, as shown in figure 7. Figure 7: Average Disposition Timelines by Entity for REO Dispositions, January 2008 through June 2012 (by Year): [Refer to PDF for image: 3 multiple line graphs] Foreclosure Sale to REO Sale: Year: 2008; FHA: 310; Enterprise 2: 229 days; Enterprise 1: 182 days; VA: 377 days. Year: 2009; FHA: 349 days; Enterprise 2: 203 days; Enterprise 1: 177 days; VA: 291 days. Year: 2010; FHA: 313 days; Enterprise 2: 180 days; Enterprise 1: 178 days; VA: 225 days. Year: 2011; FHA: 357 days; Enterprise 2: 220 days; Enterprise 1: 226 days; VA: 246 days. Year: 2012; FHA: 358 days; Enterprise 2: 225 days; Enterprise 1: 235 days; VA: 252 days. Foreclosure Sale to Initial Valuation[A]: Year: 2008; FHA: 156 days; Enterprise 2: 66 days; Enterprise 1: 35 days. Year: 2009; FHA: 176 days; Enterprise 2: 58 days; Enterprise 1: 51 days. Year: 2010; FHA: 171 days; Enterprise 2: 51 days; Enterprise 1: 63 days. Year: 2011; FHA: 189 days; Enterprise 2: 73 days; Enterprise 1: 85 days. Year: 2012; FHA: 238 days; Enterprise 2: 90 days; Enterprise 1: 93 days. Initial Valuation to REO Sale[A]: Year: 2008; FHA: 188 days; Enterprise 2: 163 days; Enterprise 1: 147 days. Year: 2009; FHA: 195 days; Enterprise 2: 146 days; Enterprise 1: 125 days. Year: 2010; FHA: 148 days; Enterprise 2: 129 days; Enterprise 1: 115 days. Year: 2011; FHA: 174 days; Enterprise 2: 147 days; Enterprise 1: 141 days. Year: 2012; FHA: 126 days; Enterprise 2: 135 days; Enterprise 1: 142 days. Sources: GAO analysis of FHA, Fannie Mae, Freddie Mac, and VA data. Note: The REO sale date represents the closing date of the REO sale. Each of the timelines included data from 337,755 (100.00 percent of total) dispositions for one enterprise and 703,853 (100.00 percent) for the other enterprise. For FHA and VA, the calculation of the foreclosure sale to REO sale timelines included data from 344,396 (92.94 percent) and 58,558 (100.00 percent) dispositions, respectively. For FHA, the foreclosure sale to initial valuation timeline included data from 344,395 (92.94 percent) dispositions and the initial valuation to REO sale timeline included data from 369,576 (99.73 percent) dispositions. The calculations did not include all dispositions for each entity because of missing values for certain data elements. [A] VA did not have data available for its REO properties' initial valuation dates. [End of figure] Some of this difference may be due to the time taken to list a property for sale. From January 2008 through June 2012, FHA took an average of 32 days to list a property for sale after receiving the value assessment, compared to 13 days for enterprise 1 and 24 days for enterprise 2. In the first half of 2012, however, FHA's average had fallen to 18 days. Similarly, FHA took slightly longer to sell its properties once they were listed--an average of 136 days compared to 119 for enterprise 1 and 118 for enterprise 2. But in the first half of 2012, FHA's average time from listing to sale was shorter than the enterprises'. FHA officials indicated that its lagging performance prior to 2012 may have been due, at least in part, to limitations in their ability to adjust the level of homeownership center staff resources in response to increases in inventory that occurred in the years prior to 2012.[Footnote 30] FHA's recent performance relative to the enterprises may also reflect continued progress in implementing its new M&M III contractor program structure that began in the middle of 2010. We tested certain factors that were to some degree under FHA's and the enterprises' control to explore whether these factors helped to explain the differences in REO disposition performances. We included these factors (time from foreclosure sale to REO disposition, ratio of initial list price to initial valuation amount, disposition method, and buyer type) in additional regression models to determine whether their inclusion significantly changed the results. These models indicated that the length of time from the foreclosure sale to the REO sale was associated with FHA's lower net execution rate relative to the enterprises. Specifically, when our regression models included the time from the foreclosure sale to the REO sale, FHA's relative performance deficit in terms of the independently assessed value net execution rate was eliminated completely. Such a result indicates that the longer time FHA requires to acquire and sell properties could be an important factor in explaining differences in execution rate performance. However, an actual causal relationship is difficult to isolate and prove, as additional factors--such as deteriorated property conditions or variations in market conditions within ZIP code areas--that were not incorporated into our regressions could also explain both the performance difference and the difference in the average total disposition time between FHA and the enterprises. FHA and the Enterprises Disposed of Most REO Properties through Retail Sales, Which Produced Higher Returns Than Other Methods: Aligning with their stated preferences for retail sales, the predominant share of FHA's and the enterprises' REO property dispositions were through retail sales of individual properties to owner-occupants or investors (figure 8). From January 2007 through June 2012, retail sales were about 97 percent of FHA's dispositions, about 99 percent of enterprise 1's, and 91 percent of enterprise 2's. Enterprise 2 had the highest share of bulk and auction sales over the time period, representing almost 7 and 2 percent of its total dispositions, respectively. Less than 4 percent of FHA's dispositions from January 2007 through June 2012, on average, were for programs related to its housing mission goals (e.g., discounts and donations). Dispositions through the enterprises' programs that market properties to nonprofits and public entities (e.g., discounts and donations) accounted for less than 0.5 percent of total dispositions from January 2007 through June 2012. Figure 8: Sales of REO Properties by Type of Sales Method, January 2007 through June 2012: [Refer to PDF for image: 3 pie-charts] FHA: Discount donation, other: 3.5%; Retail: 96.5%. Enterprise 2: Discount donation, other: 0.1%; Bulk: 2.0%; Auction: 6.5%; Retail: 91.4%. Enterprise 1: Discount donation, other: 0.3%; Bulk: 0.1%; Auction: 1%; Retail: 98.5%. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. Note: FHA held several auction sales in 2011 and 2012 through which it disposed of a total of 326 properties representing 0.08 percent of its dispositions for the overall period. The calculation of sales method types included data from 417,301 (100.00 percent of total) dispositions for FHA, 352,798 (99.99 percent) for one enterprise, and 741,623 (100.00 percent) for the other enterprise. VA did not have property-level data available on the type of sales methods it used. [End of figure] Retail sales also generated higher returns than other disposition methods. For example, FHA's independently assessed value and initial list price execution rates (net sales proceeds as a share of property values or list prices) were both 30 percent higher for its retail sales than for its mission program dispositions for sales from January 2007 through June 2012. The enterprises, particularly enterprise 2, also generally had higher execution rates for retail sales than other disposition methods over the same period. Specifically, enterprise 2's independently assessed value execution rate for retail sales was 52 percent higher than for its auction sales, 95 percent higher than for its bulk sales, and 62 percent higher than for its nonprofit and public entity sales. For enterprise 1, retail sales were 36 percent, 40 percent, and 1 percent higher, respectively, than sales using these methods. FHA Sales to Owner-Occupants Net Higher Returns Than Those to Investors, but FHA Had a Smaller Share of These Dispositions Than the Enterprises: We also examined FHA's performance in selling REO properties to different types of buyers. Our analysis of data for all dispositions from January 2007 through June 2012 showed that FHA achieved higher returns on sales to owner-occupants than on sales to investors and other buyers, although it had a smaller share of these types of dispositions than the enterprises. FHA's independently assessed value and initial list price net execution rates both were 25 percent higher for sales to owner-occupants than for sales to nonowner-occupant investors. The enterprises experienced smaller return premiums-- ranging from 10 to 19 percent--from owner-occupant sales as measured by these sales return measures. FHA's and the enterprises' overall returns on owner-occupant sales were generally comparable. Specifically, FHA's net execution rate based on independently assessed value was the same as enterprise 1's and four percentage points less than enterprise 2's. Based on initial list price, FHA's net execution rate was 2 percentage points less than enterprise 1's and the same as enterprise 2's. Yet FHA sold a smaller share of its properties to owner-occupant buyers than the enterprises. Specifically, about 58 percent of FHA's sales from January 2007 through June 2012 were to owner-occupant buyers, compared to 63 and 68 percent for the enterprises. FHA's share of owner-occupant sales increased to about 64 percent in the first half of 2012. Figure 9 shows FHA's and the enterprises' percentage of REO property sales to owner-occupants, based on our analysis of data from sales of REO properties from January 2007 through June 2012. Figure 9: Aggregate Share of REO Dispositions by Type of Buyer for All Sales, January 2007 through June 2012: [Refer to PDF for image: 3 pie-charts] FHA: Other: 42%; Owner occupant: 58%. Enterprise 2: Other: 37%; Owner occupant: 63%. Enterprise 1: Other: 32%; Owner occupant: 68%. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. Note: The calculation of buyer types included data from 417,280 (99.99 percent of total) dispositions for FHA, 352,807 (99.99 percent) for one enterprise, and 732,659 (98.79 percent) for the other enterprise. VA did not have property-level data available on the type of buyer. [End of figure] Properties that FHA sold to owner-occupants also had higher average sales prices and were sold more quickly than properties that were sold to other buyers. For example, the average price of FHA's sales to owner-occupants from January 2007 through June 2012 was more than $77,000, compared to less than $50,000 for sales to nonowner- occupants. Likewise, it took FHA an average of 42 fewer days to dispose of properties sold to owner-occupants during the same time period than were needed to complete sales to investors and other buyers, as measured from the foreclosure sale date. The average sale prices of the enterprises' owner-occupant sales were around twice the average amount of their sales to investors and other buyers, but the timelines were similar for both types of sales. Results from our regression models also indicated that the type of buyer was associated with FHA's lower independently assessed value net execution rate--which gauges success in maximizing a property's sale price. When we controlled for differences in the share of sales to owner-occupants in our regression models, FHA's performance deficit relative to the enterprises for independently assessed value net execution rate was reduced. However, this association does not necessarily mean that this factor caused the performance difference. For example, additional factors--such as deteriorated property conditions or the existence of certain amenities that might attract owner-occupants--that were not controlled for in our regressions could also explain the differences in performance and share of owner- occupant sales between FHA and the enterprises. REO Properties That Had Been Repaired or Were in Good Condition Netted Higher Returns, but FHA Repaired Few Properties before Selling Them: Based on our analysis of all dispositions from January 2007 through June 2012, the enterprises repaired more properties than FHA and experienced higher returns than on properties that they did not repair.[Footnote 31] Our review of data for all REO dispositions during this period showed that the enterprises spent at least $1,000 on repairs for 29 percent (enterprise 1) and 23 percent (enterprise 2) of the properties they sold. FHA, however, spent at least that amount on only about 5 percent of its properties. Based on our analysis, we found that properties repaired by the enterprises netted higher independently assessed value net execution rates, after accounting for repair costs, and also achieved higher list price net execution rates. [Footnote 32] Specifically, the enterprises' net execution rates based on independently assessed value--including the cost of repairs--were 3 to 4 percentage points higher over the entire period for properties with at least $1,000 in repair costs than for properties with repair costs less than that amount.[Footnote 33] The difference for the net execution rate based on list price over the entire period was 3 percentage points for each of the enterprises. However, the enterprises' properties with at least $1,000 in repair costs sold an average of 33 to 47 days more slowly than properties with lower repair costs, as measured from the initial valuation date. These differences may reflect the time required to complete repairs or a greater willingness to market the property for a longer period. Our analysis also showed that FHA netted higher returns on sales of REO properties that were in better condition--that is, that met minimum property standards to qualify for FHA insurance. To be eligible for an FHA-insured loan, a property must be in a condition and location free of known hazards and adverse conditions that could affect occupants' health and safety or the structural soundness of any improvements or that could impair the use and enjoyment of the property. For example, FHA requires properties to have adequate heating, hot water, and electricity. Based on our analysis of data for all REO dispositions from January 2007 through June 2012, FHA received higher sales returns for properties that were eligible for FHA insurance (eligible) than it did for properties that were deemed ineligible because their condition did not meet these standards (ineligible). FHA's independently assessed value net execution rate was 12 percentage points higher for eligible properties than for ineligible properties for all dispositions during this period. Furthermore, eligible properties sold faster than ineligible properties, which took an average of 88 additional days--26 percent longer--from the foreclosure sale date to sell. FHA staff told us that while they generally did not repair REO properties to increase the sale value, some properties are repaired to address health and safety concerns and to preserve the property's condition. FHA officials also noted that while the agency might conduct these types of repairs when necessary, FHA does not repair properties specifically to meet its minimum property standards. [Footnote 34] FHA officials explained that FHA had a long-standing policy of not repairing properties. They said that the agency does not conduct repairs because of concerns about having to oversee contractors that perform the work and HUD's inability to obtain volume discounts on replacement appliances or other home fixtures because of the agency's preference for using small contractors. They also said that having to comply with HUD procurement guidelines and the Davis- Bacon Act made it more difficult for FHA to engage in construction projects to repair properties and increase sale returns.