This is the accessible text file for GAO report number GAO-08-41 entitled 'U.S. Postal Service Facilities: Improvements in Data Would Strengthen Maintenance and Alignment of Access to Retail Services' which was released on December 10, 2007. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to Congressional Requesters: United States Government Accountability Office: GAO: December 2007: U.S. Postal Service Facilities: Improvements in Data Would Strengthen Maintenance and Alignment of Access to Retail Services: GAO-08-41: GAO Highlights: Highlights of GAO-08-41, a report to congressional requesters. Why GAO Did This Study: Continued financial challenges and increased competition call for the U.S. Postal Service to manage its 34,000 facilities as efficiently and cost-effectively as possible. GAO and others have identified key facility management challenges, including the need to (1) capture and maintain accurate facility data, (2) adequately maintain facilities, and (3) align retail access with customer needs. This report assesses Postal Service efforts to overcome these challenges and implement leading federal practices. To conduct this study, GAO analyzed postal data and documents, visited 58 facilities, and interviewed postal officials. What GAO Found: To address the challenge of capturing and maintaining accurate facility management data, the Postal Service developed the Facility Database, but the database does not conform to the Postal Service’s goals or to leading federal practices; specifically, it does not include data needed to measure performance on managing facilities or have the capacity to track such data over time. Further, a database analysis by GAO revealed data reliability problems, including duplicative and contradictory data. In addition, major Postal Service departments do not use the database as a consolidated data source for managing postal facilities. The Postal Service has attempted to improve the database, but many problems remain. To address the challenge of maintaining its facilities, the Postal Service has begun assessing the condition of the facilities but has neither determined the extent of its maintenance projects nor strategically prioritized the projects. A Postal Service inspection of 651 randomly selected postal facilities revealed that two-thirds were in less than “acceptable” condition, but the Postal Service had not documented the full extent of its maintenance projects backlog. After the inspection, the Postal Service initiated a program to assess the condition of all of its facilities—a necessary first step to improving their condition. In addition, the Postal Service lacks the data needed to implement leading federal practices, such as considering a facility’s importance and value when prioritizing its maintenance projects. Due to funding constraints, the Postal Service currently focuses exclusively on emergency and urgent repairs—at the expense of a less costly preventive maintenance approach. To address the challenge of aligning access to postal retail services with customer needs, the Postal Service has expanded access in underserved areas but has done less to address overserved areas. Leading federal practices identify criteria for “rightsizing” facility networks—such as considering facilities’ importance and utilization—but the Postal Service does not consider these criteria. GAO’s analysis shows wide variation in the number of postal retail facilities among comparable counties, and a number of facilities GAO visited appeared to merit consideration for closure based on one or more of the federal criteria. If the Postal Service begins collecting data that reflects criteria based on leading federal practices, it may be able to close facilities and adjust access to retail services according to customer needs. What GAO Recommends: To improve facility management, GAO recommends that the Postal Service consider whether it is more cost-effective to make its Facility Database reliable or to replace it. If the Postal Service chooses to retain the database, GAO recommends that it establish internal controls, measure facility management performance, and track trends. GAO also recommends better prioritization of maintenance projects and initiating a criteria-based approach to assist in identifying and closing unneeded retail facilities consistent with leading federal practices. In its comments, the Postal Service chose to retain the database and establish controls but not to track performance or trends. It agreed in principle to prioritize maintenance but not to initiate a criteria-based approach to identify and close unneeded retail facilities. To view the full product, including the scope and methodology, click on [hyperlink, http://www.GAO-08-41]. For more information, contact Katherine A. Siggerud at (202) 512-2834 or siggerudk@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: FDB Does Not Meet the Postal Service's Goal or Leading Federal Practices for Capturing and Maintaining Accurate Facility Management Data: The Postal Service Has Initiated Actions to Assess the Condition of Its Facilities but Lacks a Strategic Approach for Prioritizing Maintenance Projects: The Postal Service Has Expanded Alternative Access to Its Services but Lacks Performance Data to Identify Potentially Unneeded Retail Facilities for Closure: Conclusions: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Executive Order 13327: Federal Real Property Asset Management: Appendix II: Objectives, Scope, and Methodology: Appendix III: Comments from the Postal Service: GAO Comments: Appendix IV: GAO Contact and Staff Acknowledgments: Tables: Table 1: Facility Database Entries for the Lubbock (Texas) Main Post Office as of July 2007: Table 2: Postal Areas, Districts, and Cities Visited: Table 3: Regression Coefficients Used to Establish the Average Number of Retail Postal Facilities in Counties Based on Each County's Population, Land Area, and Degree of Urbanization: Figures: Figure 1: Types of Facilities Operated by the Postal Service: Figure 2: Postal Service Area and District Boundaries: Figure 3: Vacant, Possibly Leasable Space in Postal Service Facilities Not Listed in the Facility Database: Figure 4: Sources of Facility Information in the Postal Service's 2006 Annual Report: Figure 5: Illustration of the Postal Service's Three-Part Facility Condition Assessment Program: Figure 6: Examples of Maintenance Issues at Postal Facilities We Visited: Figure 7: Pictures Illustrating the Size, Condition, and Utilization of the Downtown Fort Worth Station, Texas, as of February 2007: Figure 8: Distribution of Owned, Leased, and Contracted Postal Retail Facilities by County Based on Population, Land Area, and Degree of Urbanization: Figure 9: Examples of Privately Owned Businesses That Contract to Provide Retail Postal Services: Figure 10: Automated Postal Center with Signage Advertising Its Around- the-Clock Availability: Figure 11: Examples of Damage Resulting from a Partial Roof Collapse in December 2005 That Caused an Emergency Suspension of Operations at the Central Station in McKeesport, Pennsylvania: Figure 12: The "Store of the Future" Located behind Security at the Pittsburgh International Airport, Pennsylvania: Abbreviation: FDB: Facility Database: Federal Council: Federal Real Property Council: Postal OIG: U.S. Postal Service Office of the Inspector General: [End of section] United States Government Accountability Office: Washington, DC 20548: December 10, 2007: The Honorable Susan M. Collins: Ranking Member: Committee on Homeland Security and Governmental Affairs: United States Senate: The Honorable Thomas R. Carper: Chairman, Subcommittee on Federal Financial Management, Government Information, Federal Services, and International Security: Committee on Homeland Security and Governmental Affairs: United States Senate: With about 34,000 owned and leased facilities, the U.S. Postal Service is the third-largest landholding entity in the federal government. Most of its facilities provide retail services, such as mail and package delivery, consistent with its mission to make postal services accessible to everyone in the country. Other facilities house the Postal Service's mail processing, mail delivery, vehicle maintenance, and administrative functions. This expansive facility network is costly to operate and poses significant property management challenges. Changes in mail volumes, increasing employee compensation and benefits costs, a more competitive marketplace, and legislative reforms enacted in December 2006 have necessitated that the Postal Service increase its efficiency and cut costs. The Postal Accountability and Enhancement Act, enacted in December 2006,[Footnote 1] provided tools and mechanisms that can be used to establish an efficient, flexible, and financially sound Postal Service, but also introduced a rate cap for many postal services. Traditionally, the Postal Service set rates based on costs, but now an annual limitation will be placed on percentage changes in rates for market-dominant (as opposed to competitive) products, linked to the change in the Consumer Price Index--All Urban Customers.[Footnote 2] Real property management is a challenge not only for the Postal Service but also for other federal agencies. In 2004, an executive order established the Federal Real Property Council (Federal Council)--an interagency group of federal landholding agencies--to collect and disseminate information on leading practices for managing federal facilities. One such leading federal practice requires federal agencies bound by the executive order to establish a current and reliable inventory of their facilities. Furthermore, these agencies are required to capture and report performance-related data about their facilities in four areas: (1) importance to achieving the agency's mission, (2) utilization rate (extent to which the facility is fully utilized), (3) physical condition, and (4) annual operating costs. These data, termed "facility management performance measures," are used to measure an agency's performance in managing its facilities. Although not bound by the executive order,[Footnote 3] the Postal Service has voluntarily assigned a representative to the Federal Council and adopted some leading federal practices. Additional information on the executive order and the Federal Council is available in appendix I. Previous analyses by GAO and others have identified a number of key challenges to managing federal facilities, including a lack of reliable and useful data on facilities, facilities in disrepair, and facility networks that are not properly aligned with customer needs.[Footnote 4] You requested that we analyze these challenges as they apply to the Postal Service. Accordingly, this report addresses the Postal Service's efforts to (1) capture and maintain accurate facility data, (2) adequately maintain the condition of postal facilities, and (3) align access to retail services with its customer needs. This report identifies the Postal Service's goals and actions for managing each of these challenges and assesses the Postal Service's progress in overcoming the challenges and, as applicable, its implementation of leading federal practices. To accomplish our objectives, we visited 58 postal facilities, interviewed postal officials to learn about postal facility management, reviewed relevant postal documents, and assessed the Postal Service's implementation of relevant leading federal practices for managing its facilities. The observations derived from our site visits cannot be generalized to the population of postal facilities nationwide. We chose the facilities we visited to achieve balance with regard to geographic location, area demographic trends, facility type, revenue, and physical condition. Our work also included the following: * To assess the Postal Service's progress in overcoming the challenge of capturing and maintaining accurate data on its facilities, we analyzed the Postal Service's Facility Database (FDB) and assessed the reliability of the data it contained on two dates--October 5, 2006 and July 7, 2007. To assess FDB data, we interviewed postal officials to obtain an understanding of the data, the structure of the database, the sources of the data, and known database issues and limitations; reviewed related Postal Service documents and reports by the Postal Service's Office of the Inspector General (Postal OIG); and electronically tested the database for completeness, obvious errors, and inconsistencies. In addition, we attempted to verify selected facility data as of October 5, 2006, during our visits to 58 postal facilities. Our assessment of FDB data as of July 7, 2007--our second assessment--was designed, in part, to determine the impact of the Postal Service's efforts to improve FDB. * To assess the Postal Service's progress in overcoming the challenge of maintaining its facilities in adequate condition, we analyzed maintenance data nationwide and attempted to validate maintenance data at the 58 facilities that we visited. * To assess the Postal Service's progress in overcoming the challenge of aligning access to its retail services with customer needs, we assessed the extent to which the number of retail postal facilities varies among counties with similar attributes. To do so, we first identified the number and location of Postal Service retail facilities at the county level. Then, we statistically compared the number of postal retail facilities in each county, given the county's population, land area, and degree of urbanization, with the number of retail facilities in comparable counties. The variation in the number of postal retail facilities was determined by subtracting the number of retail facilities in a county with the number of facilities in comparable counties, as determined by the statistical comparison. However, we excluded counties without an urban center of at least 10,000 people from the output of our analysis because the Postal Service may need to maintain more postal retail facilities in these counties to fulfill its mission of providing universal access to postal services. Our statistical analysis explains almost 80 percent of the variation in the number of retail postal facilities among the counties. We conducted our work from July 2006 through December 2007 in accordance with generally accepted government auditing standards. Appendix II contains additional information on our scope and methodology. Results in Brief: The Postal Service developed the Facility Database (FDB) in 2003 to address the challenge of capturing and maintaining facility data, but it has not accomplished this goal. The FDB data that we assessed are not reliable, and FDB is not used as a centralized source for facility data as planned. We found numerous data reliability problems, including duplicate facility entries, multiple facility entries with the same function at the same address, and inconsistent information on square footage and ownership. We determined that these problems had three basic causes. First, the two primary Postal Service systems that feed data directly into FDB--the Address Management and the Facility Management Systems--contain inaccurate information. Second, data from both of these systems were incorrectly linked, creating incorrect data in FDB. For example, we found a post office that had four different entries in FDB because the facility entry was incorrectly linked to multiple facility entries in the Facility Management System that were, in reality, located elsewhere. This error, and others, occurred because FDB does not have internal controls, such as edit checks, to preclude obvious mistakes, and a planned effort to correct these linking errors has not yet begun. Third, we found that local employees often make mistakes when they enter facility data directly into FDB. Major departments within the Postal Service do not use FDB data to manage postal facilities, in part, because of reliability problems. The Postal Service has attempted to improve the reliability of FDB data by requiring local postal employees to validate FDB data for each facility, planning to correct incorrectly linked facility entries and eventually automate the linking process, training FDB users, and issuing a new FDB user's guide, but data reliability problems remain. Moreover, even if the data in FDB were reliable, FDB would not meet leading federal practices for facility data because FDB does not contain needed data fields for tracking facility management performance and does not archive data at regular intervals, as is necessary for tracking trends. Specifically, the Postal Service does not track the four performance measures recommended by the Federal Council--facility importance, utilization rate, condition, and annual operating costs--to assist agencies in strategically managing their facility networks. We are recommending that the Postal Service consider whether FDB is useful as a consolidated facility data source and, if so, improve its reliability and usefulness by establishing internal controls, such as edit checks. In addition, we recommend that the Postal Service conform to leading