[Footnote 35] However, officials noted that in 2011 FHA had begun a small pilot program in its Atlanta homeownership center to assess the impact of repairs on properties' marketability. This program selected favorable properties--that is, relatively high-quality properties in a few counties in the Atlanta metropolitan area--to repair and officials indicated that results had generally been positive. The pilot involved about 50 completed sales and around 80 additional properties as of September 2012, according to FHA officials. However, the officials also said that FHA had not analyzed the sale prices of the repaired properties to determine whether it was achieving higher returns than it could achieve without conducting repairs. Additionally, one FHA official expressed concern that the existing policy not to repair properties prevented FHA from capturing the additional returns that can come from selling repaired properties for higher prices. Instead, the official said that selling properties without making repairs intended to increase the sale value allowed investors to purchase them, make the repairs, and capture the additional returns. Similarly, VA and RHS staff said that their agencies generally did very few repairs to REO properties and that most generally were sold without repairs intended to increase the sale value, largely because repairs did not generally result in higher returns. In some cases, VA has conducted minor (cosmetic) repairs in order to improve returns, according to VA staff. However, staff noted that in general the costs associated with making these repairs have not been fully recovered by the eventual sale proceeds. They further noted that some cosmetic repairs--such as fixing windows, painting, or installing new carpeting- -may increase sale returns, but major repairs often reduced returns, at least in part because of the additional costs of repairs and holding the property longer. The VA officials told us that a few years ago they conducted a small case study of repair work done for six properties and found no positive result from doing the repairs. RHS officials also said that RHS did not make repairs for the majority of its direct loan REO properties unless repairs were needed for safety. For properties in its guaranteed loan program, RHS generally has not had lenders complete cosmetic repairs but may consider repairs to increase returns on a case-by-case basis. RHS officials also said that lenders completed repairs on guaranteed properties for safety reasons and to preserve and protect the property. In contrast to FHA, VA, and RHS direct loan properties, the enterprises and the three private mortgage servicers we contacted did make case-by-case determinations on conducting cosmetic repairs to improve returns, increase the likelihood of an owner-occupant purchase, or meet neighborhood standards. Officials from these entities said that they did repairs on between about 20 to 40 percent of their REO properties, although staff from one of the private mortgage servicers indicated that they had been repairing up to 80 percent of their properties due to the lengthy foreclosure process. One of the enterprise's officials told us that it had been repairing more REO properties as a way to improve the impact on neighborhoods as well as to earn the highest possible return. For example, the officials said that about 80 percent of repaired properties were sold to owner-occupants compared to about 50 percent of unrepaired properties. They also explained that repair decisions were based on numerous factors, including neighborhood conditions, potential buyers, and the costs of the repairs. As the length of time that an unsold property remains on the market increases, the enterprise may reassess the repair decision to see if performing repairs could add value and facilitate a sale. Officials also said that the enterprise repaired properties based on expected returns, regardless of value. The other enterprise also makes decisions on whether and to what extent to do repairs on a property-by-property basis, primarily to increase returns, according to staff. This enterprise's staff said that they viewed repairing properties as a way to maximize the properties' value and increase the chances of selling them to owner-occupants. They said that the enterprise also tried to ensure that properties conformed to neighborhood standards and were competitive with other properties for sale in the area. In some areas where the potential for vandalism was high, they said that the enterprise would be less likely to make repairs early in the REO process but would complete them just prior to closing. Among the goals that FHA staff described for the agency's REO disposition program were maximizing net returns to the mortgage insurance fund and increasing home ownership, but FHA may be failing to take advantage of the opportunity for increased financial returns by not repairing more properties. FHA's policy to limit repairs only to those related to health and safety concerns may in part explain why it sells fewer properties to owner-occupants than is the case for the enterprises. Repairing properties only to address health and safety concerns would not necessarily result in a property that meets standards for FHA eligibility, and as a result FHA may be selling fewer properties to owner-occupants, many of whom may be interested in FHA loans. As we have shown, FHA's sales of eligible properties yield higher returns than those that are not in an eligible condition. If FHA repaired ineligible REO properties to make them eligible, the agency might be able to realize higher sales returns, avoid the holding costs related to the longer disposition time frames for ineligible properties, and further its mission of increasing home ownership.[Footnote 36] FHA Uses a Single Assessment of Market Value in Setting List Prices, but Other Entities Use Multiple Sources: FHA uses only one input to set list prices--an appraisal, or professional appraiser's estimate of the property's fair market value based on market research and analysis as of a specific date. Other entities use additional, generally accepted methods for establishing a listing price for their properties, including obtaining an estimate from a real estate broker--known as a broker's price opinion (BPO). BPOs are estimates of the market value of a particular property prepared by a real estate broker, agent, or sales person. In addition, market values for properties can also be estimated using automated valuation models (AVMs), computerized programs that estimate property values using proprietary and public data, such as tax records and information kept by county recorders and multiple listing services, and other real estate records. FHA's marketing contractors set REO property list prices at the appraisal value, although their marketing contractors also have access to BPOs. FHA's regulations require the use of an independent appraiser when setting a price for an REO property, and FHA staff told us that properties typically are listed at the appraised value.[Footnote 37] Based on our review of property dispositions from January 2007 through June 2012, the list prices of more than 98 percent of FHA's properties equaled the appraised market value. However, FHA field staff told us that the agency's marketing contractors often also ordered BPOs to evaluate and review list prices and were required to obtain BPOs when requested by FHA. These additional valuations are not used to change the list price; however, FHA staff said that the listing brokers used BPOs to evaluate and support list price reductions when properties did not sell. Staff from one of FHA's four homeownership centers noted that its marketing contractors often have properties' listing brokers complete a BPO during the listing process to assess the accuracy of the appraisal value used in setting the list price. These staff said that if the BPO value differed greatly from the appraisal value, the marketing contractor might discuss the valuation with the appraiser and request that the value be reconsidered. Although they said that they could only recall one or two instances of appraisers changing their valuations after these discussions, the staff at this office considered comparing the appraisal value to a BPO value to be an effective practice on a case-by-case basis. A review of the plans of selected marketing contractors showed that contractors were to obtain BPOs during the listing process and also when considering subsequent list price reductions. For example, some contractors' marketing plans called for an initial BPO when the property was listed for sale and every 30 days thereafter as a way to evaluate the appraised value and appraiser performance and to analyze market data. The contracts between FHA and its marketing contractors also state that the contractor must obtain an independent BPO when directed by FHA staff. In contrast, the enterprises, VA, RHS, and the three private mortgage servicers we interviewed all use at least two methods--either an appraisal and BPO or two BPOs--to estimate the market value of their REO properties as part of determining a list price. The enterprises also use an AVM to provide an additional value estimate and incorporate additional information and analysis beyond the supplemental valuation information into their list price decisions. For example, the enterprises produce list price guidance based on factors such as location, market conditions, comparable sales, REO sales trends, and input from listing agents. Following this guidance, the enterprises may set a list price above or below the estimated market value based on whether the property is located in a depreciating or appreciating market. For property dispositions from January 2007 through June 2012, one enterprise set the initial list price for less than 1 percent of its properties at the independent BPO value, and 28 percent were within 5 percent of the BPO value.[Footnote 38] The other enterprise set initial list prices at the appraised market values for fewer than 10 percent of the REO properties that it disposed of during this period, and 28 percent of initial list prices were within 5 percent of the appraised value. Using multiple information sources could improve the accuracy of FHA's market value estimates and list prices. Officials from other federal housing agencies and the enterprises said that multiple inputs increased accuracy by providing a range of independently valued assessments. Providers of appraisals and BPOs use different approaches to valuing properties and combining the two methods produces the best results, according to officials from some entities. For example, appraisals are to be conducted by trained and certified independent professionals with no interest in the outcome of the sales, but these appraisals focus on past sales and listings and may not reflect current price trends. BPOs, although conducted by brokers who may have an interest in the outcome, may reflect more knowledge of the properties and local markets. In addition, appraisers may have difficulty finding comparable property sales in some rural areas, and officials said that appraisals are more costly to obtain than BPOs. If estimates from different sources vary, entities reconcile them to produce a market value estimate that reflects a broader and more diverse base of information and analysis than an estimate from a single source. Our analysis of the enterprises' reconciled value estimates--which incorporate all of their market value inputs such as appraisals, BPOs, and AVMs--indicated that the reconciled values generally were lower than independent value assessments reflecting a single source such as an appraisal or BPO and accordingly reflected final sale prices somewhat more accurately. For all dispositions from January 2007 through June 2012, the enterprises' reconciled value estimates were closer to gross sale prices than their independent value assessments were by 1 and 7 percentage points overall. The use of multiple valuation methods could help FHA more accurately estimate the market values of its REO properties, increasing the likelihood of selling properties more quickly and at prices that best reflect current market conditions. FHA officials indicated that BPOs and AVMs could reduce costs and increase the accuracy of FHA's market value assessments by better reflecting recent market trends. A senior official from FHA's single family housing program also said that using AVMs could improve FHA's ability to identify the most appropriate marketing and disposition strategies for certain properties by providing more accurate and timely market value estimates. In early 2013, FHA's Santa Ana homeownership center began a pilot program to evaluate the use of AVMs in validating appraised market values, according to agency officials. The officials explained that the pilot program uses a model that incorporates results from multiple AVMs to assess the independent appraisals. They said that the model has helped to identify opportunities for FHA to increase list prices based on market analysis. The officials also said that they were considering other options to establish accurate list prices and reduce risks from appraised market values that were unnecessarily low relative to market conditions. They stated that FHA was working with HUD's Office of General Counsel (OGC) to determine if the regulatory requirement that list prices be based on an appraisal allowed them to be based on multiple sources that include an appraisal rather than solely on the appraised market value. FHA's Use of Nonmarket-Based Price Reductions Differs from Other Entities' Practices and May Contribute to Lower Returns: FHA generally did not take into account market conditions when reducing the list prices for REO properties that do not sell. FHA's marketing contractors determine when and by how much the list prices should be reduced. FHA's marketing contractors create plans for each of the geographic contract areas in which they operate that describe how they intend to market and sell FHA properties and submit these plans to FHA's homeownership centers for approval. The marketing plans include a schedule identifying time frames and percentage thresholds for reducing list prices. For example, a schedule might indicate that for properties that have been listed for sale for between 30 to 60 days, the list price should be set at 90 percent of appraised value. These schedules vary by marketing contractor and FHA homeownership center. For example, the schedules' quantitative thresholds for the amount and timing of price reductions can differ, although most plans use one of two standard amounts. Some schedules describe a reduction of "up to" a certain percentage of the appraised value or current list price, while others specify that properties will be listed for an amount "no less than" a percentage of the appraised value or current list price. Our analysis of each of the 23 marketing plans used by FHA's marketing contractors showed that all but one used a schedule for price reductions. However, FHA lacks a clear and consistent policy for how price reductions should be conducted, allowing each of its homeownership centers to approve marketing contractors' plans for reducing list prices. FHA headquarters and homeownership center officials whom we interviewed differed on the extent to which price reductions required review and approval by FHA staff. Officials from FHA's headquarters said that marketing contractors had to provide supporting documentation to justify why a price reduction was necessary and receive prior approval from homeownership center staff for all proposed reductions, even if they were following a schedule. However, officials we spoke with at each of the centers said that their marketing contractors typically did not need and did not obtain prior review and approval from FHA staff as long as they followed the approved price reduction schedules in their marketing plans. They added that marketing contractors only needed to provide documentation to FHA homeownership center staff and receive homeownership center approval for price reductions that exceed marketing plan thresholds. Further, staff from homeownership centers described to us different practices that they followed when considering price reductions in excess of marketing plan schedules. For example, the Atlanta and Denver offices allowed exceptions to the schedule for individual properties, but the Santa Ana office does not allow any price reductions in excess of the schedule, according to staff. Analysis of FHA's REO property dispositions indicated that marketing contractors generally followed the price reduction schedules systematically when reducing properties' list prices. To assess the extent to which list price reductions on FHA's properties followed these schedules, we analyzed list price data from FHA's REO property dispositions from June 2011 through June 2012 and compared the price changes to the schedules in contractors' marketing plans. Based on this analysis, we found that almost half of the properties that FHA sold over the period had at least one price reduction. Of these, about 75 percent of the initial reductions were for a scheduled amount.