federal practices by measuring facility management performance and tracking trends. To address the challenge of adequately maintaining its facilities, the Postal Service has initiated actions to better understand the facilities' condition but does not know the magnitude of its maintenance backlog or how to strategically prioritize its maintenance projects. Postal Service officials said the Postal Service has historically underfunded its maintenance needs, resulting in the deterioration of its facilities. A 2005 contractor assessment of 651 randomly selected postal facilities revealed that two-thirds of these facilities were in less than "acceptable" condition, including 22 percent that were rated in "poor" condition. However, the Postal Service does not know the magnitude of its maintenance challenge because it has not yet fully assessed the condition of its network of about 34,000 facilities. As a first step toward learning more about the condition of all postal facilities, locally based Postal Service employees completed self-assessments of over 29,000 facilities in 2007. In addition, over the next 3 years, the Postal Service plans to conduct more comprehensive assessments of its larger facilities. Numerous Postal Service officials told us that insufficient funding has caused the Postal Service to focus exclusively on reactive maintenance--that is, "emergency" and "urgent" repairs--at the expense of routine maintenance to prevent problems. A reactive approach is ultimately more expensive, in part, because it shortens the useful life of equipment and facilities and necessitates more costly future repairs. Consistent with leading federal practices, the Postal Service recently recognized the need for a preventive approach to maintaining its facilities, including regular assessments of their condition. However, the Postal Service has not adopted leading federal practices by systematically incorporating facility management performance measures--facility importance, utilization rate, condition, and annual operating costs-- into its maintenance prioritization process. Because the facility assessments will provide the Postal Service with more complete information on its facility needs, we are recommending the Postal Service prioritize maintenance projects based on a facility's overall performance. To address its challenge of aligning access to retail services with customer needs, the Postal Service has taken steps to expand access to its services in underserved areas but has not achieved a goal established in its 2002 Transformation Plan (a plan that identified steps to guide future Postal Service operations) of proactively identifying unneeded retail facilities in overserved areas consistent with leading federal practices. The persistence of this alignment challenge is demonstrated by our analysis of the location of postal facilities nationwide, which shows wide variation in the number of postal retail facilities between counties of similar population, land area, and degree of urbanization. Specifically, our analysis shows that many counties with at least one urban center with a population of at least 10,000 people have far fewer or far more postal retail facilities than the average for comparable counties. To address this challenge, the Postal Service has expanded access to its retail services in underserved areas by establishing retail alternatives and setting goals for the use of those alternatives, such as the sale of postal services through privately operated contract facilities, the Internet, and automated postal centers. However, the Postal Service does not proactively identify and close unneeded retail facilities in overserved areas. Postal officials said the Postal Service is not pursuing its 2002 Transformation Plan goal of identifying and closing low-value, redundant postal retail facilities because of legal and political restrictions[Footnote 5] on closing post offices. Instead, the Postal Service only considers closing postal facilities under emergency circumstances, such as a forced evacuation due to a natural disaster.[Footnote 6] The Postal Service could better identify potentially unneeded retail facilities for closure consistent with its goals and leading federal practices by assessing each facility's importance to the Postal Service's mission in conjunction with the facility's utilization rate, condition, and annual operating costs. During our site visits, we identified a number of facilities that did not appear critical to the Postal Service's mission and that also merited consideration for closure based on the other leading performance measures for federal facilities. We are recommending that the Postal Service institute a more proactive, criteria-based approach to assist in identifying and closing unneeded retail facilities. In its written comments on a draft of this report, the Postal Service agreed with our two recommendations regarding FDB; agreed, in principle, with our recommendation to prioritize maintenance projects based on facility performance; but disagreed with our three remaining recommendations. In agreeing with our two recommendations regarding FDB, the Postal Service indicated that it (1) had already decided to retain FDB rather than to replace it with a new, more reliable database and (2) would establish additional controls to improve its reliability. Furthermore, although the Postal Service stated that it prioritizes maintenance projects adequately, it agreed, in principle, with our recommendation related to the need to prioritize maintenance projects based on a facility's overall performance. The Postal Service disagreed with our two recommendations related to adopting leading federal practices for measuring facility management performance and tracking performance trends over time, noting that it had concerns about adopting these practices due to the challenges created by its unique mandate to provide universal mail service on an almost daily basis and its provision of services and products that are in direct competition with the private sector. We revised our recommendations to emphasize that the Postal Service should develop performance measures that may better meet its needs as long as the measures are consistent with the spirit of those developed by the Federal Council. Finally, the Postal Service disagreed with our recommendation to institute a criteria-based approach to identify and close unneeded retail facilities as part of the congressionally required June 2008 facility plan in part because of the difficulty in establishing criteria that could be applied to all retail locations. Recognizing that closure decisions involve case-by- case considerations, we clarified our recommendation to indicate that the results of a criteria-based approach would "assist" the Postal Service in identifying candidate retail facilities for possible closure. The Postal Service's comments are reprinted in appendix III. In separate correspondence, the Postal Service also provided minor technical comments, which we incorporated, as appropriate. Background: To provide, consistent with its mission, postal services to everyone in the United States, the Postal Service operates an expansive network of facilities throughout the nation. Our analysis of Postal Service data shows there are approximately 34,000 active owned or leased postal facilities, most of which provide retail services. The Postal Service 2006 annual report indicates that its facilities are valued at approximately $21 billion. In addition, according to the Postal Service, it paid approximately $860 million for capital projects and maintenance repairs in fiscal year 2006. About three-fourths of postal facilities are leased, and in fiscal year 2006, the Postal Service paid over $1 billion to lease these facilities. The Postal Service is responsible for maintaining all of its owned facilities and many of its leased facilities, but maintenance responsibilities vary with the specific lease and can change when a lease is renegotiated. Employees at postal retail facilities provide services related to First- Class Mail, Insured and Registered Mail, Parcel Post, Priority Mail, and other services, such as post office box rentals and money- order purchases. Some retail facilities also provide space for other functions, such as receiving and sorting mail for delivery. Retail facilities--whether owned or leased--fall into one of three categories: (1) main post offices, where local postmasters oversee retail operations in the geographic area; (2) postal stations located within a municipality's corporate limits; and (3) postal branches located outside a municipality's corporate limits. According to Postal Service data, main post offices account for almost 75 percent of all retail facilities, but in large communities there may be more stations and branches. The Postal Service also operates nonretail facilities, such as mail processing facilities, vehicle maintenance facilities, and administrative offices. Figure 1 provides information on the different types of postal facilities. Figure 1: Types of Facilities Operated by the Postal Service: [See PDF for image] This figure contains six photographs that depict the following information: Main post offices: Main post offices house a community’s postmaster and serve as the primary retail facility in a community. Postal stations and branches: Postal stations and branches provide additional access to retail services within a community. Personnel in these facilities report to the postmaster at the area’s main post office. Carrier annexes: Carrier annexes provide space for mail carriers to sort mail for their routes. Some main post offices, stations, and branches also provide space for carrier operations. Mail processing facilities: At mail processing facilities (including processing and distribution facilities, bulk mail centers, and air mail centers), postal personnel process mail prior to delivery. Vehicle maintenance facilities: At vehicle maintenance facilities, postal personnel service Postal Service vehicles. Administrative and other facilities: Administrative and other facilities provide office space for such purposes as administrative, management, customer service, and training. Source: GAO. [End of figure] Besides the retail facilities it owns or leases, the Postal Service reported in its 2006 annual report that it operates about 3,950 privately owned and operated facilities, known as either "contract postal units" or "community post offices," which provide retail postal services. Contract postal units are operated by nonpostal employees in privately operated businesses, such as convenience stores, grocery stores, greeting card stores, and pharmacies. Community post offices are contract postal units that are located in small communities and function as main post offices. The Postal Service reported that there were 3,014 contract postal units and 937 community post offices throughout the nation. The Postal Service has no responsibility for maintaining these privately operated retail facilities. Responsibility for managing postal facilities is distributed across the Postal Service. * Three Postal Service departments in headquarters share responsibility for managing data on postal facilities--(1) Delivery and Retail, (2) Facilities, and (3) Intelligent Mail and Address Quality. Each of these departments tracks different data on postal facilities depending on its needs. Delivery and Retail manages FDB, which was developed to consolidate information on all postal facilities. Facilities manages the Facility Management System, which is used to manage facility acquisitions and capital and expense projects. Intelligent Mail and Address Quality administers the Address Management System, which contains the addresses of about 160 million delivery locations nationwide, including postal facilities. Both of these systems feed information directly into FDB. * The Postal Service's Vice President for Facilities oversees the maintenance of facilities nationwide in conjunction with eight regionally based Facilities Service Offices,[Footnote 7] which manage the maintenance activities in the Postal Service's areas. * The Postal Service's Vice President for Delivery and Retail--in conjunction with the nine area vice presidents, 80 district managers, and almost 24,000 local postmasters throughout the country--has nationwide responsibility for aligning the Postal Service's retail network with customer needs (see fig. 2). The district managers are responsible for making recommendations to open or close a postal retail facility, while the Vice President for Delivery and Retail is responsible for acting on the recommendations. Figure 2: Postal Service Area and District Boundaries: [See PDF for image] This figure is a map of the United States depicting the following postal service areas and district boundaries: Northeast; New York Metro; Eastern; Capitol Metro; Southeast; Great Lakes; Southwest; Western; Pacific. Source: U.S. Postal Service (information); Map Resources (map). [End of figure] In 2001, we placed the Postal Service on our list of agencies and programs designated as high risk due to, among other factors, financial and operational challenges.[Footnote 8] The Postal Service responded by issuing a plan to "transform" its operations--the 2002 Transformation Plan. Specific strategies outlined in the plan called for the Postal Service to align its retail network with customer needs by promoting alternative access to retail services and closing "low-value, redundant" postal retail facilities in overserved areas. To obtain information needed to align access to retail services with customer needs, the Postal Service indicated that it would establish a national facility database and develop a criteria-based methodology to determine which facilities to close. The criteria were to include factors, such as a facility's proximity to other postal facilities, the number of households and delivery points in a community, and indicators of retail productivity. In 2004, we reported that the Postal Service had not yet developed criteria for making changes to its retail facility network, including facility closures and consolidations.[Footnote 9] The Postal Service updated its Transformation Plan in 2005 for fiscal years 2006 through 2010, indicating that it intended to reduce its long-term facility repair costs by doing more focused, routine facility assessments and preventive maintenance. The postal reform legislation enacted in December 2006 provided additional opportunities to address the challenges the Postal Service faces in adapting to an increasingly competitive environment. The act provides tools and mechanisms to help control costs, including the costs of aligning its facility network with its customer needs. The act also requires the Postal Service to develop a plan by June 2008 to "rationalize" its network of facilities, remove excess capacity from the network, and identify anticipated cost savings and other benefits associated with network rationalization. In 2007, we removed the Postal Service's high-risk designation because the Postal Service addressed several concerns we raised when we originally placed the Postal Service on the high-risk list in 2001 and because of the passage of postal reform legislation. FDB Does Not Meet the Postal Service's Goal or Leading Federal Practices for Capturing and Maintaining Accurate Facility Management Data: Developed in 2003, FDB has not achieved the Postal Service's anticipated goal or conformed to a leading federal practice that calls for a consolidated source of accurate facility data because the data entered into FDB are not reliable. As a result, the Postal Service indicated that several major Postal Service departments do not use FDB for aggregate facility information, partly because of concerns about its reliability. These concerns result from, among other things, inaccurate data entered into the systems that feed into FDB, problems with how the systems are linked to form an FDB facility entry, and mistakes in entering data directly into FDB. The Postal Service has taken steps to improve the database, but systemic problems remain. In addition, FDB does not meet leading federal practices for tracking facility management performance and trends because FDB does not include facility management performance measures or provide for tracking trends over time. FDB Does Not Meet the Postal Service's Goal Because the Data Are Not Reliable: While the Postal Service created FDB in 2003 to achieve its goal for a consolidated source of accurate facility data, it has not achieved that goal due to data reliability problems. According to the Postal Service, it established FDB because its prior use of multiple databases caused confusion, data inaccuracies, problems in decision making, and higher costs. However, based on our analysis of selected data fields, FDB data problems make it unreliable as a centralized source of aggregate facility data even after performing data-cleaning techniques on the raw data. Specifically, our assessment revealed the following nationwide problems with the data's reliability as of October 