[Footnote 39] Most price reduction schedules based this amount on one of two specific percentages. Only one of FHA's four homeownership centers--Denver--preferred its marketing contractors to base list price reductions on evaluations of market conditions rather than on a schedule. Denver center officials said that they encouraged marketing contractors to base price reduction decisions on evaluations of individual property-level market data. The Denver center also requires its marketing contractors to provide supporting documentation for all price reductions on REO properties, and its staff review the documentation for a small sample of these reductions, according to officials. Furthermore, one contractor's price reduction schedule for the Denver center included a threshold range for price reductions rather than a specific percentage. One marketing contractor in the region did not have a price reduction schedule in its marketing plan and instead analyzed market conditions, the appraisal, and other information to determine a new list price. None of the marketing contractors in the other centers used a range for price reductions or had marketing plans that did not include a price reduction schedule. Officials from one homeownership center indicated that automated reductions were easier, quicker, and required fewer resources. Those from another center noted that following such reductions helped to find the right price in a structured fashion. In contrast, officials from the Denver center identified several disadvantages to using systematic price reductions and said that they attempted to have the schedules removed from contractors' marketing plans at the beginning of the M&M III program. However, they explained that officials at FHA headquarters at that time strongly resisted the change. Although differences in disposition performance cannot be attributed solely to pricing practices, the Denver homeownership center performed better both in terms of sales returns and speed of sales than either the Philadelphia or Atlanta centers even when we controlled for regional differences, as we discuss later in the report. And compared to the enterprises' performance in the states where each center operates, the Denver center's performance compared more favorably with the enterprises' performance than did that of the other centers, which lagged the enterprises. While homeownership centers generally reduced REO properties' list prices with similar frequency, some differences existed in the degree to which contractors' price reductions followed marketing plan schedules. Based on our analysis of data on all property dispositions from June 2011 through June 2012, each of the homeownership centers reduced prices for close to the overall average of 47 percent. However, the Denver center reduced prices by scheduled amounts less frequently than the other centers, consistent with its preference that contractors base price reduction decisions on evaluations of market conditions. For example, almost 90 percent of the Santa Ana center's initial price reductions were for a scheduled amount, compared with 58 percent for the Denver center. The Atlanta and Philadelphia centers had figures of 81 percent and 76 percent, respectively. Other federally related housing entities with REO inventories generally based their decisions to reduce prices on evaluations of property-level information and market conditions. According to officials from the enterprises and VA documents that we reviewed, these entities used individual assessments of market conditions rather than predetermined schedules when considering the timing and amount of list price revisions. These entities also defined thresholds to identify when reductions required additional levels of approval. RHS follows a schedule for its direct loan program, but agency officials acknowledged limitations with this approach including a lack of flexibility. Market participants that we interviewed, including real estate brokers, an industry consultant, and a nonprofit organization, identified disadvantages with systematic price reductions, such as the potential for interested buyers to adjust the timing and amount of their bids in anticipation of a discounted price. The limited use of price reduction schedules may provide certain advantages by establishing clear benchmarks for determining when to evaluate a property's market situation and which reduction amounts should require review and approval by FHA staff. However, reducing list prices based solely on a schedule may lower prices at times and by amounts that are not optimal, potentially lowering FHA's net return. For example, a property whose list price is reduced excessively or hastily may sell at a price that is unnecessarily low based on market conditions, leading to lower returns for FHA. Also, mispriced properties may take longer to sell thereby increasing FHA's holding costs. In contrast, a strategy of basing price reductions more comprehensively on evaluations of property-level information and market conditions would likely be more flexible and could provide more accurate prices. Further, because many marketing contractors already obtain BPOs and other assessments of market conditions and are sometimes required to do so, such an approach would likely not involve significant costs for FHA. FHA Could Further Improve Its Monitoring and Evaluation of REO Contractors: FHA does not have a current and complete set of policies and procedures for its current REO disposition program as required by internal control standards.[Footnote 40] Federal internal control standards require agency management to conduct monitoring of program quality and performance through the establishment and review of performance measures and indicators. Under the new contract structure that FHA introduced in 2010, the agency intended that its staff conduct specific activities to assess whether its contractors were meeting minimum requirements under the contracts, but these reviews have not been occurring. Further, while FHA is in the process of implementing procedures to better ensure that FHA homeownership centers perform consistent oversight activities, it has not implemented a critical component of the plan--a scorecard to evaluate contractor performance against standard metrics that would allow it to compare the quality of its contractors' activities. Our review also showed that FHA was not conducting as many or as frequent in-person property inspections as other entities that dispose of REO properties and that it was not taking steps to determine that listing brokers were located sufficiently close enough to the properties they were selling to ensure local market knowledge. Finally, although assigning work to contractors in part on the basis of their performance was intended to have been a key quality assurance mechanism under the new contract structure, FHA has encountered various obstacles to implementing this condition. FHA's Lack of Current Policies and Procedures for REO Disposition May Create Inefficiencies: FHA does not have a current and complete set of policies and procedures for its current REO disposition program, as required by internal control standards. These standards require formally documented policies and procedures that are clear and readily available.[Footnote 41] Such materials can be used to: * provide guidance to staff in the performance of their day-to-day activities, * help ensure that activities are performed consistently across an agency, * communicate management's directives, and: * help ensure that the agency complies with federal laws and regulations. These control standards require that policies and procedures be reviewed regularly and updated when necessary. In keeping with these requirements, the enterprises have well-documented policies and procedures for their REO disposition programs. One of the enterprises has consolidated its guidance and expectations for its staff and contractors in a single, comprehensive guide that it updates as needed. The other enterprise does so using multiple documents that are accessible from a single location. However, while FHA has an REO Disposition Handbook that outlines policies, procedures, and key controls for REO activities, the agency has not updated the handbook since 1994. FHA describes the Disposition Handbook's objectives as providing comprehensive guidance that reflects program requirements and that stresses the importance of internal controls, incorporates fiscal procedures, and provides clear statements of policy for field office staff. However, the handbook does not reflect the current structure, processes, or requirements that FHA uses to dispose of properties. Nevertheless, FHA's contracts reference it as a source of applicable guidance. FHA headquarters staff told us that the disposition handbook was outdated and did not reflect the current REO program structure, including the use of the multiple types of contractors and their responsibilities. But staff from one FHA homeownership center told us that they continued to use the handbook for policy guidance in certain areas, such as broker registration, contract extensions, rental agreements, and the closing agent monitoring checklist. Instead of updating its Disposition Handbook, FHA relies on mortgagee letters, housing notices, and contracts to document its current policies and procedures. FHA headquarters staff told us that they had not taken steps to update the program handbook and that they used these letters and notices to provide new and revised guidance to their staff and contractors. FHA officials also indicated that the terms of its contracts with service providers served to document the REO process and performance expectations. However, as mentioned earlier, these contracts refer to the outdated Disposition Handbook as a source of reliable guidance. Also, using multiple contracts rather than a single consolidated document as a source of policies and expectations for staff and contractors creates the potential for inconsistencies. In addition, FHA has a decentralized REO disposition process managed through four homeownership centers, underscoring the need for a single source of guidance on policies and procedures for headquarters and field staff. FHA's contractors may have multiple contracts overseen by different centers, and in a February 2013 report, HUD's Office of the Inspector General (OIG) found that one of FHA's REO contractors faced different procedural requirements across the various HUD operating regions in which it was active.[Footnote 42] According to HUD OIG staff, they had also found a lack of consistent expectations for contractors across homeownership centers. For example, some centers were requiring maintenance contractors to pay any homeownership association fees and other unpaid bills before sale closings, but other centers expected marketing contractors to make these payments. In some cases, FHA faced delayed sales closings for failing to pay these fees on a timely basis. Lack of consistent and updated guidance on policy and procedures may also make oversight of contractors less efficient and may increase the REO program's operating costs. FHA staff in two homeownership centers told us that they spent significant amounts of time responding to policy inquiries from contractors and seeking answers to questions about policy within the agency. Staff from one of these homeownership centers indicated that more guidance on policy matters would be beneficial and said that they needed additional support for procedures. The lack of a single, up-to-date form of guidance for the REO program leaves FHA without important internal controls, has resulted in extra costs in time and resources for FHA staff, and has created a burden for some contractors who face differing requirements across regions. Further, the lack of consistent guidance may be a factor in the execution rate performance of the four homeownership centers. For instance, some centers may be using practices that increase contractor performance that other centers have not tried. As discussed previously, our analysis of FHA's property dispositions revealed differences in performance levels across its homeownership centers. For instance, aggregate sales returns based on independently assessed value and list price execution rates for all properties disposed from January 2007 through June 2012 were 13 and 12 percentage points higher for the Santa Ana and Denver centers than for the Philadelphia center and 7 percentage points higher than for the Atlanta center. Each year within the overall period showed similar patterns with the Santa Ana and Denver centers typically having higher execution rates than the Atlanta and Philadelphia centers. Even after controlling for the average effects of certain property characteristics, such as value and changes in local housing prices, Santa Ana and Denver had higher execution rates than Atlanta and Philadelphia. Because regional housing market differences could affect performance of the homeownership centers even after controlling for certain property characteristics, we compared the performance of the homeownership centers with the performance of the enterprises in the states where each center operates. For all dispositions from January 2007 through June 2012, the Denver center's aggregate independently assessed value net execution rate was comparable to that of the enterprises in the states where it was responsible for REO property dispositions. The Santa Ana and Atlanta centers lagged the enterprises by 2 to 6 percentage points for dispositions in their respective states, while the Philadelphia center lagged the enterprises by 8 and 11 percentage points (figure 10). After controlling for the average effects of certain property characteristics, such as value, ZIP code, and changes in local housing prices, differences in the homeownership centers' performance relative to the enterprises generally persisted. However, Santa Ana's performance decreased while Philadelphia's increased such that their performance differences relative to the enterprises were similar. Denver's performance declined slightly but remained comparable to that of the enterprises and the Atlanta center's performance improved but still lagged that of the enterprises. Figure 10: Differences in FHA Homeownership Centers' and the Enterprises' Aggregate Net Execution Rates Based on Independently Assessed Value for All REO Dispositions, January 2007 through June 2012: [Refer to PDF for image: vertical bar graph] Percentage points difference: Homeownership center: Atlanta; Performance relative to Enterprise 2: -5.22%; Performance relative to Enterprise 1: -2.04%. Homeownership center: Denver; Performance relative to Enterprise 2: -0.73%; Performance relative to Enterprise 1: 2.09%. Homeownership center: Philadelphia; Performance relative to Enterprise 2: -11.05%; Performance relative to Enterprise 1: -8.02%. Homeownership center: Santa Ana; Performance relative to Enterprise 2: -5.82%; Performance relative to Enterprise 1: -3.65%. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. Note: This calculation included data from 416,929 (99.91 percent of total) dispositions for FHA, 352,490 (99.90 percent) for one enterprise, and 648,966 (87.51 percent) for the other enterprise. It did not include all dispositions for each entity because of missing values for certain data elements. VA did not have property-level data available that were necessary to calculate net execution rates. [End of figure] The average number of days to complete an REO disposition from the date that FHA acquired properties also varied, from 164 for the Denver center to 212 for the Philadelphia center for all dispositions from January 2008 through June 2012.[Footnote 43] Other time frames--such as from the initial list date to the completed sale--also illustrated performance differences among the homeownership centers, as did the results for individual years. Even after controlling for the average effects of certain property characteristics, such as value and changes in local real estate prices, these differences persisted. Because regional housing market differences also could affect the time required to dispose of properties, we compared the homeownership centers' time frame performance with that of the enterprises in the states where the centers operate. For all dispositions from January 2008 through June 2012, the Denver center's average number of days from a property's initial valuation to the completed REO sale was generally equivalent to that of the enterprises.[Footnote 44] The Santa Ana, Atlanta, and Philadelphia centers, in order, took longer to sell properties than did the enterprises in the states in which they operate (figure 11). After we controlled for the average effects of certain property characteristics, such as value, ZIP code, and changes in local housing prices, the average number of days from initial valuation to completed sale for each of the homeownership centers generally exceeded that of the enterprises to an even greater extent. The Denver center exceeded the enterprises by the least number of days, while Philadelphia exceeded the enterprises by the most. Figure 11: Difference between FHA Homeownership Centers' Average Disposition Times and the Enterprises' Times for All REO Dispositions, January 2008 through June 2012: [Refer to PDF for image: vertical bar graph] Difference in average days: Homeownership center: Atlanta; Performance relative to Enterprise 2: 29 days; Performance relative to Enterprise 1: 47 days. Homeownership center: Denver; Performance relative to Enterprise 2: 3 days; Performance relative to Enterprise 1: 0 days. Homeownership center: Philadelphia; Performance relative to Enterprise 2: 48 days; Performance relative to Enterprise 1: 56 days. Homeownership center: Santa Ana; Performance relative to Enterprise 2: 21 days; Performance relative to Enterprise 1: 32 days. Sources: GAO analysis of FHA, Fannie Mae, and Freddie Mac data. Note: The average disposition times reflect the time from initial valuation, since that date was available from FHA and the enterprises and most closely reflected when FHA's homeownership centers began managing and overseeing the disposition of REO properties. This calculation included data from 369,576 (99.73 percent of total) dispositions for FHA, 337,755 (100.00 percent) for one enterprise, and 703,853 (100.00 percent) for the other enterprise. It did not include all dispositions for each entity because of missing values for certain data elements. VA did not have the property-level data available that were necessary to calculate the time from initial valuation to disposition. [End of figure] Returns on sales of FHA's REO properties also varied across marketing contractors, as did the time that the contractors required to complete dispositions. From 2010--when FHA implemented its new contract structure--to 2012, the difference between the contractors with the best and worst execution rates based on independently assessed value was between 12 percent and 19 percent each year.