5, 2006:[Footnote 10] * 145 facility entries were exact duplicates of another facility entry (all data fields were the same); * 1,931 facility entries had multiple retail facilities listed at the same address (e.g., a main post office and a station were listed with the same address); * 1,288 facility entries had different amounts of square footage listed for the facility;[Footnote 11] * 892 facility entries had conflicting information on whether the facility is owned or leased;[Footnote 12] * 1,216 facility entries had conflicting data on the amount of the facility's rent; and: * 509 facility entries listed as having staffed more retail windows than reportedly exist at the facility. The Lubbock (Texas) Main Post Office illustrates several of the data reliability problems we found. Specifically, in FDB data reported as of July 2007, the main post office is listed four times--each with different (1) square footage amounts, (2) ownership information, and (3) lease payment amounts. Except for information on the annual rent-- which the Postal Service considers sensitive--table 1 displays these FDB fields for the Lubbock Main Post Office. Table 1: Facility Database Entries for the Lubbock (Texas) Main Post Office as of July 2007: Name: 1; Lubbock; Subtype: Main Post Office; Address: 1515 Crickets Ave., Lubbock, TX 79402-9998; Interior square footage: 85,257; Owned or leased: Owned. Name: 2; Lubbock; Subtype: Main Post Office; Address: 1515 Crickets Ave., Lubbock, TX 79402-9998; Interior square footage: 7,896; Owned or leased: Leased. Name: 3; Lubbock; Subtype: Main Post Office; Address: 1515 Crickets Ave., Lubbock, TX 79402-9998; Interior square footage: 5,600; Owned or leased: Leased. Name: 4; Lubbock; Subtype: Main Post Office; Address: 1515 Crickets Ave., Lubbock, TX 79402-9998; Interior square footage: 50; Owned or leased: Leased. Source: U.S. Postal Service FDB. [End of table] We discussed several specific examples of the data reliability problems we found with postal officials who, after researching multiple data sources, provided explanations for the specific examples we found. For example, with respect to the Lubbock Main Post Office, postal officials said that the first entry in table 1 is correct while the other three entries actually represent other postal facilities at different locations in Lubbock. In addition, using another data source, they told us that entry 2 is actually a carrier annex; entry 3 is a warehouse; and entry 4 is a postal vending machine. Using other data sources, the Postal Service officials also provided explanations for other specific problems we identified. However, without additional site visits, we cannot determine whether the explanations provided were accurate.[Footnote 13] We also identified incorrect FDB information for 24 of the 58 postal facilities we visited. For example, during our sites visits, we found that information on vacant leasable square footage, which the Postal Service asks local employees to document, was inaccurate in FDB. At least six of the facilities we visited had vacant space that local employees said could be leased, but these facilities were not listed as having vacant, leasable space in FDB. These facilities and their vacant space are shown in figure 3. Postal officials acknowledged our examples and noted there are few incentives for local officials to report facilities' vacant, leasable space in FDB. Figure 3: Vacant, Possibly Leasable Space in Postal Service Facilities Not Listed in the Facility Database: [See PDF for image] This figure contains six photographs that depict vacant, possibly leasable space in Postal Service Facilities not listed in the facility database: Circle City Station, Indianapolis, Indiana: * Vacant area: A large portion of the second floor; * Status: Postal officials said the Postal Service never built out the second floor because the space was not needed and could be subleased or returned to building owner; * Status listed in FDB: No vacant leasable space. Denton Main Post Office, Texas: * Vacant area: Entire second floor of the large post office; * Status: Postal officials said half of the building was occupied by other federal agencies that moved out about 10 years ago and that the space could be leased; * Status listed in FDB: No vacant leasable space. Downtown Finance Station, Gary, Indiana: * Vacant area: The basement (pictured) is completely vacant, and the second floor is used once per month or less for training; * Status: Postal officials said the Postal Service never used more than just the main floor and could lease the excess space; * Status listed in FDB: No vacant leasable space. Fort Worth Downtown Station, Texas: * Vacant area: Second floor (pictured) and basement are vacant. Third floor used periodically for storage and training; * Status: Postal officials said most of the building has been vacant since the mail processing function was removed years ago; * Status listed in FDB: No vacant leasable space. Richland Station, Dallas, Texas: * Vacant area: Much of the second floor of this 53,000-square-foot post office; * Status: Postal officials said the office space has been vacant for years, and another portion (pictured above) has not been occupied since the Postal Service purchased the building in 1989; * Status listed in FDB: No vacant leasable space. East Chicago Main Post Office, Indiana: * Vacant area: The entire second floor, which consists of several offices; * Status: Postal officials said it has been vacant for years and could be leased; * Status listed in FDB: No vacant leasable space. Source: GAO. [End of figure] Postal officials acknowledged that the Postal Service has not analyzed the reliability of FDB data but expressed confidence that the problems we found affect a small percentage of the Postal Service's facilities. While we cannot definitively determine the overall magnitude of FDB's data reliability problems, the cumulative effect of the problems we found could significantly distort the reporting of aggregate facility statistics. For example, conflicting ownership data in FDB, as illustrated for the Lubbock facility in table 1, could cause the Postal Service's annual rent obligation in FDB to vary by as much as $82 million, or more than 8 percent of the Postal Service's reported rent obligation for fiscal year 2006. Inaccurate ownership data are of particular concern because the Postal Service used FDB for its aggregate ownership statistics in its 2006 annual report. Postal Officials Do Not Rely Exclusively on FDB for Reporting on the Postal Service's Facilities: Although the Postal Service developed FDB over 5 years ago to provide a consolidated source of facility data, the Postal Service continues to operate and use various facility data sources. Furthermore, while postal officials acknowledged that postal staff did not use FDB initially because the data were not reliable, our analysis demonstrates that FDB is still not sufficiently reliable for use as a consolidated data source for postal facilities. The Postal Service indicated that several major Postal Service departments do not use FDB for aggregate facility information, partly because of concerns about its reliability. Thus, as shown in figure 4, instead of exclusively relying on FDB for facility information in its 2006 annual report, the Postal Service used multiple sources of data, including its Address Management System, for quantifying its retail and delivery facilities by type. Figure 4: Sources of Facility Information in the Postal Service's 2006 Annual Report: [See PDF for image] This figure is an image of a portion of the Postal Service's 2006 Annual Report, showing sources of information. The following data is depicted: Financial Section Part I: Available Information: The United States Postal Service is not subject to the informational requirements of the Securities Exchange Act of 1934 (Exchange Act). We do not file periodic reports, proxy statements or other information with the SEC. Financial and other information can be accessed on the "About USPS and News" section of our website at [hyperlink, http://www.USPS.com]. Information on our website is not incorporated by reference in this document. We make available, free of charge, copies of our annual report, as soon as reasonably practicable after it is produced. Requests for copies should be sent to: United States Postal Service: Public Affairs and Communications: 475 L'Enfant Plaza, SW: Washington, DC 20260-3100: Item 2 — Properties: Real Estate: Our facilities range in size from 50 square feet to 34 acres under one roof, and support retail, delivery, mail processing, maintenance, administrative and support activities. Real Estate Inventory (Actual numbers): Leased Facilities, 2006: 25,567; Leased Facilities, 2005: 25,772. Real Estate Inventory (Actual numbers): Owned Facilities, 2006: 8,437; Owned Facilities, 2005: 8,399. Real Estate Inventory (Actual numbers): GSA/Other Government Facilities, 2006: 408; GSA/Other Government Facilities, 2005: 417. Real Estate Inventory (Actual numbers): Total Real Estate Inventory, 2006: 34,412; Total Real Estate Inventory, 2005: 34,588. Real Estate Inventory (Actual numbers): Annual Rent Paid to Lessors (dollars in millions), 2006: $1,002; Annual Rent Paid to Lessors (dollars in millions), 2005: $934. Source: Facility Database (as fed from the Facility Management System): The majority of our small and medium sized facilities support the retail and delivery operations located in virtually every community across this country. Our retail and delivery operations are supported by 32,875 leased or owned facilities. We also provide retail services through 3,951 Contract Postal Units and Community Post Offices where the facility is owned and maintained by the contractor. Retail and Delivery Facilities (Actual numbers): Post Offices, 2006: 27,318; Post Offices, 2005: 27,385. Retail and Delivery Facilities (Actual numbers): Classified Branches, 2006: 1,522; Classified Branches, 2005: 1,540. Retail and Delivery Facilities (Actual numbers): Classified Stations, 2006: 3,457; Classified Stations, 2005: 3,498. Retail and Delivery Facilities (Actual numbers): Carrier Annexes, 2006: 578; Carrier Annexes, 2005: 584. Retail and Delivery Facilities (Actual numbers): Contract Postal Units, 2006: 3,014; Contract Postal Units, 2005: 3,116. Retail and Delivery Facilities (Actual numbers): Community Post Offices, 2006: 937; Community Post Offices, 2005: 1,019. Retail and Delivery Facilities (Actual numbers): Total Retail and Delivery Facilities, 2006: 36,826; Total Retail and Delivery Facilities, 2005: 37,142. Source: Address Management System. Our larger facilities typically support mail processing operations which move millions of pieces of mail across the country on a daily basis. Processing Facilities, (Actual numbers): Processing and Distribution Centers, 2006: 269; Processing and Distribution Centers, 2005: 269. Processing Facilities, (Actual numbers): Customer Service Facilities, 2006: 195; Customer Service Facilities, 2005: 195. Processing Facilities, (Actual numbers): Bulk Mail Centers, 2006: 21; Bulk Mail Centers, 2005: 21. Processing Facilities, (Actual numbers): Priority Mail Processing Centers, 2006: 11; Priority Mail Processing Centers, 2005: 11. Processing Facilities, (Actual numbers): Annexes, 2006: 66; Annexes, 2005: 66. Processing Facilities, (Actual numbers): Surface Transfer Centers, 2006: 17; Surface Transfer Centers, 2006: 14. Processing Facilities, (Actual numbers): Airmail Processing Centers, 2006: 77; Airmail Processing Centers, 2005: 79. Processing Facilities, (Actual numbers): Remote Encoding Centers, 2006: 12; Remote Encoding Centers, 2005: 15. Processing Facilities, (Actual numbers): International Service Centers, 2006: 5; International Service Centers, 2005: 5. Processing Facilities, (Actual numbers): Total Processing Facilities, 2006: 673; Total Processing Facilities, 2005: 675; Source: Physical count of facilities. We also have approximately 1,000 other types of facilities including administrative, vehicle maintenance and miscellaneous support facilities. Sources: U.S. Postal Service (information); GAO (analysis). [End of figure] FDB Reliability Problems Are Caused by Inaccuracies in Feeder Applications, Application Linkage Problems, and Data Entry Mistakes: We found that FDB's data reliability problems were caused by (1) errors in the systems that feed into FDB, (2) problems with how the systems are linked to form an FDB facility entry, and (3) errors in inputting FDB data. First, as stated earlier, some information is fed directly from both the Address Management and Facility Management systems into FDB, causing errors in either system to automatically feed into FDB. While we did not fully evaluate the Address Management System--the Postal Service's database for all delivery points in the country--we found errors in the Address Management System data that the Postal Service used in its 2006 annual report. Specifically, we found instances of (1) duplicate entries for the same facility; (2) multiple facilities with the same function (e.g., main post office) listed at the same address; and (3) contractor and Postal Service-operated facilities listed at the same address. During our site visits, we also found that the address contained in the Address Management System and fed into FDB for the O'Hare Terminal 2 Finance Station in Chicago, Illinois, was incorrect. In addition to containing errors, in December 2006, the Postal OIG reported that the Address Management System was incomplete because it did not contain information on all postal facilities.[Footnote 14] During our site visits, we also found errors that were fed directly into FDB from the Facility Management System. For example, the square footage in FDB for the Colleyville Main Post Office in Texas was incorrect because it did not reflect the sale of a significant portion of land that occurred 3 years prior to the entry. Postal Service officials acknowledge that errors in the Address Management and Facility Management systems feed directly into FDB. Second, many of the FDB errors we found resulted when facility entries in the Address Management and Facility Management Systems were incorrectly linked to create an FDB facility entry. FDB facility entries are created by linking facility entries in the Address Management and Facility Management Systems. However, these systems do not use the same convention for naming facilities, and therefore, they cannot be linked automatically to create an FDB facility entry. Instead, according to Postal Service officials, postal employees must manually link facility entries in the two systems. In some instances, the manual process resulted in linkage errors which, according to postal officials, caused some of the duplicate facility entries and contradictory information in FDB that we identified. For example, the Cumberland Main Post Office in Maryland, which we visited, had four entries in FDB--one for the Main Post Office and the other three for other postal facilities in Cumberland that were incorrectly linked to the main post office's address. The Postal Service corrected the FDB entry for this facility when we brought this problem to the Postal Service's attention, but the problem remains at other locations. For example, the Lubbock Main Post Office, which was discussed previously (see table 1), was incorrectly linked to multiple facility entries within the Facility Management System. The error occurred because FDB lacks internal controls, such as edit checks, to prohibit postal staff from linking an FDB facility to multiple facility entries in the Facility Management System. According to Postal Service officials, there were 762 facilities in FDB as of September 2007 that were mistakenly linked to multiple Facility Management System facility entries. Third, we found that some of the errors in FDB were caused by local employees entering incorrect information directly into FDB. Although information on a facility's address, size, and ownership is fed automatically into FDB from the Address Management and Facility Management Systems, other data are entered manually into FDB by local postal employees. For example, local employees enter the number of retail windows (e.g., sales areas) at each facility and the number of those windows that are typically staffed. Our analysis revealed that local officials make numerous mistakes entering this and other information into FDB. For example, 509 retail postal facilities are listed in FDB as staffing more retail windows than reportedly exist at the facility. Implementing an edit check in FDB would eliminate this type of reporting error. As shown in figure 3, FDB data on vacant, leasable space entered by local employees are also often in error. The Postal Service Has Taken Actions to Improve FDB Data, but Problems Remain: While postal officials are aware of errors in FDB and have taken several actions to improve the quality of the data, these actions have not yet corrected all of the problems we identified. First, in response to a 2006 recommendation by the Postal OIG,[Footnote 15] the Postal Service has started requiring local employees to validate and correct FDB data periodically for their facilities. While this validation process could, in our view, help identify and correct some of the errors we found, the Postal Service's validation completed in February 2007 was not entirely successful for a number of reasons. Specifically, mistakes that could have been corrected locally were often not corrected. For example, the retail postal facility data available as of May 2007, that had been validated by a postal employee on February 7, 2007, indicated the facility staffed twice as many retail windows as exist at the facility. We discussed this data discrepancy with Postal Service officials, and they corrected the facility's FDB entry. While this specific discrepancy was corrected, our assessment of FDB data as of July 7, 2007, identified 354 instances of this problem. However, even if local employees corrected these and other inaccuracies, problems would remain because local employees cannot correct errors in FDB fed directly from other systems, such as Facility Management System data on square footage and ownership status. Errors of this type can only be corrected by personnel administrating the Facility Management or Address Management Systems. Second, to avoid manual linking errors and to improve the accuracy of the linkage used to create FDB facility entries, postal officials are planning to automatically link the Address Management and Facility Management Systems' facility entries. According to the Postal Service, Address Management System administrators must first apply the standard facility naming convention used by the Facility Management System and thus create a unique identifier for linking facility entries in the Address Management and Facility Management Systems. Automation could help reduce the frequency of future linking mistakes, but the Postal Service must correct existing errors before automating the process. Postal officials had expected to begin correcting existing errors in October 2006; however, as of September 2007, the effort has not yet begun. Finally, the Postal Service responded to a Postal OIG recommendation [Footnote 16] for improving the quality and completeness of data entered into FDB by completing field training sessions for FDB users and issuing a new FDB user's guide in 2007. These actions are too recent to gauge their effectiveness. The Postal Service's Facility Data Do Not Meet Leading Federal Practices for Tracking Facility Management Performance and Trends: Even if the data in FDB were reliable, they would not meet leading federal practices for facility data because FDB (1) does not contain fields for the four performance measures recommended by the Federal Council--a facility's importance, utilization rate, condition, and annual operating costs--and (2) does not allow for tracking trends. The federal leading practices are intended to, among other things, help agencies measure their progress in managing their facilities and identify properties for disposal or investment. Postal Service officials said none of the Postal Service's facility databases, including FDB, were designed for these purposes. In addition, they noted the Postal Service is not bound by the executive order on federal real property asset management and, consequently, is not required to adopt leading federal practices, such as the implementation of performance measures. Even if the Postal Service collected data on its performance, it could not measure its performance over time because it does not retain or archive FDB data at regular intervals (e.g., annually). The Postal Service Has Initiated Actions to Assess the Condition of Its Facilities but Lacks a Strategic Approach for Prioritizing Maintenance Projects: The Postal Service has initiated actions to assess the condition of its facilities, but has not yet assessed the magnitude of its maintenance backlog or strategically prioritized its maintenance projects--a leading federal practice. According to postal officials, the Postal Service has historically underfunded its maintenance needs, resulting in the deterioration of its facilities. While there is some evidence that many postal facilities are in less than acceptable condition, the magnitude of the challenge is unknown. To learn more about the condition of all its facilities, the Postal Service has started implementing self-assessments conducted by local employees for small facilities and more intensive assessments for larger facilities. Numerous Postal Service officials told us that insufficient funding has caused the Postal Service to focus solely on urgent repairs--instead of routine, preventive maintenance--which could lead to more costly repairs over time. Consistent with a leading federal practice, agencies can maximize the value of maintenance funding by using facility management performance data to identify and prioritize their greatest maintenance needs, but the Postal Service cannot adopt this practice because it does not systematically capture the necessary data. The Postal Service Does Not Currently Know the Extent of Its Maintenance Needs but Has Initiated Actions to Assess the Condition of Its Facilities: The Postal Service has not comprehensively assessed the condition of its facilities, but the amount it has recently spent on facility maintenance--$712 million from fiscal year 2003 through fiscal year 2006--has been insufficient, postal officials said, to address its facility maintenance needs. Postal officials also said several years of underfunding have caused postal facilities to deteriorate and many are in need of repairs. Evidence supporting the officials' statements includes a 2005 assessment conducted by a contractor of 651 randomly selected owned and leased postal facilities. According to this assessment, two-thirds of the facilities were in less than "acceptable" condition, including 22 percent that were in "poor" condition.[Footnote 17] However, the Postal Service will not know the magnitude of the deterioration or the extent of its maintenance backlog until it fully assesses the condition of all its facilities. In 2007, to begin assessing the condition of its facilities, the Postal Service requested local employees to conduct self-assessments of their facilities. Local employees responded to the request by assessing over 29,000 facilities and identifying 73,500 maintenance needs estimated to cost $236 million. While these assessments provide additional information on the condition of postal facilities, they are not comprehensive because thousands of facilities were not assessed and the local employees are not formally trained to conduct facility assessments. Over the next 3 years, the Postal Service plans to conduct a more comprehensive, three-part program to assess the condition of all its facilities. The first part of the program involves facilities that are less than 6,500 square feet. For these facilities, the Postal Service plans to ask local employees to complete annual self-assessments similar to the ones completed in 2007. The second part of the program involves facilities with 6,500 to 100,000 square feet of interior space. For these facilities, the Postal Service plans to conduct more detailed condition assessments once every 3 years using contract building inspectors beginning in the summer of 2007.[Footnote 18] According to Postal Service officials, the Postal Service established the 6,500-square-foot threshold because the most complicated, important facilities are generally larger. The third part of the program involves the Postal Service's largest mail processing facilities. For these facilities, the Postal Service plans to use on-site postal maintenance staff to conduct annual physical assessments of the buildings and, once every 5 years, employ architectural and engineering firms to conduct more thorough assessments. According to the plans for the program, local Postal Service employees or facility inspectors will enter information collected from each assessment into a new database called the Infrastructure Condition Assessment Model, which will allow contract facility inspectors to estimate the urgency and cost of each identified repair project. The new database will feed into the Postal Service's existing maintenance tracking system and, according to postal officials, will be used to budget and prioritize urgent maintenance projects for repair. Figure 5 illustrates the Postal Service's three-part facility condition assessment program. Figure 5: Illustration of the Postal Service's Three-Part Facility Condition Assessment Program: [See PDF for image] This figure displays photographs of the three-part facility condition assessment program, and depicts the following data: Facilities with less than 6,500 square feet: * Annual self-assessment performed by local postal employees. Information is fed to the Infrastructure Condition Assessment Model database: Contract facility inspectors: * Review maintenance projects; * Estimate repair costs. Facility service office staff: * Review and prioritize maintenance projects; * Determine maintenance responsibility; * Track and oversee repairs. Facilities with 6,500 to 100,000 square feet: * Physical inspections by contract building inspectors; * Facilities to be inspected once every 3 years. Information is fed to the Infrastructure Condition Assessment Model database: Contract facility inspectors: * Review maintenance projects; * Estimate repair costs. Facility service office staff: * Review and prioritize maintenance projects; * Determine maintenance responsibility; * Track and oversee repairs. Largest mail processing facilities: * Annual inspections by postal maintenance staff; * Physical inspections by contracted architectural and engineering firms once every 5 years; Information is fed to the Infrastructure Condition Assessment Model database: Contract facility inspectors: * Review maintenance projects; * Estimate repair costs. Facility service office staff: * Review and prioritize maintenance projects; * Determine maintenance responsibility; * Track and oversee repairs. Source: U.S. Postal Service (information); GAO (analysis). [End of figure] The Postal Service Reacts to Emergency and Urgent Maintenance Issues but Recognizes the Need for More Preventive Maintenance: Postal Service officials with responsibility for facility maintenance at the national, area, and district levels said that the Postal Service has underfunded its maintenance for years and suspects that this underfunding has resulted in deteriorating facilities and a large maintenance backlog. Postal officials told us that this insufficient funding has caused the Postal Service to focus exclusively on reactive maintenance--that is, "emergency" and "urgent" repairs--at the expense of routine repairs. In addition, according to the Postal OIG, insufficient funding for repairs and maintenance may be hampering the Postal Service's ability to adopt a preventive maintenance approach.[Footnote 19] A 2000 Postal OIG report described the Dallas Downtown Station in Texas as deteriorated and attributed its deterioration to deferred maintenance that "increased the risk of injury to Postal Service employees and customers, and has compromised Postal Service property and the safety and security of the mail."[Footnote 20] The Postal OIG recommended immediate evacuation of the facility until needed repairs could be made. When we visited the Dallas Downtown Station in 2007, the Postal Service had repaired the facility at a cost of $12 million. The Postal Service may have avoided some of those costs if it had done more preventive maintenance. Other facilities we visited had not yet been repaired--including facilities with chronically leaking roofs and visible interior and exterior damage. Figure 6 illustrates maintenance issues we observed during our site visits: Figure 6: Examples of Maintenance Issues at Postal Facilities We Visited: [See PDF for image] This figure shows four photographs, with the following captions: Photo #1: Back door of the West Indianapolis Finance Station, Indianapolis, Indiana. Photo #2: Evidence of a leaking roof in Tollston Station, Gary, Indiana. Photo #3: Damaged ceiling tiles in the 15th Avenue Station, Gary, Indiana. Photo #4: Damaged floor in the Carnegie Main Post Office, Carnegie, Pennsylvania. Source: GAO. [End of figure] In its Strategic Transformation Plan for 2006 through 2010, the Postal Service established a goal of reducing its repair costs "through more focused routine building assessments and better planning to fix small problems as soon as possible."[Footnote 21] In a May 2007 report, the Postal OIG concurred with this approach, recommending that the Postal Service adopt more preventive maintenance practices by, among other things, regularly assessing postal facilities to identify repair needs and better leverage the Postal Service's limited financial resources.[Footnote 22] This approach is consistent with a recommendation made by the National Research Council of the National Academies[Footnote 23] in 1990 that suggested federal agencies regularly assess the condition of their facilities and do preventive maintenance to avoid costly future repairs.[Footnote 24] A reactive maintenance approach is ultimately more expensive, partly because it shortens the useful life of equipment and facilities and necessitates, among other things, more costly future repairs. To that point, a 2004 National Research Council study cited an estimate that each dollar in deferred maintenance results in a long-term liability of $4 to $5 for future repair costs.[Footnote 25] The Postal Service Has Not Established a Strategic Facility Maintenance Approach in Accordance with Leading Federal Practices: The Postal Service bases maintenance priorities on urgency. For example, roof issues take priority over nonstructural interior maintenance needs. While urgency is important for prioritizing maintenance spending, leading federal practices consider other important measures, such as a facility's (1) importance to an agency's mission, (2) utilization rate, (3) condition, and (4) annual operating costs. The Postal Service's three-part assessment program will provide data on the condition of its facilities, but the Postal Service will not be able to prioritize repairs strategically since it does not capture data on its facilities' importance, utilization rate, and annual operating costs to inform its maintenance decisions. For example, a Postal Service official who is responsible for managing maintenance throughout a large geographic area told us that he cannot consider a facility's importance to the Postal Service's mission when prioritizing maintenance projects because the Postal Service does not capture this information. Furthermore, the Postal Service does not currently know the replacement value of its facilities--essential information for evaluating a facility's overall condition. Adopting federal facility management performance measures would help the Postal Service establish a strategic approach to facility maintenance by allowing the agency to better identify its most important facilities, prioritize its maintenance needs, and allocate its maintenance funds accordingly. The consequences of not considering a facility's importance, utilization rate, and annual operating costs were evident at the Downtown Fort Worth Station in Texas, which we visited. The Postal Service spent about $1 million to repair it in fiscal year 2006 even though the station remains in deteriorating condition, is largely vacant, and does not appear critical to the Postal Service's mission since the remaining retail and carrier functions could be housed elsewhere in a smaller facility. Local postal officials said the Postal Service has considered disposing of the station for years but, instead, repaired it because no decision had been made on whether to retain it. Figure 7 illustrates the size, condition, and utilization of the Downtown Fort Worth Station at the time of our visit. Figure 7: Pictures Illustrating the Size, Condition, and Utilization of the Downtown Fort Worth Station, Texas, as of February 2007: [See PDF for image] This figure is a display of five photographs, with the following captions: Photo #1: The Downtown Fort Worth Station. The Postal Service primarily occupies only the first floor. Photo #2: Plywood, partially secured by a wood plank, covers a broken window. Photo #3: Vacant work floor space with obsolete processing equipment. Photo #4: Vacant work floor space. Photo #5: Obsolete mail sorting equipment fills an unused room. Source: GAO. [End of figure] The Postal Service Has Expanded Alternative Access to Its Services but Lacks Performance Data to Identify Potentially Unneeded Retail Facilities for Closure: To address the challenge of aligning retail access with customer needs, the Postal Service has expanded alternative access to its services in underserved areas but has done less to curtail services in overserved areas. Our analysis shows wide variation in the number of postal retail facilities among counties of similar population, land area, and degree of urbanization--demonstrating the Postal Service's challenge of placing facilities where they are needed. To address this challenge in underserved areas, the Postal Service has expanded access to retail services through alternative access options and has set goals for the use of these options. However, the Postal Service has not actively pursued a goal in its 2002 Transformation Plan to proactively identify and close unneeded retail facilities in overserved areas. This inaction does not conform to leading federal practices, which suggest that an agency consider closing facilities that are not critical to achieving the agency's mission, are in poor condition, are not fully utilized, or are costly to operate relative to their revenue. Number of Postal Retail Facilities Varies among Counties with Similar Characteristics: The number of facilities that provide postal retail services varies widely among counties of similar population, land area, and degree of urbanization, according to our statistical analysis of the distribution of postal retail facilities nationwide.