[Footnote 45] Of the seven marketing contractors, one had the best execution rate in each of the last 2 years and another had the worst rate for each of those years. Likewise, the difference between the contractors with the best and worst time frame for the number of days from acquisition to completed sale ranged from 11 percent to 40 percent each year. [Footnote 46] Again, one contractor--not the same one that had the best execution rate--had the best time frame in each of the last 2 years and another--the same one that had the worst execution rate in each of the last two years--had the worst time frame for each of the last 2 years. These differences raise questions about the guidance that FHA provides to its homeownership centers and contractors tasked with managing and selling REO properties. For instance, FHA has not identified optimal practices and included them in a single consolidated handbook, although some centers could be using practices that could benefit others. According to an FHA headquarters official, staff in each homeownership center analyze their office's performance against the overall agency goals of reducing the time that REO properties are in inventory and the time it takes to list properties for sale, steps that should decrease the costs associated with dispositions. According to this official, each center analyzes this data monthly and takes corrective action as deemed necessary, and recently the performance across homeownership centers on these timeline measures have been similar. The official also noted that homeownership center staff complete standard contractor monitoring activities monthly to identify and address potential problems with disposition performance. Although having each homeownership center evaluate its own performance is an important internal control step, it does not replace an independent performance assessment across all of FHA's centers, nor does it address the causes of any differences in performance across centers. FHA has yet to conduct any analysis to identify differences in execution rate performance across homeownership centers and the factors that may account for such differences, although doing so could help to improve performance at all centers and reduce costs across the REO program. FHA Has Yet to Implement a Uniform System for Evaluating Contractor Performance: Under the new contract structure--known as M&M III--created in 2010, FHA expected that its staff and an oversight contractor would conduct a number of specific activities to monitor its REO maintenance and marketing contractors' performance, including (1) assessing whether its contractors are meeting minimum contractual requirements, and (2) using standard metrics in a scorecard to evaluate the level of contractors' performance. Federal internal control standards require agency management to monitor program quality and performance through the establishment and review of performance measures and indicators. Also, HUD contracting standards and guidelines require the periodic evaluation of a contractor's performance to help ensure that services conform to the contract's quality and quantity requirements.[Footnote 47] Under the new M&M III program structure FHA homeownership center staff were expected to evaluate maintenance and marketing contractors monthly to determine whether they were meeting minimum contract standards. Staff were expected to perform this analysis using a tool-- the performance requirements summary--that assessed the contractors against several minimum standards. However, staff at FHA's four homeownership centers have not been performing the systematic reviews envisioned in HUD guidelines and the M&M III program structure to determine whether contractors are meeting minimum performance requirements. FHA staff in the four centers told us that they had not been using the planned assessment tool as intended and instead had just been reviewing the quality of contractors' performance more informally and subjectively. FHA homeownership center staff explained that they did not complete formal performance requirements summary reports to be shared with contractors because FHA did not have a standard reporting mechanism. Instead, they informally assess contractors by examining performance trends, reports on properties exceeding suggested time frames for disposition, property inspections, and public feedback. FHA homeownership center staff indicated that the tool was not available because the methodology for producing it was to have been developed by the oversight monitor contractor, but the initial firm chosen for this role did not produce results that FHA deemed usable and FHA did not renew its contract when it expired in 2011. FHA headquarters officials also said that the original contractor oversight plans were not implemented because this contract was not renewed. By not completing these assessments, FHA has not systematically or uniformly determined whether contractors have been performing as intended. Without a comprehensive system to evaluate whether contractors are meeting minimum performance standards, FHA risks not being able to ensure the most efficient and effective disposition of its properties. Additionally, FHA has failed to implement another critical component of its M&M III program structure--a uniform tool known as a performance scorecard--that was to have been used to compare the level of contractor performance with that of other contractors. FHA intended to use contractor performance scorecards to determine which contractors would continue to receive new assignments of REO properties, and how many they would receive. However, FHA officials said that they never implemented use of a contractor performance scorecard because of the terminated relationship with its oversight monitor contractor that was responsible both for developing the scorecard and for the actual monitoring. While the performance requirements summary was a tool to identify minimum contractor performance, the scorecard would allow FHA to evaluate the level of contractor performance using standard metrics and to better compare the relative quality of a contractor's activities against that of other contractors. For example, other entities use a scorecard to rank contractors on their overall performance as well as on certain component metrics that together comprise their overall score. Component metrics used by these entities include measures such as the average time to complete certain tasks or services or the results of oversight inspections. More recently, FHA has taken some steps to increase the consistency of its monitoring activities. A 2012 report by HUD's OIG found that staff in FHA's four homeownership centers had developed their own contractor oversight procedures that had led to inconsistent oversight of REO contractors.[Footnote 48] During the course of the OIG audit, FHA headquarters staff developed standardized plans--one for monitoring maintenance contractors and one for monitoring marketing contractors-- that the homeownership centers were to begin using in June 2012. Each of these two standardized monitoring plans includes various contract monitoring tasks, such as measuring performance in key areas and reviewing disposition status reports for properties that exceed certain REO processing time frames (e.g., that have been in the REO inventory for more than a year).[Footnote 49] As part of these plans, FHA included a contractor performance scorecard that its staff had developed. However, FHA officials said that the scorecard was not implemented due in part to difficulties renewing FHA's contract with the provider of its REO data management system. To assist in its development of a standard method of evaluating contractor performance, including scorecards, FHA officials said that a new contractor was hired in September 2012. FHA officials said that the scorecards would likely require the approval of each of the maintenance and marketing contractors, as it was not included in their original contracts as a basis for performance evaluation. In contrast, the enterprises and private sector mortgage servicers that we interviewed had been using scorecards to evaluate and compare contractor performance and as a basis for assigning work to contractors. These scorecards generally tracked a variety of metrics related to quality and time frames, such as the number of days that a property was listed for sale, different measures of sales returns, and completion of maintenance and repair work. For example, one of the enterprises uses performance results from monthly scorecards and quarterly report cards to ascertain whether its contractors are meeting its standards for performance. Its officials and those from two private mortgage servicers we spoke with said that they also used scorecard results to make decisions about reducing or ending their use of poorly performing contractors. Officials from some of these entities said that they also used scorecards to compare the performance of individual contractors to the performance of all contractors in a similar geographic area. These officials also said that the contractors knew how the scorecards were being used to assess their performance and that their business relationship with the company and the volume of work they received depend on the assessments. In the absence of a scorecard, FHA homeownership center staff indicated that they were taking steps to evaluate contractors individually and to provide feedback on their performance. But homeownership centers were using different processes that were inconsistent, fragmented, and informal. For example, staff at two centers said that they relied on individual measures of performance such as case reviews and summary reports of properties' progression through different stages of the disposition process. Staff from one of these centers explained that while they had a sense of whether a contractor was doing a good job or not, they did not have the ability to formally compare performance across contractors. Staff from one center told us that some staff members had created a scorecard-like tool to evaluate the performance of contractors for which they had oversight responsibilities, but had been told by HUD contracting officials that they could not share the results with the contractors until FHA introduced a standard scorecard nationwide. However, staff from a different office said that they had shared certain individual contractor performance information with their contractors. Without a functioning, standardized scorecard, FHA does not have a uniform tool for evaluating the overall level of its contractors' performance and cannot effectively make distinctions about relative performance differences across contractors or tell contractors how their performance compares to their peers. This shortcoming also limits FHA's ability to identify and address underperforming contractors and creates the risk that FHA cannot ensure the most efficient and effective disposition of its properties. FHA's Contractor Oversight Lacked Key Activities Used by Other Entities: Our review also showed that FHA was not conducting certain contractor oversight activities performed by some other entities that dispose of REO properties. Specifically, FHA was not conducting as many or as frequent in-person property inspections as other entities and was not taking steps to ensure that listing brokers were close enough to the REO properties they were chosen to market to know local market conditions and efficiently access the properties. One of the ways that FHA's oversight activities varied from other federally related housing entities and private mortgage servicers was the extent to which it conducts in-person property inspections. HUD's Contract Monitoring Guide states that inspections are the best way to determine the quality of a contractor's performance. FHA and other entities typically have their contractors visit properties regularly to inspect the properties' condition and to perform routine maintenance such as lawn cutting. However, they also have their own staff or third-party contractors that visit properties for oversight purposes, including ensuring that contractors are performing their required duties and maintaining properties to expected standards. According to staff from one of FHA's homeownership centers, performing in-person property inspections is critical because they allow FHA staff to review contractor performance, identify problems needing resolution, and conduct quality assurance checks, especially in the absence of a uniform scorecard. FHA's standard monitoring plans include property site visits to help ensure that maintenance contractors are conducting their own routine inspections and maintaining the condition of assigned properties in accordance with their contractual requirements. They also are meant to help ensure that marketing contractors are following required standards and procedures when conducting sales activities. FHA's plans call for in-person inspection of 2 percent of properties three times per year.[Footnote 50] However, FHA homeownership center staff had varying interpretations of FHA headquarters' expectations for the amount and timing of property inspections. For example, staff in some centers told us that they aimed to inspect 2 percent of properties annually, while staff in another center said they targeted 6 percent of properties annually. The timing of the inspections also varied across the homeownership centers from a certain percentage each month to a certain percentage in three of the four quarters of the fiscal year. According to FHA officials, a lack of adequate travel funds and staff capacity has created challenges for homeownership centers in conducting in-person property inspections. Staff at some centers said that a lack of available funds could delay some inspections until the end of the fiscal year, when funds might become available. Others noted that in the past they had inspected more properties within proximity to the homeownership center when travel funds were not available. However, waiting to conduct inspections until the end of the fiscal year and restricting them to a limited geographic area limits their effectiveness, as the inspections may not target properties in certain locations or contractors equally. Further, contractors may become aware that they are unlikely to have properties inspected early in the year or in certain locations. To supplement the in-person inspections, FHA attempts to use other means of monitoring whether its contractors are complying with its expectations, but the effectiveness of these efforts is also uncertain. For example, FHA homeownership center staff may conduct reviews of the evidence, such as photographs of the property, that contractors submit to document that they have performed routine inspections and other activities. However, in 2012 HUD OIG audits found that contractors could upload pictures that do not accurately depict a property's condition or incomplete reports that limited the effectiveness of these reviews.[Footnote 51] To supplement its own staffs' monitoring efforts, FHA's contractors that are responsible for marketing REO properties also complete some inspections as part of their quality control plans. Additionally, property listing agents employed by the marketing contractors inspect the work of maintenance contractors for all properties that are listed and sold. However, the effectiveness of these reviews may be limited because FHA staff told us that the property listing agents have often been reluctant to submit negative reports on maintenance contractors' performance because of fear of damaging working relationships with these other contractors. The number of in-person inspections that FHA completes may not be sufficiently effective to ensure that FHA's contractors are conducting their activities in compliance with contractual requirements. In multiple reports issued between March 2012 and February 2013, HUD's OIG found that contractors responsible for maintaining and marketing FHA's REO properties were often not performing the required work at all or were not performing to the expected level of quality.[Footnote 52] For example, a September 2012 HUD OIG report examined a Las Vegas, Nevada, FHA contractor and found that it did not secure or properly maintain 40 percent of the 96 properties that the OIG examined. [Footnote 53] Another report from September 2012 reviewed 125 properties nationwide and determined that FHA's contractors did not properly maintain 75 of them, as evidenced by unmaintained yards, unclean conditions, lack of security, and water leaks.[Footnote 54] The review also found that for 100 of the 125 properties, FHA's maintenance contractors did not conduct routine inspections in a timely manner.