[Footnote 26] Specifically, we developed a regression model to determine the average number of post offices for counties of comparable population, land area, and degree of urbanization. To determine the extent of variation among counties, we compared the actual number of post offices in each county with the averages derived from our regression analysis.[Footnote 27] While our analysis was not intended to consider all relevant factors, including retail sales volumes or the capacity of postal retail facilities, it does explain almost 80 percent of the variation in the number of retail postal facilities between counties. According to our analysis, some counties have far fewer or far more postal retail facilities than other counties of comparable population, land area, and degree of urbanization. For example, Hoke County, North Carolina, has 1 retail postal facility, whereas the average comparable county has 10 such facilities. Conversely, Fayette County, Pennsylvania, has 63 postal retail facilities--over 400 percent more retail facilities than the average county of comparable population, land area, and degree of urbanization. The wide variability in the number of retail postal facilities in comparable counties suggests that access to postal services among comparable counties also varies. Such variation is inconsistent with the Postal Service's Transformation Plan goal and the leading federal practice of aligning access to facilities with customer and service needs. Figure 8 shows the distribution of owned, leased, and contracted postal retail facilities by county based on population, land area, and degree of urbanization. Figure 8: Distribution of Owned, Leased, and Contracted Postal Retail Facilities by County Based on Population, Land Area, and Degree of Urbanization: [See PDF for image] This figure is a map of the United States indicating the distribution of owned, leased, and contracted Postal Retail Facilities by county based on population, land area, and degree of urbanization. Counties on the map are shaded to indicate their inclusion in one of the following categories: 1) Counties without an urban center of 10,000 people or more were excluded from this analysis; 2) Approximately the same number of postal retail facilities as comparable counties: 1,313. 3) Fewer postal retail facilities than comparable counties: 280[A]. 4) More postal retail facilities than comparable counties: 260[B]. Source: GAO (information); Map Resources (map). Note: Appendix II contains additional information on our scope and methodology. [A] Counties with "fewer" facilities than counties with comparable population, land area, and degree of urbanization have 9 fewer postal retail facilities on average than comparable counties, which represents the lowest 10 percent. [B] Counties with "more" facilities than comparable counties have 14 more postal retail facilities on average than comparable counties, which represents the highest 10 percent. [End of figure] The Postal Service Has Expanded Access to Alternative Retail Options in Underserved Areas: Aware that some communities, particularly in growing areas, have insufficient access to postal retail services, the Postal Service established a goal in 2002 for increasing access to postal services through alternative options. The Postal Service favors alternative retail options over building new postal-operated facilities because the Postal Service does not incur construction or operating costs when providing services through these alternative access options. Thus, in its 2006 annual report, the Postal Service indicated it had contracts with almost 4,000 private operators to provide access to postal retail services.[Footnote 28] While contract postal units are not always located in underserved areas, our analysis shows that their presence increased access to postal services in hundreds of counties nationwide. Without contract postal units, these counties would have had fewer retail facilities than the average for counties of comparable population, land area, and degree of urbanization. For example, Butler County, Ohio, had 16 retail facilities operated by the Postal Service- -7 fewer than the average for comparable counties. However, the Postal Service's agreement to operate seven contracted facilities there brought the total number of retail postal facilities in the county to 23, which was the average for comparable counties. Figure 9 shows a variety of examples of privately owned businesses that contract to provide postal services. Figure 9: Examples of Privately Owned Businesses That Contract to Provide Retail Postal Services: [See PDF for image] This figure is a display of four photographs depicting examples of privately owned businesses that contract to provide Retail Postal Services. The four businesses shown are: * Sweet Time Café and General Store; * Simpson Realty; * Denton Mail-n-More; * Shop-n-Save. Source: GAO. [End of figure] Other alternative retail access options include the Postal Service's: * stamps on consignment program, which allows businesses, such as drug stores and grocery stores, to purchase stamps from the Postal Service and retain a share of the proceeds; * package pickup and stamps sold online, which are available through the Postal Service's Web site; and: * automated postal centers, which provide access to most postal services from 2,500 centers located in postal-operated retail facilities; some automated postal centers are available 24 hours a day, 7 days a week, as indicated by the signage for the automated center shown in figure 10. Figure 10: Automated Postal Center with Signage Advertising Its Around- the-Clock Availability: [See PDF for image] This figure is a display of four photographs which depict various Automated Postal Centers with signage advertising its around-the-clock availability. Source: GAO. [End of figure] To improve its efforts to increase retail access for communities with insufficient access, the Postal Service is developing a model that will identify underserved areas--called the Model to Optimize Retail Effectiveness. According to Postal Service officials, the Postal Service developed the model in response to our 2004 recommendation that it develop criteria for making changes to its retail network[Footnote 29] and expects to finalize the model by the end of 2007. Postal Service officials also said the model will take several factors into consideration--including the relative location of competitors, costs, the extent of customer satisfaction, and population growth--in order to target areas that would benefit most from increased access to postal services. According to postal officials, FDB is an important source of information for this effort, which suggests FDB data reliability problems could adversely affect the model's output. Postal officials said the local Postal Service employees would verify the model's findings to ensure their accuracy. Once it identifies and verifies the underserved areas, the Postal Service could decide to, among other actions, increase facility hours, expand advertising of existing alternative access options, add automated postal centers, or contract with a private business to open a contract postal facility. While the Postal Service could also acquire new postal retail facilities, Postal Service officials said this is the least preferable option given the high costs of traditional postal facilities. To emphasize the importance of expanding alternative options for postal services, the Postal Service set a goal to accomplish at least 40 percent of its retail transactions through alternative access options by 2010. Postal Service's Actions to Align Retail Access in Overserved Areas Are Limited to Closing Some Vacated Facilities and Reducing Staff: From 2002 through the end of 2006, the Postal Service closed 795 facilities after placing them on emergency suspension. The Postal Service places a facility on emergency suspension when, among other reasons, severe maintenance problems create health or safety risks to employees and customers that require the Postal Service to vacate the facility. Other reasons for an emergency suspension include the retirement or resignation of a community's sole postal employee; a building owner's decision not to renew the Postal Service's lease; or a forced evacuation due to fire, flood, or other natural disaster. When a postal retail facility is placed on emergency suspension, the district has 90 days to decide whether to reopen, close, or consolidate the facility. However, the actual closure can take years. Specifically, 44 of the 159 facilities under emergency suspension that were slated for closure as of May 2007 have had their operations suspended for more than 5 years. In addition, it is not clear what criteria local managers apply when deciding whether to reopen or close a facility that is on emergency suspension. We visited one of these stations, the McKeesport Central Station in Pennsylvania, which had been on emergency suspension for over a year following a partial roof collapse in December 2005 (see fig. 11). The local postmaster told us the Postal Service planned to reopen the station once the landlord repaired the roof, even though the sole employee assigned to the station has been reassigned, customers have not complained about the lack of service, the station had low revenue, and the station is located approximately 1 mile from the McKeesport Main Post Office. Figure 11: Examples of Damage Resulting from a Partial Roof Collapse in December 2005 That Caused an Emergency Suspension of Operations at the Central Station in McKeesport, Pennsylvania: [See PDF for image] This figure is a series of three photographs depicting examples of damage resulting from a partial roof collapse in December 2005 that caused an emergency suspension of operations at the Central Station in McKeesport, Pennsylvania. The pictures depict damage around a wall and pipes, a damaged rest room, and the front door to the facility with a notice of closing posted. Source: GAO. [End of figure] In locations where demand for postal services is low (as measured by low revenue), the Postal Service sometimes chooses to reduce operating costs by reducing the staffing to just one postal employee--an approach that could place postal employees at risk unless the Postal Service installs needed security upgrades. We visited some single-employee facilities that had not received upgrades the Postal Service identified as necessary in order to protect the safety of its employees and customers. For example, when staffing at the two postal stations we visited in Indiana was reduced to one employee, the Postal Service inspected the facilities and identified numerous security deficiencies at both. More than 6 months after the inspections, however, none of the identified security upgrades had been completed and several of the upgrades were listed as deferred. The sole employee at another postal station we visited in Indiana complained in a 2001 letter to her supervisor about the lack of postal-identified security measures, such as security cameras, at her facility. Six years later, no actions have been taken to install additional security measures, such as a video- monitoring system or a pull-down gate to help secure the front register when the employee goes to other areas of the station. According to the manager of the station, who is located at a nearby postal retail facility, installing security systems at the station would increase costs--a result that would be incompatible with the goal of decreasing the station's operating costs. Lack of Criteria and Facility Management Data Limits the Postal Service's Ability to Proactively Identify Unneeded Retail Facilities for Possible Closure: In 2002, the Postal Service established a goal of reducing the number of "redundant, low-value" retail facilities in order to lower its operating costs. Establishing such a goal suggests that these facilities are less important to the Postal Service's mission of providing universal access to postal services than other facilities. To implement this goal, in 2002, the Postal Service lifted a moratorium on closing retail postal facilities but has not (1) provided a definition for "redundant, low-value" retail facilities; (2) established a goal for their reduction; or (3) identified unneeded facilities for possible closure, including those with low revenue. According to Postal Service officials, pursuing retail facility reductions is difficult because of legal restrictions on and political pressures against closing retail facilities. For example, legal restrictions preclude the Postal Service from closing a small post office solely because it is operating at a deficit. To close a post office, the Postal Service is required to, among other things, formally announce its intention to close the facility, analyze the impact of the closure on the community, and solicit comments from the community. While the Postal Service closes some retail facilities placed on emergency suspension, its reliance on other factors, such as the loss of a lease or severe maintenance problems, to drive decisions about closing retail facilities is inconsistent with leading federal practices which call for a targeted, criteria-based approach to closure decisions. More specifically, leading federal practices require applicable agencies to "rightsize" their facility holdings by, among other things, closing facilities that are (1) not critical to their mission, (2) in poor condition, (3) not fully utilized, or (4) costly to operate relative to their revenue. While considering these criteria is essential for rightsizing a facility network, the Postal Service cannot consider them because it does not capture the data needed to do so. During our review, we visited a number of postal facilities that appeared to merit consideration for closure based on one or more of these criteria. Furthermore, none of these facilities housed carriers or mail processing functions and each had low sales and was located near other retail facilities. Our site visits included the following types of facilities in urban locations: * Facilities that contribute little to the Postal Service's mission of providing universal access to postal services. For example, Station C in downtown Dallas, Texas, provides access to postal services only for the people that work in the secured federal building, and the station is located just 1 half-mile from another retail facility in downtown Dallas. Consequently, Station C's annual retail sales of about $282,000 in fiscal year 2007 ranked among the lowest in the Dallas District. * Facilities in poor condition. We observed maintenance issues at a number of postal stations that do not appear critical to the Postal Service's mission based on their low revenue and proximity to other retail facilities. For example, the Postal Service recently renewed the lease for the Wilkinsburg Station in Pittsburgh, Pennsylvania, even though the facility's façade is starting to pull loose from the building, the roof has numerous leaks, and sewage backs up throughout the facility's plumbing system. Postal Service officials said the Postal Service did not consider closing the station before renewing the lease, even though it does not appear critical--as it earns below- average revenue and is located about 2 miles from another retail facility in Wilkinsburg. * Facilities not fully utilized. One example is the Downtown Station in Fort Worth, Texas--a large, four-story building where the Postal Service conducts retail and carrier operations on the main floor but does not use the other floors (see fig. 6). * Facilities that are costly to operate relative to their value. We visited several postal stations with annual sales below $200,000, which is approximately what the Postal Service requires from each of the 2,500 automated postal centers located within postal retail facilities.[Footnote 30] For example, the Gary Downtown Finance Station, Indiana, which is one of seven postal retail units operating in the city, had annual revenue of $163,000 in fiscal year 2006. Exclusively retail, this station does not support any mail delivery functions beyond some post office boxes. * Facilities with high operating costs relative to their revenue. For example, the "Store of the Future" in the Pittsburgh International Airport, Pennsylvania (see fig. 12), costs more per square foot ($95) for the Postal Service to lease than any other facility in the Pittsburgh District, yet it has below-average sales revenue. Postal Service officials said the facility's revenue is low because it is located behind airport security check points. Figure 12: The "Store of the Future" Located behind Security at the Pittsburgh International Airport, Pennsylvania: [See PDF for image] This figure is a photograph of the "Store of the Future" inside the Pittsburgh International Airport. Source: GAO. [End of figure] Conclusions: The Postal Accountability and Enhancement Act has significant near-and long-term financial implications for postal operations. Consequently, it is imperative that the Postal Service manage its facilities as efficiently and cost-effectively as possible. However, the Postal Service cannot begin to overcome its facility management challenges until it improves the quality of its facility data. To date, FDB has failed to serve as a reliable system for inventorying postal facilities or for measuring their performance, and its weaknesses are so great that it cannot be reliably used for basic facility management purposes, such as the Postal Service's annual reporting. While efforts are under way to improve FDB, it remains unreliable and unusable for measuring performance. Instead of solving the Postal Service's problem of having multiple, overlapping databases, FDB appears to have compounded the problem by adding an additional data source to those already available. Postal officials with responsibility for facility maintenance at the national, area, district, and local levels said that the Postal Service has underfunded its maintenance for years and suspects that this underfunding has resulted in deteriorating facilities and a large maintenance backlog. However, the Postal Service does not know the extent of the problem because it has not comprehensively tracked and analyzed data on the condition of its facilities. This situation may be changing because the Postal Service has recognized the importance of compiling data for facility maintenance and has initiated a comprehensive facility assessment. Conducting a comprehensive facility assessment is a necessary first step toward improving the condition of postal facilities, but it will initially add tens of thousands of new maintenance projects to the Postal Service's maintenance backlog. Funding constraints will require the Postal Service to take an incremental approach in order to reduce the backlog. The Postal Service will have to make difficult choices about what repairs to make and what repairs to defer. These choices would be easier if the performance data on the Postal Service's facilities, such as a facility's importance, utilization rate, or costs, were reliable. One way to minimize maintenance costs is to reduce the number of facilities that must be maintained. In locations where new services are needed, the Postal Service is developing alternative access options to avoid new facility costs, but it has not identified or closed unnecessary postal retail facilities. Moreover, although the Postal Service has set a goal of shifting 40 percent of its retail sales to alternative access options, it has not set any similar targets for reducing its vast network of post offices, stations, and branches. Instead, it relies on emergency suspensions and staffing reductions to curtail operations at some facilities. However, this approach does not conform to leading federal practices because the closures are not linked to the facilities' performance. In addition, the staffing reductions potentially place the remaining postal employees at risk. The expected increase in the use of alternative access options, combined with financial necessity, suggests the need to consider additional closures of brick-and-mortar postal facilities. To properly consider the closure opportunities, the Postal Service will need to know which retail facilities to retain and which facilities are no longer important to its retail mission. With improved facility data, the Postal Service could assess a facility's importance and information on other relevant factors, such as a facility's utilization rate and condition, to identify closure possibilities and justify any closure decisions. Recommendations for Executive Action: To improve the Postal Service's management of its facilities, we are making the following six recommendations: * To strengthen the reliability and usefulness of the Postal Service's facility data, the Postmaster General should (1) direct the Vice President of Delivery and Retail to determine, in consultation with the Vice Presidents of Facilities and Intelligent Mail and Address Quality, whether it is more cost-effective to make FDB a reliable source for consolidated data on its facilities or to replace it with a new, more reliable database. If the Postal Service decides to retain FDB, the (2) Postal Service should take steps to improve its reliability and usefulness by establishing internal controls, such as edit checks, to preclude obvious mistakes. * To conform to leading federal practices, the Postmaster General should direct the Vice President for Delivery and Retail to (3) measure facility management performance consistent with the spirit of those developed by the Federal Council; and (4) begin tracking facility management performance trends. * To improve facility management and reduce long-term facility costs, the Postmaster General should, consistent with leading federal practices, (5) direct the Vice President for Facilities to prioritize maintenance projects based on a facility's overall performance, including measures consistent with the spirit of those developed by the Federal Council; and (6) direct the Vice President for Delivery and Retail to institute a proactive, criteria-based approach to assist in identifying unneeded retail facilities for possible closure as part of the June 2008 facility plan required by the Postal Accountability and Enhancement Act. Agency Comments and Our Evaluation: The Postal Service provided its written comments on a draft of this report by letter dated November 19, 2007. These comments are summarized below and are included, in their entirety, as appendix III to this report. The Postal Service agreed with our two recommendations to improve FDB's reliability; agreed, in principle, with our recommendation to prioritize maintenance projects based on a facility's overall performance; but disagreed with our three remaining recommendations. In separate correspondence, the Postal Service also provided minor technical comments, which we incorporated, as appropriate. In agreeing with our two recommendations regarding FDB, the Postal Service indicated that it had already determined that it is in its best interest and more cost-effective to retain FDB rather than to replace it with a new, more reliable database. Second, while the Postal Service believes FDB is, as a whole, very reliable, it agreed to establish additional controls to improve its reliability and usefulness. Furthermore, although the Postal Service stated that it prioritizes maintenance projects adequately, it agreed, in principle, with our recommendation to prioritize maintenance projects based on a facility's overall performance. The Postal Service also reiterated actions it has taken, including the initiation of its Facility Condition Assessment program, to better identify and prioritize the agency's maintenance workload and budget. The Postal Service disagreed with our three remaining recommendations, including our two recommendations to conform to leading federal practices by measuring facility management performance and tracking performance trends. Regarding these two recommendations, the Postal Service noted that it considers leading federal practices developed by the Federal Council, but reiterated that it is not required to adopt them.[Footnote 31] The Postal Service also noted that its mandate to provide universal mail service on an almost daily basis and its provision of services and products that are in direct competition with the private sector pose unique challenges for the Postal Service. Thus, the agency indicated that it had concerns about adopting leading federal practices. We continue to believe that measuring and tracking performance in such areas as a facility's importance, utilization, condition, and operating costs are critical to effective facility management. We also recognize the uniqueness of the Postal Service's business and mission. Consequently, we revised these recommendations to emphasize that the Postal Service should develop and track performance measures that may better meet its needs--as long as the measures are consistent with the spirit of those developed by the Federal Council. Finally, the Postal Service disagreed with our recommendation to institute a criteria-based approach to identify and close unneeded retail facilities as part of the June 2008 facility plan required by the Postal Accountability and Enhancement Act. The Postal Service stated that it had purposely removed mention of retail facility closures from its 2005 update of its Transformation Plan due to its difficulty in establishing criteria that could be applied to all retail locations. Instead, the Postal Service indicated that it intends to continue to assess its retail facilities on a case-by-case basis. We agree that developing criteria for identifying unneeded retail facilities is difficult and that such criteria cannot be uniformly applied to the universe of all postal retail facilities. As a result, we clarified our recommendation to indicate that the results of a criteria-based approach would "assist" the Postal Service in identifying candidate retail facilities for possible closure. The Postal Service further indicated that the focus of its forthcoming facility plan is on its mail processing network, not its retail facilities. While the Postal Service does not currently envision developing such a criteria-based approach as part of its congressionally required facility plan, we continue to believe that doing so would provide an excellent opportunity for the agency to begin pursuing its 2002 goal of identifying and closing redundant, low value retail facilities. We are sending this report to the congressional requesters and their staffs. We are also sending copies to the Postmaster General and other interested parties. We will make copies available to others upon request. In addition, the report will be available at no charge on the GAO Web site at [hyperlink, http://www.gao.gov]. If you or your staffs have any questions regarding this report, please contact me at siggerudk@gao.gov or (202) 512-2834. Contact points for our Office of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff that made key contributions to this report are listed in appendix IV. Signed by: Katherine A. Siggerud: Director, Physical Infrastructure Issues: [End of section] Appendix I: Executive Order 13327: Federal Real Property Asset Management: In 2004, the President signed Executive Order 13327: Federal Real Property Asset Management in response to our designation of federal real property as a governmentwide high-risk area.[Footnote 32] The executive order and the initiative to implement the order were intended to ensure that agencies maintain facility inventories appropriate to their needs, cost, and physical condition to support each agency's mission and objectives. This order applies to 24 executive departments and agencies,[Footnote 33] but not to the Postal Service.[Footnote 34] The executive order established the Federal Real Property Council (Federal Council), which includes representatives from the Office of Management and Budget and applicable agencies. The Federal Council develops guidance, serves as a clearinghouse for leading federal practices, and facilitates agency efforts to implement the executive order. Although not bound by the executive order, the Postal Service has voluntarily adopted some of the Federal Council's leading practices. Specifically, the Postal Service submits some data to the General Services Administration for its annual real property inventory and has designated an individual to represent the Postal Service on the Federal Council. The Federal Council has developed guidance that, among other actions, requires applicable agencies to collect 24 data elements for transmission to the General Services Administration to use in compiling an annual inventory of federal facilities. This data collection is intended to, among other things, provide agencies with data needed to assess their performance based on four key performance measures and determine whether their facilities are properly aligned with customer and service needs. The following four performance measures are intended to help agencies determine if the continued use of each of their facilities is justified based on their: 1. importance to achieving the agency's mission; 2. utilization rate (extent to which the facility is used); 3. physical condition as measured by a "condition index";[Footnote 35] and: 4. annual operating costs, including the recurring maintenance and repair costs for each facility in an agency's inventory. [End of section] Appendix II: Objectives, Scope, and Methodology: Our overall objectives were to identify the Postal Service's efforts to address three key facility challenges related to (1) capturing and maintaining accurate facility data, (2) adequately maintaining postal facilities, and (3) aligning access to retail services with customer needs. Specifically, this report identifies the Postal Service's goals and actions for managing each of these challenges and assesses its progress in overcoming the challenges and, as applicable, in implementing leading federal practices. To address these objectives, we visited 58 postal facilities and 4 contract postal facilities in nine districts covering three of the Postal Service's nine areas. The observations derived from our site visits cannot be generalized to the population of postal facilities nationwide. We chose these locations to achieve geographic balance and to include areas with growing and declining populations. We chose the specific facilities within these areas to achieve a range in the type, revenue, condition (indicated by the number of open maintenance requests in the Postal Service's maintenance tracking system), and size. This selection process led us to visit of each of the following types of postal facilities: main post office, postal station, postal branch, contract postal facility, carrier annex, processing and distribution center, air mail center, bulk mail processing center, vehicle maintenance facility, and administrative facility. We toured each facility, interviewed local employees, and collected data on facility operations. Table 1 identifies the specific areas, districts, and cities we visited. Table 2: Postal Areas, Districts, and Cities Visited: Postal area: Eastern; District: Greensboro; City and state: Greensboro, NC. Postal area: Eastern; District: Greensboro; City and state: Chapel Hill, NC. Postal area: Eastern; District: Greensboro; City and state: Carrboro, NC. Postal area: Eastern; District: Greensboro; City and state: Durham, NC. Postal area: Eastern; District: Baltimore; City and state: Cumberland, MD. Postal area: Eastern; District: Pittsburgh; City and state: Uniontown, PA. Postal area: Eastern; District: Pittsburgh; City and state: McKeesport, PA. Postal area: Eastern; District: Pittsburgh; City and state: Pittsburgh, PA. Postal area: Eastern; District: Pittsburgh; City and state: Carnegie, PA. Postal area: Great Lakes; District: Chicago; City and state: Chicago, IL. Postal area: Great Lakes; District: Central Illinois; City and state: Bedford Park, IL. Postal area: Great Lakes; District: Northern Illinois; City and state: Lombard, IL. Postal area: Great Lakes; District: Northern Illinois; City and state: Palatine, IL. Postal area: Great Lakes; District: Greater Indiana; City and state: East Chicago, IN. Postal area: Great Lakes; District: Greater Indiana; City and state: Gary, IN. Postal area: Great Lakes; District: Greater Indiana; City and state: Indianapolis, IN. Postal area: Southwest; District: Forth Worth; City and state: Colleyville, TX. Postal area: Southwest; District: Forth Worth; City and state: Forth Worth, TX. Postal area: Southwest; District: Forth Worth; City and state: Denton, TX. Postal area: Southwest; District: Dallas; City and state: Dallas, TX. Source: GAO. [End of table] To assess the Postal Service's progress in overcoming the challenge of capturing and maintaining accurate facility data, we analyzed all facility entries (over 36,000) contained in the Facility Database (FDB) as of October 5, 2006, and assessed the reliability of the data it contains. To assess the reliability of FDB data, we (1) interviewed Postal officials to