[Footnote 55] Furthermore, this OIG audit revealed that FHA's maintenance contractors nationwide were paid for inspections for which they had not completed the required documentation and may not have conducted. One of FHA's homeownership centers developed a report to identify these missing routine inspections and, in coordination with HUD's procurement office, requested reimbursement of fees paid to five of its contractors totaling more than $1.3 million for more than 10,000 inspections from June 2011 through February 2012. In contrast to the amount of in-person inspections done by FHA, other federally related housing entities and private mortgage servicers we spoke with indicated that they conducted in-person inspections of much larger percentages of their REO properties and conducted them more frequently. For example, staff from one of the enterprises told us that it performed in-person oversight inspections of 25 to 30 percent of its REO properties monthly using an independent inspection firm. [Footnote 56] The other enterprise said that it completed in-person inspections of around 35 percent of properties monthly, using both an independent inspection firm and its own staff. VA officials noted that it conducted in-person inspections on at least 10 to 20 percent of its REO properties annually and in fiscal year 2012 the agency inspected over 40 percent. In addition, staff from one of the private mortgage servicers we interviewed told us that its own field agents inspected about 40 percent of its properties, while two other private mortgage servicers said that they inspected about 7 and 10 to 12 percent of their properties on a monthly basis throughout the year. Federal internal control standards require that agency management conduct effective monitoring to assess program quality and performance over time and work to address any identified deficiencies. Other entities, whether federally related or private, found that frequent in- person property inspections were an effective way to better ensure that contractors were performing required activities and to assess the quality of their work. As a result of not conducting in-person inspections of a greater share of its REO properties and not inspecting them more frequently, FHA may not discover potential maintenance and disposition problems, potentially resulting in poorly maintained properties that sell for lower prices. In addition, FHA does not have the procedures that the enterprises have to ensure that properties are assigned to listing brokers located close enough to the properties to have sufficient knowledge of the local market. Using listing brokers that are close to and familiar with properties and the surrounding communities improves the chances that the properties will be shown as often as possible and will be well maintained. FHA's contracts require its marketing contractors to use local real estate professionals whose primary place of business is within reasonable proximity of the listed property. However, FHA does not have either a definition of "reasonable proximity" or formal guidelines or procedures for determining whether properties are assigned to local listing brokers, according to officials. FHA headquarters officials said that they had an informal goal of using brokers located within 20 miles of a listed property, but noted that consistently defining what constitutes reasonable proximity could be difficult. For example, distances that would be considered reasonable in an urban area might not be realistic in a rural location. FHA homeownership center staff said that headquarters had not yet implemented clear criteria or controls for the marketing contractors that assigned properties to brokers. In the absence of clear criteria from FHA headquarters, homeownership centers often made their own determinations on using brokers within reasonable proximity to properties. For instance, FHA staff at one center told us that after discovering as part of unrelated inspections that listing brokers for some of its properties were not local, the center and its marketing contractors had decided that "local" generally meant within 50 miles, with exceptions for sparsely populated areas. Centers also varied in their reviews of listing brokers' proximity to their listed properties. One center noted that such reviews were part of annual inspections of marketing contractors. However, officials at another center said that they did not believe the proximity of listing brokers was a major concern and did not monitor it closely and instead placed more emphasis on overall performance. Officials from one listing broker that has sold properties for two of FHA's marketing contractors said that there had been many instances of listed properties being more than 50 miles from the listing broker's office. They also noted that many of these more- distant listing brokers were not members of the listing service that includes properties for those local markets, which resulted in a lack of proper exposure for FHA's properties. Without clear guidance from FHA on the use and oversight of listing brokers, homeownership centers may continue to make their own determinations on what constitutes "reasonable proximity" to listed properties and may not be able to ensure that properties are being effectively marketed by knowledgeable agents. In contrast, the enterprises have established guidelines for the selection of local listing brokers and conduct monitoring to ensure brokers' proximity to the listed properties. For example, one of the enterprise's REO sales guidelines states that properties should be no further than 25 miles from the listing broker, although this threshold is used more often for rural areas, according to officials. In urban areas the goal is to assign a broker as close as 5 miles to the property. The officials also told us that the enterprise used reports to monitor the distances between listing brokers and their assigned properties and addressed situations involving longer distances on a case-by-case basis. Officials from the other enterprise emphasized the importance of using listing brokers located close to listed properties because of local brokers' market knowledge. This enterprise has a goal for broker proximity of about 15 to 20 miles, according to these officials, but the distances can vary in rural areas. The officials also told us that the enterprise had an Internet-based REO management system that assigned properties to listing brokers by geographic area. The enterprise's staff define the geographic areas within which its listing brokers can receive property assignments when it adds them as approved service providers and conduct reviews of listing brokers' office locations. FHA Has Not Assigned Work to Contractors Based on Evaluations of Their Performance: As part of the new M&M III program structure it introduced in 2010, FHA intended to implement a key quality control--assigning contractors work according to the quality of their performance--but has encountered obstacles to implementing this mechanism. The Federal Acquisition Regulation (FAR) stipulates a strong preference for using multiple contractors for the types of contracts that FHA has used to manage and dispose of REO properties.[Footnote 57] With respect to the types of contracts that FHA has used to manage and dispose of REO properties, the FAR provides that the agency must provide each contractor with a fair opportunity to be considered for the work. [Footnote 58] FHA designed its M&M III contract structure to include 10 geographic areas with multiple maintenance and marketing contractors operating within most areas. During the first year of the M&M III contracts, FHA assigned equal percentages of REO properties to each contractor in a contract area to satisfy the minimum guarantee under the contract.[Footnote 59] After the first year, FHA intended to use performance evaluations to help determine the shares of its REO properties within a contract area that it would assign to each of the multiple contractors operating in that area, with the high-performing contractors receiving the largest allocations of properties.[Footnote 60] FHA planned to divide property assignments within the overall contract on a percentage allocation basis, so that all individual contractors in an area would receive a minimum share of the work as long as the contractors met minimum performance requirements. However, FHA has been unable to implement this system as intended for two reasons. First, until it implements the planned scorecards or other uniform evaluation method, FHA has no way to systematically generate the information needed to assign work based on performance. Second, after implementing the new M&M III program structure, FHA encountered challenges in ensuring that its performance-based allocation contract structure complied with the FAR. In late 2011, HUD's OGC advised FHA that the process of allocating a minimum share of property assignments to each of its contractors was not compliant with the FAR rules requiring that each contractor have a fair opportunity to compete and win all the work for which it is competing. [Footnote 61] FHA officials said that the performance-based competitions that OGC determined would be compliant with federal acquisition rules, with contractors winning either all or none of the work assignments, would jeopardize the financial viability of some contractors. These officials explained that contractors have high overhead costs and could go out of business if they did not receive at least some property assignments for more than a few weeks. As the result of these obstacles to implementing performance-based allocations as planned, FHA has continued to assign properties to marketing contractors based on equal allocations in a contract area. It also assigns work to maintenance contractors solely on the basis of the cost. FHA staff told us that they could not identify an acceptable alternative to assigning work among multiple contractors that was also compliant with federal acquisition rules. Instead, FHA plans to award new contracts that give all property assignments in a contract area to a single contractor for at least a year. FHA staff said that if the contractor performed poorly or was unable to provide the necessary services, the property assignments could be shifted to a contractor in a neighboring area by redefining the contract area. However, the practical difficulties and challenges involved in redefining contract areas and reassigning all properties among contractors could make implementing this option difficult. Further, it is not clear that FHA has fully explored other options that are compliant with federal acquisition rules. In terms of performance incentives, FHA officials said that its discretion in determining whether to renew annual contracts was the most important. However, this particular incentive may be less powerful than the frequent reallocation of work envisioned under the M&M III contract structure. According to officials, FHA has never failed to renew an annual contract for the 50-plus maintenance and marketing contracts that have been part of the current contract structure since 2010. FHA staff told us that under the maintenance and marketing contracts' terms, they could reassign properties to another contractor or suspend property assignments to a poorly performing contractor, but these options have rarely been used. According to agency officials, since 2010 FHA has suspended a contractor's assignments only once for a period of 1 month and has not yet reassigned properties from one contractor to another on the basis of poor performance. FHA officials said that they had not defined standard criteria for the number of instances of deficient performance that would be required before these actions were taken. Rather, FHA staff perform a risk assessment, and the contractor is given an opportunity to address any deficiencies. FHA has tried to create other incentives for superior performance. It pays marketing contractors a percentage of a property's sale price at the time of sale based on the disposition price and time frame. These payments are higher for properties with sale prices above a set percentage of the initial listing price and within a certain time frame. Over the second half of 2012, almost one-third of FHA's sales met the thresholds for the higher fee amount, according to FHA data. However, FHA procurement officials told us that the bonus fee structure was not a normal contract incentive and that FHA also was considering including more typical performance incentives and disincentives in contracts. They explained that the only specific disincentive or performance penalty in the current contracts is FHA's ability to assess late fees if contractors delay sale closings. Without performance-based work assignments, however, FHA's ability to motivate contractor performance is limited because it has few other incentives and disincentives and uses them infrequently. As a result, FHA cannot ensure that its maintenance and marketing contractors are performing at the highest possible levels. In contrast, the enterprises and private market servicers we interviewed use performance-based work assignments to align contractor incentives and promote high performance. None of these entities are required to follow federal acquisition regulations. Officials from one of the enterprises explained that if one of its contractors consistently performed better than others based on scorecard assessments, the high-performing contractor would receive more work assignments, with the amount dependent on capacity considerations. This system applies to its national marketing contractors, listing brokers, maintenance contractors, and other service providers. The other enterprise also considers performance when assigning properties to contractors such as listing brokers, according to its officials. Officials from two private mortgage servicers we interviewed told us that they assigned additional work to contractors with better performance on their scorecard indicators and that poor contractor performance could lead to fewer work assignments or termination. Conclusions: The housing crisis has increased the number of REO properties in FHA's inventory. The agency's ability to effectively dispose of these properties in ways that maximize sales proceeds and minimize holding time could help increase the government's financial returns. We found that FHA's disposition performance and the time required to complete sales of REO properties lagged the performance of the government- sponsored enterprises. Our analysis of FHA's REO activities revealed that the agency was not employing some of the disposition practices that the enterprises and other housing entities used. These practices could be a factor in other entities' ability to dispose of REO properties for higher returns and with less holding time and include: * using multiple means of assessing property values to better assure that REO properties are fairly valued and thus more likely to sell faster and at the highest price, * making improvements to properties with characteristics that are more likely to result in a higher sales price if repaired, and: * basing price reductions for properties that do not sell at the original list price on market conditions rather than on a predetermined schedule. If FHA could perform as well as the enterprises in disposing of REO property, it could potentially generate hundreds of millions of dollars in additional sales proceeds and reduce maintenance and other holding costs from its future REO activities. Federal internal control standards call for agencies to have comprehensive policies and means to help ensure that program objectives are being met and that expected activities are being completed. However, FHA has not taken the actions necessary to ensure that its controls and oversight activities are effective in several areas. Specifically, * FHA lacks comprehensive guidance for its REO program and a process for updating this guidance as policies and procedures change. Having such guidance could better ensure consistent practices across homeownership center staff and uniform oversight of the numerous contractors that carry out maintenance and marketing activities. Further, having a robust revision process would allow FHA to incorporate best practices that it identifies by analyzing differences in performance across homeownership centers, something it currently does not do. * FHA has fully implemented neither mechanisms for evaluating contractors' activities against minimum expected standards nor a scorecard that would allow staff to compare contractor performance to identify high-and low-performing contractors. Addressing this issue is critical to better ensuring contractors are performing as expected and meeting program goals. * Given the FHA Inspector General's findings that REO properties were not always being well-maintained as required by the service providers' contracts, the number of in-person inspections currently being undertaken by FHA does not appear to be effective. * FHA lacks controls to help ensure that the brokers marketing its properties meet FHA contractual requirements that brokers be in reasonable proximity to their properties. As a result, FHA risks having brokers that do not have the expected level of local market knowledge and cannot conduct effective marketing activities because they are too far away. * FHA contracts generally lack incentives and disincentives that would encourage performing high-quality work consistent with other entities' practices. Implementing a more frequent performance-based assessment and assigning work on the basis of performance could improve returns on REO properties and reduce property holding times. Collectively addressing these issues could improve FHA's oversight of its contractors by, for example, ensuring that their properties are inspected regularly and that they face consequences for not meeting program requirements. Recommendations for Executive Action: To increase the potential for higher financial returns from FHA's disposition of REO properties, the Secretary, HUD, should direct the Commissioner, FHA, to identify and implement changes in current practices or requirements that could improve REO disposition outcomes, including: * requiring the use of multiple estimates of market value when determining initial list prices, * considering whether conducting repairs could increase the amount of net proceeds from specific property sales, and: * ensuring that the timing and amount of price reductions for its listed properties are made on the basis of an evaluation of market conditions rather than on standardized schedules. To improve its oversight of the REO disposition program, the Secretary, HUD, should direct the Commissioner, FHA, to: * update its REO program disposition handbook, or equivalent document, to include a current and consolidated set of policies and procedures for managing and disposing of FHA's REO properties; * establish a process for analyzing differences in disposition performance and practices across homeownership centers that can be used to periodically update this handbook or equivalent documentation to reflect current policy and procedures; * implement a mechanism for systematically reviewing contractors' compliance with minimum performance requirements through the use of standard metrics; * ensure the completion and implementation of the scorecard currently being developed, including ensuring that performance metrics included in the scorecard are consistent with those used to review contractors' compliance with minimum performance requirements; * determine more effective ways, including increased use of in-person inspections, to better ensure that contractors comply with expected requirements; * implement controls to ensure that listing brokers are located within close enough proximity to their listed properties to effectively market REO properties; and: * take steps to develop a legally acceptable means of assigning work to REO contractors that uses more frequent assessments of past performance. Agency Comments and Our Evaluation: We provided a draft of this report to HUD, FHFA, Fannie Mae, Freddie Mac, VA, and RHS for their review and comment. HUD provided written comments, which are reprinted in appendix II. Fannie Mae, Freddie Mac, and VA provided technical comments on the draft report, which we incorporated as appropriate. In a letter from the Assistant Secretary for Housing - Federal Housing Commissioner, HUD agreed with our recommendations. HUD also identified actions that it has taken or planned to take in response to our recommendations. For example, HUD wrote that FHA plans to update its REO disposition handbook. In addition, in response to our recommendation that FHA establish a process for analyzing differences in disposition performance and practices across homeownership centers that can be used to periodically update the handbook, HUD pointed to the monitoring plan that FHA has implemented for its contractors that will analyze disposition performance and practices across homeownership centers. HUD also wrote that any identified best practices will be noted, discussed, and communicated to homeownership centers and contractors. It will be important for FHA to also periodically update the handbook to reflect these changes in practices, as we recommended. HUD acknowledged that budgetary constraints affect implementation of contractor performance scorecards--critical elements in three of our recommendations--and limit its ability to make increased use of in- person property inspections that we suggested could be used to better ensure that contractors comply with expected requirements. While recognizing that FHA's scope for action may be limited by available budgetary resources, we emphasize the importance of considering not just the costs to undertake these steps but also the potential savings and improved disposition outcomes that would be realized from enhanced contractor oversight. In response to our recommendation to develop a legally acceptable means of assigning work to REO contractors that uses more frequent assessments of past performance, HUD said that FHA has taken steps in its new REO contracts to provide incentives to high-performing contractors and disincentives to lower-performing contractors by transitioning inventory among them based on performance and price. When implementing such a contract structure, we encourage FHA to consider inventory transitions on a frequent basis, such as quarterly, to align with the frequency of the scorecard performance assessments. In technical comments, the Director of Regulatory Affairs of Fannie Mae noted that the REO execution rate performance information that our report presents was inconsistent with publicly disclosed loss severity rates published by FHA. Although loss severity rates--which measure loss on a defaulted loan as a percentage of the unpaid principal balance--can be presented when discussing REO performance, we did not include such analysis in our report because this measure reflects factors beyond the control of the REO programs of these entities. For example, loss severity rates can be affected by the original loan-to- value ratio, loan amortization schedule, origination date, changes in market values, or the existence of mortgage insurance. We therefore do not use loss severity rates to assess REO performance. Fannie Mae also noted that the performance execution information that uses independent valuations may not be comparable across entities because not all REO sellers use the same valuation methodology. Our report notes that the entities use different methods for obtaining an independent valuation-- including an independent appraisal for FHA and one of the enterprises and an independent BPO for the other enterprise--and that any systematic differences between the appraisals and BPOs could affect the performance results. We also calculated execution rate results using list prices and these calculations showed similar results. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to interested congressional committees, HUD, FHFA, Fannie Mae, Freddie Mac, VA, and RHS. In addition, the report will be available at no charge on the GAO website at [hyperlink, http://www.gao.gov]. If you or your offices have any questions about this report, please contact me at (202) 512-8678 or sciremj@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix III. Signed by: Mathew J. Scirè: Director: Financial Markets and Community Investment: [End of section] Appendix I: Objectives, Scope, and Methodology: Our objectives were to examine (1) real estate-owned (REO) property disposition practices used by the Housing and Urban Development Department's (HUD) Federal Housing Administration (FHA) and other federally related housing entities and how FHA's effectiveness compares to that of these entities, and (2) how FHA oversees its REO disposition program. To examine the disposition practices, we reviewed REO program regulations, requirements, and policies to determine the goals and strategies for these activities of FHA and other federally related housing entities, including the Department of Veterans Affairs (VA), the Department of Agriculture's Rural Housing Service (RHS), and two housing government-sponsored enterprises (the enterprises)--Fannie Mae and Freddie Mac. We interviewed HUD and FHA staff in the agency's Washington, D.C., headquarters and staff in the four regional homeownership centers that oversee REO activities in their areas. We also interviewed staff from other federally related housing entities about the goals and strategies of their REO activities. We also discussed REO goals and strategies with staff from three large private- sector mortgage servicers that also acquire and dispose of REO properties. We selected these servicers because they were among the largest servicers of home mortgages. To obtain additional information on REO activities, we also interviewed staff from the National Association of Realtors, two local realtors identified as knowledgeable about REO properties by that association, and community groups or government housing entities in various cities with large REO concentrations. We also interviewed the National Community Stabilization Trust, which administers a database of REO properties for purchase by community groups, and an appraisal group responsible for promoting appraisal standards. To assess the effectiveness of FHA's REO dispositions, we obtained and analyzed REO disposition data from FHA and the other federally related housing entities, including all REO properties disposed of from January 1, 2007, through June 30, 2012, as well as properties in inventory at the end of the period.[Footnote 62] Specifically, we obtained data from the data management systems of FHA, the enterprises, VA, and RHS. We did not include RHS in our analysis of REO property disposition performance because it only obtains and manages REO properties through its direct loan program and these are very small in number compared to the other entities. For example, in the period of analysis, RHS's direct loan property dispositions represented less than 1 percent of FHA's REO dispositions. RHS also did not have property-level data available for many of the data elements that we included in our analyses. Table 1 shows the number of dispositions for each entity that occurred during this time period based on the data that we received and analyzed. For RHS, we used data on its direct loan properties because it only acquires and disposes of REO properties related to its direct loans and does not do so for properties from its guaranteed loan program. Table 1: Number of REO Dispositions for FHA and Other Entities from January 1, 2007 through June 30, 2012: FHA: 417,303; Enterprise A: 741,623; Enterprise B: 352,847; VA: 65,273; RHS: 3439. Source: GAO analysis of FHA, Freddie Mac, Fannie Mae, VA, and RHS data. Note: We do not identify the identity of the enterprises for certain analyses. We use "enterprise A" and "enterprise B" when referring to the entities' disposition volumes. [End of table] We assessed the reliability of these data by reviewing agency data documentation, interviewing officials, and testing for missing values, outliers, and obvious errors. VA and RHS did not have property-level data available that were necessary for some of the calculations in this report. In particular, VA and RHS did not have data available on initial REO valuation dates, type of sales method, and type of buyer. Additionally, VA did not have data on initial REO appraisal amounts and RHS did not have data on net sales proceeds. We excluded VA and RHS from calculations that used these data elements and noted these instances as applicable. We addressed missing data by excluding those properties with missing data elements from analyses that relied on those elements as necessary. Finally, in analyses using the net sales proceeds data element, we excluded properties with net sales proceeds of $0, as this is an unlikely value for that element and could indicate missing data. We excluded additional properties from our regression models, as described below. In no instance did we exclude more than 13 percent of all properties. After making the necessary qualifications, corrections, and related assumptions, we believe that the data were reliable for our purposes as described in this appendix. To determine the performance in maximizing the sales price of REO dispositions by the entities, FHA homeownership centers, and FHA asset managers, we calculated various execution rates. These rates included the ratio of REO dispositions' gross sales prices (for gross execution rates) or net sales proceeds (for net execution rates) to one of several estimated property valuations.[Footnote 63] The measures of property values that we used included (1) the sum of an independent assessment of a property's value--an independent appraisal or BPO conducted by a third party if the entity did not use appraisals--plus any repair costs for that property, and (2) the initial list price for the REO property.[Footnote 64] We calculated aggregate execution rates for each organization by dividing the sum of their property dispositions' net sales proceeds by the sum of the properties' independently assessed values or initial list prices. We used aggregate execution rates because, by showing the net return on the total estimated value of properties in an entity's portfolio, the aggregate rates better reflect the entity's overall performance than an average of the execution rates of individual properties. The aggregate rate calculates execution rate performance on a value- weighted basis, with higher value properties having a greater impact on the aggregate rate than low-value properties. An average of individual properties' execution rates gives equal weight to properties of different values despite their unequal effects on total net returns. To assess the timeliness of the entities' REO property dispositions, we calculated the average number of days from the foreclosure sale to the day on which they sold or otherwise disposed of the property, as well as the number of days they took to move properties between various points within the REO process. To account for the possibility that differences in performance results between FHA and the enterprises might be due to differences in the characteristics of the properties that each entity acquired and disposed of in their REO programs, we used the data on the REO property dispositions from January 1, 2007, through June 30, 2012, to create various regression models. The general regression specification for these models was: [For formula, refer to PDF] where: * y is the performance measure being assessed, such as net execution as a percent of independently assessed value or time from foreclosure to REO sale closing; * organization is an indicator variable for the entity holding the property; * X is a series of control variables related to the property and represents the relationship between variable j and the outcome variable, independent of the other variables in the model; * zip is an indicator variable for the property's ZIP code; * valuation indicates the range of the property's independently assessed value;[Footnote 65] * a, b, and g ? are the parameters of interest; and: * e represents an error term. We interacted the ZIP code and valuation category variables to allow the average effect of the valuation category on the performance measure to vary between ZIP codes. The parameters of interest were the coefficients for each organization. We used the following additional property characteristics as control variables in the regressions: * year the property was built (as a categorical variable), * number of bedrooms (as a categorical variable), * number of units for the property (as a categorical variable), * the change in the FHFA House Price Index for the property's ZIP code from the month of the initial valuation to the month one quarter after the initial valuation, and: * the change in the FHFA House Price Index for the property's ZIP code from the month of the initial valuation to one month after the initial valuation month. In addition, for regressions with an execution rate as the outcome variable, we included the month and year of REO acquisition as a categorical control variable. For regressions with the number of days from the foreclosure sale to the REO disposition as the outcome variable, we included the number of weeks from the scheduled date of the last completed payment to the date of the foreclosure sale as a categorical control variable.[Footnote 66] For the regression models with an execution rate as the outcome variable, we weighted every property by the amount of the property's independently assessed value. This was done so that the percentage point differences resulting from the regressions would be similar to overall aggregate differences for models where only the entity dummies were included. Weighting by property value also reduced the effect of heteroskedasticity that was likely present over the range of valuations.[Footnote 67] To reduce the effect of outliers on the regression estimates, we excluded certain property records that had irregular values for a few data elements. For the regression models with an execution rate based on either the independently assessed value or reconciled value as the outcome variable, we excluded properties with independently assessed value net execution rates greater than 300 percent or less than negative 300 percent. Such large or small execution rates suggest a highly inaccurate valuation or otherwise irregular data such as a very low property value or sales price. For models with execution rate based on initial REO list price as the outcome variable, we excluded properties with the highest and lowest 1 percent of values for the ratio of initial REO list price to independently assessed value for all properties. In total, we excluded fewer than 12 percent of all properties in our data for each of the net execution models. Due to these exclusions, the population of properties used in the net execution regression models was smaller than the full population that was used when calculating the aggregate results. As a result, to estimate the percentage point difference in net execution rates between FHA and the enterprises for the full population, we multiplied the percentage point differences from the aggregate execution rates by the ratio of the percentage point differences resulting from the full regression model to the percentage point differences resulting from a regression model with only entity variables. For the models where the time from the foreclosure sale to the completed REO property disposition was the performance measure, we excluded properties that had a negative value for the number of days from the foreclosure sale to the completed REO property disposition. In total, we excluded less than 13 percent of all properties for the time frame models. We took various steps to test these regression models for robustness. * We used various thresholds for the exclusion of property records from the net execution models. For example, we tested results when the models excluded properties with independently assessed values above and below certain amounts rather than properties with execution rates above and below certain amounts. * We used various specifications for certain property and market characteristics. For example, we tested models where the categorical variable for independently assessed value was not interacted with ZIP code, where it was specified as a log, and where it was specified as a polynomial. We also tested models where the change in the FHFA House Price Index in each ZIP code was specified as the change in the index in the year prior to the disposition date and in the month prior to that date. * We included additional property characteristics as a control variable in the models or excluded some of the property characteristics. For example, we tested models where a property's occupancy status at REO acquisition was included as a control variable. This variable ultimately was not included in our final models because a significant percentage of FHA's property dispositions were missing data for this variable. These models showed qualitatively similar results for the reduction in the performance difference between FHA and the enterprises relative to regressions with only entity variables. However, the final model for net execution rates had one of the largest reductions in the performance difference between FHA and the enterprises. This appeared to be due to the interaction of properties' independently assessed values and ZIP codes. To identify factors that could help explain differences in FHA's REO performance relative to that of the enterprises, we also created models that included various factors (time from foreclosure sale to REO disposition, ratio of initial list price to initial valuation amount, disposition method, and buyer type) at least partially under the entities' control and examined how their inclusion affected estimates of the performance difference between the entities as measured by net execution rates. To compare the performance of FHA homeownership centers and contractors, we computed execution rates and average time frames for the properties they managed. When examining time frames, we focused on the time frame from a property's initial valuation to completion of the disposition because initial valuation was the date available from both FHA and the enterprises that most closely reflected when FHA's homeownership centers begin managing and overseeing the disposition of REO properties. In addition, when comparing homeownership center performance, we developed regression models to account for the possibility that differences in performance results between the homeownership centers might be due to differences in the characteristics of the properties that each center disposed of in their REO programs. To further control for regional differences that our nationwide regression models may not have been able to capture, we compared the performance of the homeownership centers to the performance of the enterprises in the states where each center operates. For this analysis, we also conducted separate regressions for each homeownership center region to account for differences in the characteristics of properties between homeownership centers and the enterprises in those states. For the regression models used to compare the performance of the homeownership centers, we controlled for the same property and housing market characteristics that we included in our regression models comparing FHA's overall performance to that of the enterprises.[Footnote 68] However, we did not include ZIP code in the models directly comparing homeownership center performance because the homeownership centers operate in regions with mutually exclusive ZIP codes. To examine the price reduction strategies used by FHA, we also obtained additional data on all listing prices for properties that FHA disposed of from June 30, 2011, through June 30, 2012. To determine if price reductions occurred at a scheduled amount, we compared the price reductions reflected by these data to the price reduction schedules, if any, listed in FHA contractors' marketing plans. One contractor did not have a price reduction schedule in its marketing plan so we considered price reductions for properties managed by this contractor to be according to schedule if they occurred at one of the two standard reduction percentages that appeared in most other contractors' marketing plans. In addition, some marketing plans for one contractor specified a range for the price reduction amount. In these cases, we considered a price change to be for a scheduled amount if the price was reduced by one of the two standard reduction percentages that appeared in most contractors' marketing plans. Finally, for marketing plans that specified that price changes could be up to a certain amount, we considered a price change to be for a scheduled amount only if the price change was for the maximum amount, since the goal of this analysis was determine how often price changes were for particular, predictable, amounts. To determine how the entities oversee their REO programs, including the contractors they use to perform various REO-related activities, we interviewed staff from FHA and the housing entities. We also reviewed program regulations, requirements, and policies related to oversight. We also discussed REO oversight activities with the staff from the three private sector mortgage servicers. [End of section] Appendix II: Comments from the Department of Housing and Urban Development: U.S. Department of Housing and Urban Development: Assistant Secretary for Housing: Federal Housing Commissioner: Washington, DC 20410-8000: June 12, 2013: Mr. Mathew J. Scirè: Director: Financial Markets and Community Investment: Government Accountability Office: 441 G Street, NW: Washington, DC 20548-0001: Dear Mr. Scirè: Thank you for the opportunity to respond to the Government Accountability Office (GAO) draft report GAO-13-542 of May 28, 2013, entitled, "Improving Disposition and Oversight Practices May Increase Returns on Foreclosed Property Sales " This letter conveys the Department of Housing and Urban Development (HUD) response to the audit. With regards to the recommendations found in the report, the Federal Housing Administration's (FHA) specific responses are below. But at the outset, it should be noted that FHA agrees with the GAO's recommendations. GAO Recommendation: To increase the potential for higher financial returns from FHA's disposition of real estate owned properties (RHO), the Secretary, HUD, should direct the Administrator, FHA, to identify and implement changes in current practices or requirements that could improve REO disposition outcomes, including: Recommendation 1: Require the use of multiple estimates of market value when determining initial list prices. HUD Response: FHA agrees with the recommendation to use multiple valuation tools to determine initial list price. Currently, the Santa Ana Homeownership Center (HOC) is running a pilot using appraisals, automated valuation models (AVM), and broker price opinions (BPO) to set the initial list price. In addition, HUD is scheduled to conduct a second pilot using various pricing tools as noted above with a price evaluation contractor to ascertain initial list price, acceptance/rejection of offers, counteroffers and reserve prices. Once the two pilots are completed, an analysis will be performed to determine which method is more effective in increasing FHA's net return. Lastly, the Office of Single Family Asset Management (OSFAM) forwarded a Property Disposition term sheet of comprehensive property disposition regulatory changes, in May 2013, to the FHA Commissioner (not the Administrator as stated in the GAO recommendation above) to he assigned to OGC. One of the key changes in the term sheet that FIFA is considering is regulatory changes that would allow FHA to use multiple valuation tools to establish a list price. Recommendation 2: Consider whether conducting repairs could increase the amount of net proceeds from specific property sales. HUD Response: FHA is currently conducting a pilot repair program in the Atlanta HOC with a focus on limited repairs to convert a property in fairly good condition to a move-in ready condition. FHA will also continue to review its results and to weigh the cost-benefit of an overall repair program to identify the optimal approach. Recommendation 3: Ensure that the timing and amount of price reductions for its listed properties are made on the basis of an evaluation of market conditions rather than on standardized schedules. As a part of the pilot referencing pricing tools, FHA is exploring ways it can enhance its current list price reduction procedures. HUD Response: FHA is currently revising its market analysis language in the Performance Work Statement to clarify and emphasize that reanalyzing properties should not result in automatic price reductions. Instead, a reduction must be based on market conditions and summary data from the initial listing. GAO Recommendation: To improve its oversight of the REO disposition program, the Secretary, HUD, should direct the Administrator, FHA, to: Recommendation 4: Update its REO program disposition handbook, or equivalent document, to include a current and consolidated set of policies and procedures for managing and disposing of FHA's REO properties. HUD Response: The Office of Single Family Housing has a plan of action in place and a vendor under contract to help with the update (including electronic updates) of all of the office's outdated handbooks. The REO Property Disposition handbook is a priority on the plan. Recommendation 5: Establish a process for analyzing differences in disposition performance and practices across homeownership centers that can be used to periodically update this handbook or equivalent documentation to reflect current policy and procedures. HUD Response: The Office of Single Family Housing has implemented a monitoring plan for its Asset Managers and Field Service Managers to analyze differences in disposition performance and practices across 1-10Cs. The Principal Government Technical Representatives (PGTRs) review the plans and prepare a monthly report for senior management. When best practices are identified in these plans, they are noted, discussed during the REO Directors' weekly conference call, and communicated to all four HOCs and contractors. Recommendation 6: Implement a mechanism for systematically reviewing contractors' compliance with minimum performance requirements through the use of standard metrics. HUD Response: The implementation of the Scorecard that is in development will ensure a mechanism for systematically reviewing contractors' compliance with minimum performance requirements. Recommendation 7: Ensure the completion and implementation of the scorecard currently being developed, including ensuring that performance metrics included in the scorecard are consistent with those used to review contractors' compliance with minimum performance requirements. HUD Response: FHA agrees with the recommendation pertaining to the Scorecard implementation; however, the implementation is subject to budgetary constraints (i.e., Working Capital funds). Recommendation 8: Determine more effective ways, including increased use of in-person inspections. to better ensure that contractors comply with expected requirements. HUD Response: FHA consistently explores opportunities to increase the effectiveness of its oversight efforts. However, budgetary and resource restraints limit the feasibility of increasing the number of in-person inspections conducted. Recommendation 9: Implement controls to ensure that listing brokers are located within close enough proximity to their listed properties to effectively market REO properties. HUD Response: The Office of Single Family Housing will work with the Office of General Counsel (OGC) to determine if there is an effective way of defining "close proximity." This definition has been a long-standing issue for OGC. Recommendation 10: Take steps to develop a legally acceptable means of assigning work to REO contractors that uses more frequent assessments of past performance. HUD Response: FHA has taken steps to promote competition and ensure sufficient coverage in a single award environment. The 3.5 Asset Manager and 3.6 Field Service Manger contracts include a realignment clause, 11.2 Option To Increase/decrease The Geographic Service Area. This will allow FHA to transition inventory based on performance and price (i.e., available under a vendor's contract ceiling). The H.2 clause is intended to incentivize exceptional performance by allowing the highest performing contractors the opportunity to expand their geographic contract areas, and at the same time, provide a disincentive to lower performing contractors by reducing their geographic footprint. Also, the Office of Single Family Asset Management is currently working to finalize scorecards to aid in the review of performance on a quarterly basis. We appreciate the efforts of the GAO to review the REO inventories and suggest steps to improve disposition and oversight practices to increase returns on foreclosed property sales. Sincerely, Signed by: Carol J. Galante: Assistant Secretary for Housing-Federal Housing Commissioner: [End of section] Appendix III: GAO Contact and Staff Acknowledgments: GAO Contact: Mathew J. Scirè, (202) 512-8678, or sciremj@gao.gov: Staff Acknowledgments: In addition to the individual named above, Cody Goebel, Assistant Director; Kevin Averyt; Stephen Brown; Emily Chalmers; William R. Chatlos; Robin Ghertner; DuEwa Kamara; John Karikari; Jon Menaster; Alise Nacson; Jessica Sandler; Jena Sinkfield; Jack Wang; William T. Woods; and Ethan Wozniak made key contributions to this report. [End of section] Footnotes: [1] FHA also acquires REO properties from lenders that obtained them through a process known as deed-in-lieu of foreclosure whereby a delinquent borrower avoids foreclosure but is required to transfer title to the property to the lender. [2] Unlike its direct loan program, RHS does not take possession of REO properties for the loans that it guarantees but rather oversees their disposition by lenders with the goal of minimizing losses. [3] While we collected and analyzed data on all dispositions during the period, some dispositions were missing certain data elements and could not be included in all of our calculations. For more detailed information on the share of dispositions by entity that were included in our calculations, please see the notes for figures 4-11. [4] We did not attempt to verify the extent to which these other entities complied with their policies and practices. [5] The Mutual Mortgage Insurance Fund, which supports this program, is required by law to contain sufficient reserves and funding to cover the estimated future payment of claims on foreclosed mortgages and other costs. [6] Conventional mortgages do not carry government insurance or guarantees. [7] FHA requires servicers to consider, and when qualified implement, home retention and disposition options in a specific order with specified time frames. [8] The Protecting Tenants at Foreclosure Act of 2009 protects certain tenants from immediate eviction by new owners who acquire residential property through foreclosure. The provisions are scheduled to expire on December 31, 2014. 12 U.S.C. §§ 5201 note, 5220 note. [9] FHA refers to its maintenance contractors as Field Service Managers and its marketing contractors as Asset Managers. [10] The activities performed by the mortgagee compliance manager include reviewing property inspections to ensure the property is in conveyance condition, resolving conveyance exceptions, and providing guidance to mortgagees related to pre-and postconveyance responsibilities. [11] 24 C.F.R. § 291.1(a)(2). [12] The HUD Secretary's Office of Strategic Management and Planning defines strategic goals for FHA's REO property disposition in its Management Action Plan. The Neighborhood Stabilization Program was established to try to reduce inventories of foreclosed homes by providing funding to state and local governments to, among other things, acquire and rehabilitate or demolish foreclosed and abandoned homes. [13] If the property is deemed uninsurable, the exclusive owner- occupant purchase period is reduced to 5 days from the initial listing date. [14] Revitalization areas are HUD-designated geographic areas authorized by Congress under provisions of the National Housing Act and are intended to promote revitalization through expanded home ownership opportunities. [15] Net sales proceeds represent the cash received by the seller at settlement, or the contract sales price less deductions for items such as seller concessions and broker commissions. Independently assessed value represents the estimated value from an independent appraisal for FHA, one of the enterprises, and VA. For the other enterprise it represents an independent broker's price opinion, because according to its officials the enterprise does not obtain appraisal values for all of its properties. We do not identify the identity of the enterprises for certain analyses; rather, we use "enterprise 1" and "enterprise 2" to refer to each of the entities. [16] FHA and the enterprises each obtain a value assessment from independent parties. Although the enterprises also use additional inputs when determining a property's value, we used only the independent appraisals and broker's price opinions for this calculation, since they provided a more consistent basis for comparison. However, because the independently assessed value for FHA and one enterprise came from an appraisal, whereas the independently assessed value for the other enterprise came from an independent broker's price opinion, any systematic differences between the appraisals and broker's price opinions may be reflected in the independently assessed value execution rate results. [17] We calculated the aggregate net execution rate by totaling the net proceeds received from each disposition and dividing them by the total independently assessed values for the properties disposed of during the period. For execution rates using independently assessed value, we added any actual repair costs to the assessed value in the ratio's denominator to approximate the value of the property as repaired. [18] See appendix I for more information on how we controlled for differences in the characteristics of each entity's REO properties. [19] The gross sales price is the contract price before deductions for items such as seller concessions and broker commissions. [20] The estimated additional revenue is based on a 5-percentage-point difference in net execution rate after controlling for certain property characteristics and a total of 114,300 dispositions in 2011 at an average appraised value of $80,000. We used enterprise 2's independently assessed value net execution rate in this calculation because this enterprise and FHA both used appraisals to value their properties, which provided the most consistent basis for comparison. [21] Although the REO data necessary to calculate sales return performance were not available from VA and RHS, VA data on the overall REO disposition time frame were available and reliable for our purposes. [22] As noted previously, FHA acquires its REO properties later than other entities. The average time from the foreclosure sale to FHA's acquisition date was 161 days on average for properties that it disposed of from January 2008 through June 2012. [23] About 13 percent of the REO property dispositions for which we received data from FHA were missing a foreclosure sale date. Because the majority of these records were for dispositions in 2007, we used January 2008 as the starting point for our analysis of REO disposition time frames. [24] See appendix I for more information on how we controlled for differences in the characteristics of each entity's REO properties. [25] This amount may vary as it is based on an estimate of FHA's average daily holding costs of $38. FHA said that this amount generally represented the period beginning when it accepts conveyance of an REO property, but we assumed that holding costs for the preconveyance period were similar to that for the postconveyance period. The estimated extra holding costs associated with extended REO disposition timelines was based on 2011 dispositions of 114,300 properties and average days from foreclosure to REO sale of 340 for FHA, 201 for enterprise 1, and 208 for enterprise 2. [26] For example, see GAO, Vacant Properties: Growing Number Increases Communities' Costs and Challenges, [hyperlink, http://www.gao.gov/products/GAO-12-34] (Washington, D.C.: Nov. 4, 2011); and Mortgage Foreclosures: Additional Mortgage Servicer Actions Could Help Reduce the Frequency and Impact of Abandoned Foreclosures, [hyperlink, http://www.gao.gov/products/GAO-11-93] (Washington, D.C.: Nov. 15, 2010). [27] VA did not have data available for the dates of its REO properties' initial value assessments. [28] FHA officials noted that its timeline could also have been affected by the fact that one of its largest servicers delayed the conveyance of a significant number of foreclosed properties in recent years because of uncertainty about a review of foreclosure processing by federal banking regulators. These officials explained that FHA had no requirements on the amount of time servicers were allowed to file claims and convey foreclosed properties to FHA. [29] See GAO, Single-Family Housing: Opportunities to Improve Federal Foreclosure and Property Sale Processes, [hyperlink, http://www.gao.gov/products/GAO-02-305] (Washington, D.C.: Apr. 17, 2002). [30] In 2011, we reported that while FHA's REO property inventory increased by 85 percent from the end of 2006 to the end of 2010, its homeownership center staff increased by 13 percent from 2006 to 2010. See GAO, Federal Housing Administration: Improvements Needed in Risk Assessment and Human Capital Management, [hyperlink, http://www.gao.gov/products/GAO-12-15] (Washington, D.C.: Nov. 7, 2011). [31] For our purposes, we defined repaired properties as those that had at least $1,000 in total repair costs. [32] FHA's net execution rates were lower for repaired properties than for unrepaired properties, but FHA only conducts repairs when they are deemed necessary for health and safety reasons rather than to increase property values as do the enterprises, according to agency officials. [33] Because the decision to repair a property is based on various property characteristics, it is possible that those property characteristics that led the enterprises to do repairs are responsible for the higher execution rates. Still, as discussed later, the enterprises and private market servicers we interviewed told us that repairing certain properties led to higher returns. [34] Under the 203(k) rehabilitation mortgage insurance program, however, FHA sells properties to borrowers with a repair escrow account for repairs, including those that might be necessary to qualify for FHA insurance. This program enables borrowers to finance both the purchase of a property and the cost of its rehabilitation through a single mortgage or to finance the rehabilitation of their existing property. [35] The Davis-Bacon Act requires that all laborers and mechanics employed by contractors and subcontractors performing on certain federal contracts must be paid at least the prevailing wage rate, including fringe benefits, in the local area in which they are employed, as determined by the Secretary of Labor. In addition, contractors are required to pay these workers weekly and submit weekly certified payroll records to the contracting or administering agency. 40 U.S.C. §§ 3141-3144, 3146-3148. [36] Any reduction in holding costs would depend on whether the time needed to complete repairs exceeded the additional time typically required for FHA to sell ineligible properties. [37] See 24 C.F.R. § 291.100(b), which requires that the list price assigned to an REO property be based upon an appraisal conducted by an independent real estate appraiser using nationally recognized industry standards for the appraisal of residential property. [38] Because one of the enterprises generally does not obtain an appraisal, we compared the initial list prices of its properties to the values from the independent BPO. The percentages for both enterprises were similar when calculated only for properties with less than $1,000 in repair costs. [39] For marketing plans that specified a range for the reduction or did not contain a schedule, we considered a price change to be for a scheduled amount only if the price was reduced by one of the two reduction percentages specified in most other plans. For marketing plans that specified that price changes could be up to a certain amount, we considered a price change to be for a scheduled amount only if the price change was for the maximum amount. [40] See GAO, Standards for Internal Control in the Federal Government, [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: Nov. 1, 1999). [41] The plans, methods, and procedures used to meet missions, goals, and objectives are a fundamental component of internal control, which itself is an integral component of an organization's management and seeks to ensure the effectiveness and efficiency of agency operations. See GAO, Standards for Internal Control in the Federal Government, [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. [42] HUD OIG, Ofori & Associates, PC, Hartford, CT, Single Family REO Contract Administration, 2013-BO-1001 (Boston, MA: Feb. 19, 2013). [43] Because of missing data for many FHA property dispositions in 2007, we used January 2008 as the starting point for our analysis of REO disposition time frames. [44] We compared the time from initial valuation to REO sale since that period is consistent for each of the entities and most closely approximates the period for which FHA's homeownership centers have responsibility for property disposition activities. [45] Data for 2012 are from January 1ST to June 30TH. [46] Data for 2012 are from January 1ST to June 30TH. [47] HUD, Office of the Chief Procurement Officer, Contract Monitoring Desk Guide (Washington, D.C.: March 2009), and HUD, Office of the Chief Procurement Officer, Handbook 2210.3 Procurement Policies and Procedures, Revision 9 (Washington, D.C.: April 2009). [48] This work was finalized and issued as HUD OIG, Real Estate-Owned Management & Marketing III Program, Washington, D.C., 2012-LA-0003 (Los Angeles, CA: Sept. 18, 2012). [49] The contract monitoring tasks are analysis of the adequacy of contract funding levels; measurement of contractor performance in key areas; training of homeownership center staff on the monitoring of REO contractors; training of REO contractors on FHA's performance expectations; review of REO contractors' entry of data and documentation into FHA's property management database; visiting REO properties to inspect contractors' work; visiting REO contractors' work sites to evaluate the appropriateness of office technology and personnel; developing and implementing communication between homeownership center and headquarters staff; reviewing property status reports from FHA's property management database; and reviewing disposition status reports for properties that exceed certain REO processing time frames (such as properties that have been in REO inventory for more than a year). [50] Specifically, the plans state that REO properties selected for inspection should include 2 percent of newly acquired properties from the most recent quarterly period and 2 percent of other properties that do not yet have an accepted sale offer. The plans further state that the frequency should be three times per year--January through September. [51] HUD OIG, Single Family REO Contract Administration: Ofori & Associates, PC, Hartford, CT, Did Not Always Comply With Its REO Contract and Marketing Plan Requirements, 2013-BO-1001 (Boston, MA: Feb. 19, 2013). [52] See, for example HUD OIG, Innotion Enterprises, Inc., Las Vegas, NV, Single Family REO Contract Administration, 2012-LA-1010 (Los Angeles, CA: Sept. 12, 2012); HUD OIG, HUD's Oversight - REO M&M III Contract Administration, LA-12-0019 (Los Angeles, CA: Mar. 30, 2012); HUD OIG, Evaluation of HUD's Management of Real Estate-Owned Properties, IED-12-001R (Washington, D.C.: Apr. 19, 2012); HUD OIG, Single Family REO Contract Administration: Ofori & Associates, PC, Hartford, CT, Did Not Always Comply With Its REO Contract and Marketing Plan Requirements, 2013-BO-1001 (Boston, MA: Feb. 19, 2013). [53] HUD OIG, Innotion Enterprises, Inc., Las Vegas, NV, Single Family REO Contract Administration, 2012-LA-1010 (Los Angeles, CA: Sept. 12, 2012). [54] HUD OIG, Real Estate-Owned Management & Marketing III Program, Washington, D.C., 2012-LA-0003 (Los Angeles, CA: Sept. 18, 2012). [55] The HUD OIG report noted that FHA's maintenance contracts stated that, at a minimum, properties must be inspected once every 2 weeks. In some cases, it found that the routine inspections were late by 1 to 33 days. [56] This enterprise selects the inspection sample from available REO properties, which does not include properties in eviction, redemption, litigation, or with disposition activities on hold for other reasons. The inspections are documented and reviewed by the enterprise's staff each month. [57] Federal Acquisition Regulation (FAR) § 16.504(c). [58] FHA has used multiple award indefinite-delivery, indefinite- quantity contracts which allow FHA to award contracts to multiple contractors for their services, but provide flexibility in both quantities and delivery scheduling. FAR § 16.501-2. When FHA determines its specific service requirements, it awards a task order to one of the contractors for those requirements. FAR § 2.101. See also, FAR § 16.505(b)(1). [59] Placing an order to satisfy a minimum guarantee is an exception to the fair opportunity process. FAR § 16.505(b)(2)(i)(D). [60] While contractor performance can be used as a basis for determining future awards of work, the FAR requires agencies to also consider cost. FAR § 16.505(b)(ii)(E). [61] FAR § 16.505(b)(1). In addition, the FAR provides that an agency may not use an allocation or designation method that would not result in fair consideration being given to all awardees prior to placing an order. FAR § 16.505(b)(1)(ii)(B). [62] We collected information on all single family REO properties defined as having 1-4 units that were disposed of during the period. [63] Net sales proceeds represent the cash received by the seller at settlement, or the contract sales price less deductions for items such as seller concessions and broker commissions. [64] For FHA, one of the enterprises, and VA, we used the estimated value from an independent appraisal as the independently assessed value. For the other enterprise, we used an independent broker's price opinion as this enterprise's officials said that it does not obtain appraisal values for all of its properties. [65] We used valuation categories of $0 to less than $25,000; $25,000 to less than $50,000; $50,000 to $100,000; $100,000 to less than $150,000; $150,000 to less than $200,000; and greater than $200,000. [66] We included this variable despite 13 percent of FHA's dispositions missing foreclosure sale dates because those properties already were excluded from the regressions with the days from the foreclosure sale date to the disposition date as the outcome variable. [67] The variance in the execution ratios was generally higher for lower value properties. [68] FHA's homeownership centers comprise distinct geographical areas and the effects of property characteristics, such as the number of bedrooms, may differ across geographical areas. Thus, we also completed regressions comparing homeownership center performance without using property characteristics such as the number of bedrooms, initial REO valuation, year built, and the number of units as explanatory variables. These regression results were qualitatively similar to the results of regressions that included those variables. [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 website [hyperlink, http://www.gao.gov]. Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, 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 website, [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. Connect with GAO: Connect with GAO on facebook, flickr, twitter, and YouTube. Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts. Visit GAO on the web at [hyperlink, http://www.gao.gov]. To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Website: [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: Katherine Siggerud, Managing Director, siggerudk@gao.gov: (202) 512-4400: U.S. Government Accountability Office: 441 G Street NW, Room 7125: Washington, DC 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, DC 20548. [End of document]