obtain an understanding of the data, the database structure, the sources of the data, known issues or limitations in the database, and relevant Postal Service documentation; (2) reviewed related reports by the Postal Service Office of Inspector General (Postal OIG); and (3) performed electronic testing of the database for completeness, obvious errors, and inconsistencies. We also attempted to verify the data in FDB during our 58 facility visits. Prior to our visits, we identified potential data reliability problems and, during our visits, documented information related to the existence and causes of those problems. In addition, we corroborated our observations on FDB reliability issues through interviews with and documentation obtained from postal officials in the Delivery and Retail department. To determine the reasons for the data reliability issues in FDB, we interviewed Postal Service personnel in the Delivery and Retail department responsible for managing the facilities data, local postal officials, and postal officials from the Postal OIG. We also reviewed key Postal Service documents related to the facilities data, including the FDB user's guide, the Postal Service's 2002 and 2006 Transformation Plans, and reports issued by the Postal OIG and GAO. Our analysis was based principally on the FDB data as of October 5, 2006, but we also analyzed FDB data obtained on July 7, 2007, in part, to assess the impact of the Postal Service's efforts to improve FDB since October 6, 2006. To identify examples of leading federal practices related to our objectives, we reviewed Executive Order 13327: Federal Real Property Asset Management, related documentation from the Federal Council, and previous GAO reports. We also interviewed a Federal Council official and attended a Federal Real Property Association conference in 2006 that focused on the implementation of the executive order. To assess the Postal Service's progress in overcoming the challenge of maintaining its facilities in adequate condition, we analyzed data drawn from the Postal Service's maintenance tracking system (the Facility Single Source Provider) as of March 7, 2007. We assessed the reliability of the data by (1) interviewing knowledgeable Postal Service officials and (2) validating the accuracy of the data at the facilities we visited. We also observed conditions at the 58 facilities we visited and interviewed local postal officials. Additionally, we analyzed key Postal Service documents related to facility maintenance, including the Postal Service's Strategic Transformation Plan for 2006 through 2010, presentations related to its facility assessment effort, and an audit report issued by the Postal OIG. To assess the Postal Service's progress in overcoming the challenge of aligning access to retail services with customer needs, we assessed the variance in access to retail postal facilities among similar counties using an ordinary-least squares regression model. We designed the model to predict the number of postal retail facilities (main post offices, stations, branches, and contract units) in each U.S. county based on each county's population (in 2000), land area (in square miles), and degree of urbanization (as defined by the U.S. Census Bureau).[Footnote 36] First, we determined the number of retail facilities in each county by conducting a unique address analysis, which counted only one retail facility at each address within FDB to avoid counting the same retail facilities more than once. The regression coefficients from the regression model are presented in table 3. Table 3: Regression Coefficients Used to Establish the Average Number of Retail Postal Facilities in Counties Based on Each County's Population, Land Area, and Degree of Urbanization: County characteristic[A]: Intercept; Estimate: 1.861[B]; Standard error of the estimate: 0.292. County characteristic[A]: Population in 2000 (in thousands); Estimate: 0.318[B]; Standard error of the estimate: 0.016. County characteristic[A]: Degree of micropolitan urbanization[C]; Estimate: 0.618; Standard error of the estimate: 0.539. County characteristic[A]: Population micropolitan; Estimate: - 0.142[B]; Standard error of the estimate: 0.018. County characteristic[A]: Degree of metropolitan urbanization[C]; Estimate: 6.581[B]; Standard error of the estimate: 0.358. County characteristic[A]: Population metropolitan; Estimate: - 0.276[B]; Standard error of the estimate: 0.016. County characteristic[A]: Land area in square miles (in thousands); Estimate: 0.279[B]; Standard error of the estimate: 0.033. Source: GAO. Notes: Adjusted R-squared = 0.79 N (counties in the United States and Puerto Rico) = 3,213: [A] Because the rural degree of urbanization serves as the baseline of the model, there is no separate estimate for rural urbanization. To obtain the predicted value for rural counties, we used the following regression equation for the number of postal retail facilities in rural counties: post offices in rural counties = 1.861 + 0.318 (population in thousands) + 0.279 (land area in thousands of square miles) + model error. [B] Indicates statistical significance at the 0.001 level. [C] Degree of urbanization is specified as a dummy variable in this model: 1 if micropolitan, 0 otherwise, and 1 if metropolitan, 0 otherwise. [End of table] The R-squared statistic, an indicator of model fit or explanatory power, was 0.79. This means the model explains 79 percent of the variation in the number of post offices between counties. We performed various model specification and diagnostic tests to determine the appropriateness of our model specifications (e.g., interaction effects and degree of urbanization as dummy variables), as well as to determine the existence and influence of outliers. We identified two potential outliers, but did not eliminate them from our analysis because they did not materially affect our results.[Footnote 37] To determine the variation in the number of postal retail facilities between counties, we examined the residuals from the ordinary-least squares regression model. The residuals consist of the difference between the actual number of postal retail facilities and the number predicted by the model. The number predicted by the model can be considered the average number of facilities offering postal retail services in counties of similar population, land area, and degree of urbanization because the model, in deriving the estimates, considers these factors across all 3,213 counties in the United States and Puerto Rico. We then rank ordered the counties on the difference between the actual and the predicted number of postal retail facilities from low to high (i.e., from fewer than predicted to more than predicted). For the map in figure 7, we considered the top 10 percent of counties in this rank order as having more post offices than comparably sized and populated counties and the bottom 10 percent of counties as having fewer post offices than comparable counties. We limited this portion of our analysis to counties with an urban center of at least 10,000 people--defined as metropolitan and micropolitan areas--because the Postal Service may need to maintain more postal retail facilities in these areas to fulfill its mission of providing universal access to postal services. We interviewed postal officials at the facility, district, area, and headquarters levels about how retail facilities are aligned with customer needs and obtained documentation relating to alternative service options and emergency suspensions that resulted in facility closures. We also reviewed key Postal Service documents, including the policies and procedures for closing postal facilities, federal regulations governing postal facility closures, the Postal Service's 2002 and 2005 Transformation Plans, audit reports issued by the Postal OIG, and previous GAO reports. We conducted our work from July 2006 through December 2007 in accordance with generally accepted government auditing standards. [End of section] Appendix III: Comments from the Postal Service: Note: GAO comments supplementing those in the report text appear at the end of this appendix. United States Postal Service: 475 L'Enfant Plaza SW: Washington, DC 20260: [hyperlink, http://www.usps.com]: November 19, 2007: Katherine A. Siggerud: Director, Physical Infrastructure Issues: United States Government Accountability Office: Washington, DC 20548-0001: Dear Ms. Siggerud: Thank you for providing the U.S. Postal Service with the opportunity to review and comment on the draft report titled U.S. Postal Service Facilities: Improvements in Data Would Strengthen Maintenance and Alignment of Access to Retail Services, (GAO-08-41). Data Integrity and Reliability: The report accurately reflects that our expansive facility network poses significant real property management challenges. To meet this challenge, the Postal Service relies on a number of systems to capture, monitor, and track information about each of our facilities. However, this report focuses almost exclusively on the data and reliability of our Facilities Database (FDB). The following provides information on each of the systems identified in the report and clarification as to its purpose: * Address Management System—maintains the master postal addressing database. * Electronic Facilities Maintenance System (eFMS)-manages and tracks construction, repair and alteration projects, and all leasing activities. * Facilities Database—serves as a repository/information database of facility information. The theme throughout the report is that FDB has not met the Postal Service's goal for a single, consolidated source of accurate facility data; the report also seems to target FDB's demise. We want to make it very clear that FDB was never intended to be what the report claims. As stated above, FDB is a repository of facility information. However, the report dismisses FDB without any exploration of how the information it contains is actually used throughout the Postal Service, and states that several internal organizations "do not use FDB for aggregate facility information, partly because of concerns about its reliability." Did the GAO base this conclusion, in part, on the various sources used in our annual report to account for facility information? If so, please be aware the data for our annual report properly extracts this information from the appropriate source, not FDB alone. [See comment 1] Recommendations 1 and 2 state: To strengthen the reliability and usefulness of the Postal Service's facility data, the Postmaster General should direct the Vice President of Delivery and Retail to (1) determine, in consultation with the Vice Presidents of Facilities and Intelligent Mail and Address Quality, whether it is more cost-effective to make FDB a reliable source for consolidated data on its facilities or to replace it with a new, more reliable database. If the Postal Service decides to retain FDB, take steps to improve its reliability and usefulness by (2) establishing internal controls, such as edit checks, to preclude obvious mistakes. [See comment 2] We agree with recommendation 1 and, as such, determined it is in our best interest and more cost-effective to retain FDB. We believe the report's recommendation to consider replacement of FDB is based on an overly ambitious goal for this repository of facility information, and to replace it would be fiscally irresponsible. We also recognize there are issues with any database; therefore, we agree with recommendation 2, even though we believe FDB is, as a whole, very reliable. We are already working to improve the reliability of FDB by putting additional controls in place to minimize errors whenever data is captured from feeder systems and during any data-entry process performed by local management. On several occasions, when confronted with the report's findings, our staffs were able to explain most of the data reliability issues brought to their attention with the use of the aforementioned systems. The Postal Service uses these systems in concert to achieve its mission. Errors which were brought to our attention during these discussions with GAO personnel were corrected, and we would benefit tremendously if all the identified issues cited in the report were shared with us, as requested previously. In addition, while the report specifically states that it could not determine the overall magnitude of FDB's data reliability, it then speculates that errors in FDB could cause the Postal Service's rent obligation to vary by as much as $82M. GAO has not provided data to allow us to evaluate this conclusion, and in any event, the Lease Payment System within eFMS is used to determine all USPS rent payments, not FDB. [See comment 3 and 4] Leading Federal Practices: Recommendations 3 and 4 state: To better conform to leading federal practices, the Postmaster General should direct the Vice President of Delivery and Retail to (3) adopt federal leading practices [sic] for measuring facility management performance, including a facility's importance to achieving the Postal Service's mission and the facility's utilization rate, condition, and annual operating costs, and (4) begin tracking facility performance management over time. We do not agree with recommendations 3 and 4. As duly noted in the report, the Postal Service is not covered by the executive order which established the Federal Real Property Council. While we voluntarily participate in meetings whenever topics are germane to the Postal Service, considerations on meeting real property management challenges for the Postal Service are very different than those for the rest of the federal government. We are mandated to provide universal mail service on an almost daily basis, but we also provide services and products that have direct competition with the private sector. The report repeatedly cites that the Postal Service does not follow leading federal practices the Council requires of covered agencies. These practices are to establish current and reliable inventory of facilities and to capture, track, and report facility management performance- related data on 1) the importance to achieving the agency's mission, 2) utilization rate, 3) physical condition, and 4) annual operating costs. While we do consider leading federal practices, we have real concerns about applying these leading federal practices, in their entirety, to Postal Service activities across the board due to the uniqueness of our mission and business. [See comment 5] Today, we consider the majority of our facilities to be important to achieving the agency's mission - from a one-person, part-time office in rural America providing a sense of community, to a facility which processes millions of pieces of mail each week. However, as noted in the report, we do close and consolidate facilities, which is an admittedly difficult process due to specific legal requirements and political realities. To determine utilization rate, we seek input from local installation heads on the availability of vacant space in their facility. Facilities Service Offices, in cooperation with our Operations personnel, determine the potential leasibility or saleability of excess space. In addition, the Postal Service now has a national revenue generation program that has identified potential disposal and developmental opportunities for excess space. We do not rely solely upon local managers to identify excess space, and FDB was never intended to be the only source to identify such space. Facility Management and Costs: Recommendations 5 and 6 state: To improve facility management and reduce long-term facility costs, the Postmaster General should, consistent with leading federal practices, (5) direct the Vice President for Facilities to prioritize maintenance projects based on a facility's overall performance, including the facility's importance, utilization rate, and costs, and not just on a project's urgency; and (6) direct the Vice President for Delivery and Retail to institute a proactive, criteria-based approach to identify and close unneeded retail facilities as part of the June 2008 facility plan required by the Postal Accountability and Enhancement Act. With regard to recommendation 5, we agree in principle with the idea of prioritizing maintenance projects and believe that we already prioritize them adequately. The report states the Postal Service focuses "solely on urgent repairs at the expense of routine, preventive maintenance;" this is simply not true. Certainly in the absence of unlimited funding, the Postal Service must balance the number of maintenance requests with available funds by prioritizing projects which range from discretionary (e.g., repainting office space) to emergency (e.g., safety of employees and customers). The report did acknowledge that for the first time on a national level, the Postal Service received deficiency information based on surveys of installation heads. And in conjunction with the Facilities Condition Assessment program of all larger facilities, data will be available to better identify and prioritize workload and budget. All facilities will be surveyed or inspected every three years to determine maintenance needs. [See comment 6] We disagree with recommendation 6, as we do not see this approach as part of the 2008 Facility Plan. The focus of the Facility Plan developed pursuant to the Postal Accountability and Enhancement Act of 2006 is our mail processing network, not retail facilities. As information, the Transformation Plan, updated in 2005 for fiscal years 2006 to 2010, purposely removed mention of retail facility closures, as it is difficult to agree and establish criteria that could be applied to all locations. As GAO noted, looking at retail facility distribution on a county-by-county basis is one evaluation option, but there are other approaches that could be considered (e.g., universal service, local infrastructure and economy, demographics, growth, competitive presence, etc.). As discussed when we met, we have learned that each facility must be assessed on a case-by-case basis, and we intend to continue to use this method. [See comment 7] The report appears to reach the conclusion that for a facility to be viable, it must meet the minimum revenue threshold that is required of an Automated Postal Center (APC). The APCs are intended for use in high growth areas where there is significant customer traffic; APCs also enable us to shift less complex transactions away from the retail counter. It is not appropriate to apply the APC's revenue criteria to our brick and mortar retail facilities. We would be glad to further discuss the GAO's criteria for identifying unneeded facilities. [See comment 8] We again request detailed information from GAO on its site visits to enable us to correct the facility-specific issues that were identified and highlighted in the report. We would also appreciate receiving all specific information on any data reliability issues found. If you or your staff wishes to discuss any of these comments further, we are available at your convenience. Sincerely, Singed by: Kathy Ainsworth: Vice President Delivery and Retail: The following are GAO's comments on the Postal Service's letter dated November 19, 2007. GAO Comments: 1. The Postal Service stated that FDB is a repository of facility information and that it was never intended to be a single, consolidated source for information on facilities. The agency's September 2006 FDB briefing to us appears to dispute this point. At that time, postal officials indicated that the agency had intended to create "one source of information for tracking all aspects of our [its] facilities" to address, among other matters, the high cost of maintaining inaccurate and redundant databases. Notwithstanding our prior understanding, we clarified our report, indicating that FDB was intended to be a "consolidated" source for accurate facility data--not a "single," consolidated data source. 2. The Postal Service questioned a statement in our report which indicated that several internal (postal) organizations "do not use FDB for aggregate facility information, partly because of concerns about its reliability." More specifically, the agency inquired whether we had drawn this conclusion on the basis of the various data sources used to compile the Postal Service's 2006 annual report. While the variety of data sources used in the annual report--none of which were exclusively FDB--is illustrative of our point, our statement was principally based on interviews with numerous officials throughout the Postal Service's major departments. 3. The Postal Service correctly noted that, after researching multiple data sources, its staff was able to provide explanations for specific examples of data reliability issues--including reasons for identified mistakes--that we brought to their attention. Although these explanations appeared plausible, as discussed in our report, without additional site visits, we cannot determine whether their explanations were accurate. Most importantly, however, regardless of whether the identified mistakes could be explained using other data sources--the fact remains that FDB contains errors that need to be corrected if it is to retained and used as a reliable source of agency data. 4. The Postal Service indicated that it would benefit tremendously if we shared information on all of the specific examples of data reliability issues we used in this report. We appreciate the Postal Service's desire to improve the reliability and usefulness of FDB and, therefore, suggest that the Postal Service replicate our statistical analyses on current data by, among other standard analytical methods, sorting its current FDB data according to each facility's name, address, and type in order to identify duplicate facility entries. If needed, we could also meet to further discuss our analytical methods for analyzing the Postal Service's data. 5. As discussed in the body of this report, we continue to believe that measuring and tracking performance in such areas as a facility's importance, utilization, condition, and operating costs are critical to effective facility management. Nevertheless, because we also recognize that the Postal Service's business and mission are unique, we revised these recommendations to emphasize that the Postal Service should develop performance measures that may better meet its needs--as long as the measures are consistent with the spirit of those developed by the Federal Council. 6. The Postal Service disagreed with our statement that it focuses "solely on urgent repairs at the expense of routine, preventive maintenance." While we removed "solely" from the applicable sentence in this report, this statement was based on evidence obtained during interviews with numerous postal employees involved with facility maintenance at the national, area, and local levels. All of these individuals stated that resource constraints have forced the Postal Service to focus on urgent repairs and to defer routine or preventive maintenance projects. Furthermore, the interviewees' comments were consistent with findings in a prior Postal OIG review which found that insufficient budgeting for repairs and maintenance may be hampering the Postal Service's ability to proactively manage its maintenance.[Footnote 38] 7. As discussed in the body of this report, we agree that developing criteria for identifying unneeded retail facilities is difficult and that such criteria cannot be uniformly applied to the universe of all postal retail facilities. As a result, we clarified our recommendation to indicate that the results of a criteria-based approach would "assist" in identifying candidate retail facilities for closure. Furthermore, although the Postal Service does not currently envision developing such a criteria-based approach as part of its congressionally required facility plan, we continue to believe that doing so would provide an excellent opportunity for the agency to begin pursuing its 2002 goal of identifying and closing redundant, low value retail facilities. 8. We are not intending to imply that in order to be viable, a facility must meet minimum revenue thresholds established for the Postal Service's Automated Postal Centers. As discussed in our report, this threshold was $198,000 in fiscal year 2007. We included information on this threshold simply as context for the amount of revenue generated by some of the postal retail facilities we visited-- all of which were in urban areas. While we agree that the threshold the Postal Service applied to the deployment of automated postal centers is not necessarily applicable to its retail facilities, the Postal Service has not established criteria based on revenue or any other criteria for analyzing the performance of its retail facilities. The absence of such action is among the reasons we are recommending that the Postal Service institute a proactive, criteria-based approach to help identify and close unneeded retail facilities. [End of section] Appendix IV: GAO Contact and Staff Acknowledgments: GAO Contact: Katherine A. Siggerud, (202) 512-2834 or siggerudk@gao.gov: Staff Acknowledgments: In addition to the individual named above, Kathleen Turner, Assistant Director; Michael Armes; Richard Bakewell; Keith Cunningham; Bess Eisenstadt; Kathy Gilhooly; Brandon Haller; Anne Izod; Dorothy Yee Leggett; Josh Ormond; and Minette Richardson made key contributions to this report. [End of section] Footnotes: [1] Pub. L. No. 109-435 (Dec. 20, 2006). [2] Notwithstanding the annual limitation, rates may be adjusted on an expedited basis due to either extraordinary or exceptional circumstances. See Section 201 of Pub.L. No. 109-435 (2006). [3] The Executive Order applies to the heads of all executive branch departments and agencies listed in 31 U.S.C. 901(b)(1) and (b)(2). The Postal Service is not included in that list. [4] GAO, Federal Real Property: Progress Made Toward Addressing Problems, but Underlying Obstacles Continue to Hamper Reform, GAO-07- 349 (Washington, D.C.: Apr. 13, 2007). [5] The Postal Service is required by statute to consider specific factors in making a determination to close a post office and to give persons served by a post office the opportunity to present their concerns regarding a proposal to close the post office. See 39 U.S.C 404(b). Additionally, the Postal Service is precluded from closing a small post office solely because it operates at a deficit. See 39 U.S.C. 101(b). [6] The Postal Service closed 795 facilities due to emergency circumstances from 2002 through April 2007. [7] The Postal Service has nine geographically based area offices. The eight Facility Service Offices are aligned with the Postal Service's areas, but the Eastern and Capital Metro areas are combined into one Facility Service Office. [8] GAO, U.S. Postal Service: Transformation Challenges Present Significant Risks, GAO-01-598T (Washington, D.C.: Apr. 4, 2001). [9] GAO, U.S. Postal Service: USPS Needs to Clearly Communicate How Postal Services May Be Affected by Its Retail Optimization Plans, GAO- 04-803 (Washington, D.C.: July 13, 2004). [10] For more information on our assessment of data reliability, see appendix II of this report. [11] According to Postal Service officials, some of this variation could be caused by the existence of different leases for the same facility, but we were unable to determine the reasons for the variations from the FDB data. [12] According to the Postal Service, some inconsistent ownership data may be associated with facilities owned by the Postal Service on leased land or facilities leased by the Postal Service on land that it owns, but this information could not be determined from the FDB data we received. [13] The Postal Service also identified other data errors related to the Lubbock facilities. Specifically, according to the Postal Service, the Facility Management System contained an outdated address for the main post office (entry 1) and an incorrect address for the location of the postal vending machine (entry 4). [14] U.S. Postal Service, Office of the Inspector General, Data Integrity Review of Address Management System Facility Data, IS-AR-07- 005 (Washington, D.C., Dec. 22, 2006). [15] U.S. Postal Service, Office of the Inspector General, Audit Report: Data Input Validation for the Facilities Database, IS-AR-06-006 (Washington, D.C., Mar. 30, 2006). [16] U. S. Postal Service, Office of the Inspector General, Audit Report: Data Input Validation for the Facilities Database, IS-AR-06-006 (Washington, D.C., Mar. 30, 2006). [17] The contractor used a long-standing industry measure, called a facility condition index, which compares the cost of repairing a facility with the cost of replacing the facility to acceptable levels. [18] The Postal Service plans to stagger the assessments by assessing one-third of the facilities with 6,500 to 100,000 square feet each year. [19] U.S. Postal Service, Office of the Inspector General, Postal Service Facilities Maintenance and Repair Costs, CA-AR-07-003 (Arlington, Va., May 14, 2007). [20] U.S. Postal Service, Office of the Inspector General, Allegations of Unsafe Working Conditions at the Dallas Downtown Station, LM-AR-00- 002 (Arlington, Va., Mar. 20, 2000). [21] U.S. Post Office, Strategic Transformation Plan 2006-2010 (Washington, D.C., September 2005). [22] CA-AR-07-003. [23] The National Research Council of the National Academies is a nonprofit institution that provides science, technology and health policy advice to the federal government. [24] National Academy of Sciences, Committing to the Cost of Ownership: Maintenance and Repair of Public Buildings (Washington, D.C., 1990). [25] National Academy of Sciences, Investments in Federal Facilities: Asset Management Strategies for the 21st Century (Washington, D.C., 2004). [26] Our analysis includes retail facilities that are owned or leased by the Postal Service, as well as privately operated contract facilities that provide postal retail services. [27] We excluded counties without an urban center of at least 10,000 people from the output of our analysis because the Postal Service may need to maintain more postal retail facilities in these areas in order to fulfill its mission of providing universal access to postal services. See appendix II for additional details on our methodology. [28] In 2006, the Postal Service reported having 3,014 contract postal units and 937 community post offices. [29] GAO-04-803. [30] As of December 2006, the Postal Service had relocated 170 automated postal centers that did not meet revenue targets. [31] As noted in appendix I of this report, the President signed Executive Order 13327: Federal Real Property Asset Management in 2004 with the goal of ensuring that agencies maintain facility inventories appropriate to their needs, cost, and physical condition to support each agency's mission and objectives. This order applies to 24 executive departments and agencies, but not to the Postal Service. However, the Postal Service has voluntarily adopted some of the Federal Council's leading practices. [32] GAO, High-Risk Series: Federal Real Property, GAO-03-122 (Washington, D.C.: January 2003). [33] The executive order applies to the Departments of Agriculture, Commerce, Defense, Education, Energy, Health and Human Services, Homeland Security, Housing and Urban Development, Interior, Justice, Labor, State, Transportation, the Treasury, and Veterans Administration, as well as to the Environmental Protection Agency, the National Aeronautics and Space Administration, the United States Agency for International Development, the General Services Administration, the National Science Foundation, the Nuclear Regulatory Commission, the Office of Personnel Management, the Small Business Administration, and the Social Security Administration. [34] The Executive Order applies to the heads of all executive branch departments and agencies listed in 31 U.S.C. 901(b)(1) and (b)(2). The Postal Service is not included in that list. [35] The condition index is calculated by dividing the cost of a facility's repair needs by the facility's replacement value. "Repair needs" is the amount necessary to ensure that a facility is restored to a condition substantially equivalent to the facility's originally intended and designed capacity, efficiency, or capability. [36] Metropolitan (metro) and micropolitan (micro) statistical areas are geographic entities defined by the U.S. Census Bureau for use by federal statistical agencies in collecting, tabulating, and publishing federal statistics. A metro area contains a core urban area with a population of 50,000 or more, and a micro area contains an urban core with a population of at least 10,000 (but less than 50,000). Each metro or micro area consists of one or more counties and includes the counties containing the core urban area, as well as any adjacent counties that have a high degree of social and economic integration (as measured by commuting to work) with the urban core. [37] There were two potentially influential outliers (counties) as detected by the Cook's D measure (DFFITS). To test the influence of outlying counties on our regression model, we ran the model without these two counties. Although the R-squared of this model was somewhat lower at 0.75 with these two counties, we retained them in the final model because their inclusion did not materially affect the results for all other counties. [38] U.S. Postal Service, Office of the Inspector General, Postal Service Facilities Maintenance and Repair Costs, CA-AR-07-003 (Arlington, Va., May 14, 2007). [End of section] GAO's Mission: The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each weekday, GAO posts newly released reports, testimony, and correspondence on its Web site. To have GAO e-mail you a list of newly posted products every afternoon, go to [hyperlink http://www.gao.gov] and select "E-mail Updates." Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability Office: 441 G Street NW, Room LM: Washington, DC 20548: To order by Phone: Voice: (202) 512-6000: TDD: (202) 512-2537: Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: [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: Gloria Jarmon, Managing Director, jarmong@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: