United States General Accounting Office GAO Report to Congressional Requesters November 2003 AVIATION SECURITY Federal Air Marshal Service Is Addressing Challenges of Its Expanded Mission and Workforce, but Additional Actions Needed GAO-04-242 November 2003 AVIATION SECURITY Federal Air Marshal Service Is Highlights of GAO-04-242, a report to Addressing Challenges of Its Expanded congressional requesters Mission and Workforce, but Additional Actions Needed To help strengthen aviation To deploy its expanded workforce by July 1, 2002, a deadline set by the security after the September 11, 2001, terrorist attacks, the Deputy Secretary of Transportation, the Service used expedited procedures Congress expanded the size and to obtain interim secret security clearances for air marshal candidates and mission of the Federal Air Marshal provided abbreviated training for them. These procedures allowed Service (the Service) and located candidates with interim clearances to work until they received their final the Service within the newly top-secret clearances. Because of a governmentwide demand for clearances, created Transportation Security nearly a quarter of the active air marshals had not received their top-secret Administration (TSA). Between clearances as of July 2003; but by October 2003, only about 3 percent were November 2001 and July 1, 2002, awaiting their top-secret clearances. To train its expanded workforce before the Service grew from fewer than the Deputy Secretary’s deployment deadline, the Service incrementally 50 air marshals to thousands, and revised and abbreviated its curriculum. its mission expanded to include the protection of domestic as well as international flights. In March The Service has begun to develop management information, policies, and 2003, the Service, with TSA, procedures to support its expanded workforce and mission, but it has not merged into the new Department of yet completed this major effort. For example, it replaced a manual system Homeland Security (DHS); and in for scheduling flight duty with an automated system, but it has not yet November 2003, it was transferred developed an automated means to monitor the effectiveness of its from TSA and merged into DHS’s scheduling controls designed to prevent air marshals’ fatigue. In addition, it Bureau of Immigration and has gathered and used information on potential security incidents and on air Customs Enforcement (ICE). GAO marshals’ reasons for separation from the Service to improve its operations looked at operational and and workforce management. However, some of this information is not clear management control issues that or detailed enough to facilitate follow-up. Finally, the Service has emerged during the rapid implemented policies needed to support its expansion. expansion of the Service, specifically addressing its (1) background check procedures and The Service is likely to face challenges in implementing changes resulting training; (2) management from its mergers into DHS and ICE, including changes to its roles, information, policies, and responsibilities, and training and to its procedures for coordinating with procedures; and (3) challenges TSA’s security organizations, as well as administrative changes. GAO’s likely to result from its mergers recent work on mergers and organizational transformations proposes into DHS and ICE. several key practices—set implementation goals, establish a communication strategy, and involve employees to obtain their ideas—and associated implementation steps that could help the Service implement such changes. GAO is making recommendations designed to improve the Service’s Training Air Marshal Candidates to Shoot from a Seated Position data on flight duty and information on separations. DHS agreed with GAO’s recommendations and expressed a commitment to continuous improvement as the Service moves forward. www.gao.gov/cgi-bin/getrpt?GAO-04-242. To view the full product, including the scope and methodology, click on the link above. For more information, contact Gerald L. Dillingham at (202) 512-2834 or firstname.lastname@example.org. Contents Letter 1 Results in Brief 3 Background 5 Expediting Background Checks and Training Enabled the Service to Meet the Deputy Secretary’s Deployment Deadline 7 Management Information, Policies, and Procedures Have Not Kept Pace with Expanded Operations 13 Key Practices and Implementation Steps Can Help Address Merger Implementation Challenges 24 Conclusions 28 Recommendations for Executive Action 28 Agency Comments 29 Appendix I Scope and Methodology 31 Appendix II Demographic Profile of the Federal Air Marshal Service 35 Appendix III Locations of the Federal Air Marshal Service’s 21 Field Offices and Training Facility 37 Appendix IV Events Affecting the Federal Air Marshal Service, September 2001 through October 2002 38 Appendix V Mission-Related Incidents Reported by Federal Air Marshals, by Broad Categories, September 15, 2001 — September 16, 2003 39 Appendix VI Key Practices and Implementation Steps for Mergers and Organizational Transformations 40 Page i GAO-04-242 Federal Air Marshal Service Appendix VII Contacts and Acknowledgments 41 GAO Contacts 41 Acknowledgments 41 Figures Figure 1: Federal Air Marshal Service Workforce by Gender, by Percentage, as of August 2003. 35 Figure 2: Federal Air Marshal Service Workforce by Age, by Percentage, as of August 2003. 35 Figure 3: Federal Air Marshal Service Workforce by Race, by Percentage, as of August 2003. 36 Abbreviations DHS Department of Homeland Security DOT Department of Transportation FAA Federal Aviation Administration IG Inspector General ICE Immigration and Customs Enforcement LEAP law enforcement availability pay OPM Office of Personnel Management PDA personal digital assistants TSA Transportation Security Administration 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. However, 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. Page ii GAO-04-242 Federal Air Marshal Service United States General Accounting Office Washington, DC 20548 November 19, 2003 The Honorable Christopher Shays Chairman, Subcommittee on National Security, Emerging Threats and International Relations Committee on Government Reform House of Representatives The Honorable Diane E. Watson House of Representatives Within 10 months of the September 11, 2001, terrorist attacks on the United States, the number of federal air marshals grew from fewer than 50 to thousands;1 and within 2 years, the Federal Air Marshal Service (the Service) underwent three organizational transfers. More specifically, the Congress, through a provision of the Aviation and Transportation Security Act, enacted on November 19, 2001,2 authorized a dramatic expansion of the Service’s mission and workforce and transferred authority over the Service from the Federal Aviation Administration (FAA) to the Transportation Security Administration (TSA). Under this legislation, the Service’s mission grew from preventing hijackings on international flights to protecting passengers, crews, and aircraft from terrorist activities on both domestic and international flights. Additionally, the Deputy Secretary of Transportation set a goal of hiring, training, and deploying thousands of new air marshals by July 1, 2002. After the passage of the Homeland Security Act, the Service moved with TSA from the Department of Transportation (DOT) to the newly created Department of Homeland Security (DHS) on March 1, 2003. Finally, in September 2003, the Secretary of Homeland Security announced that the Service would be transferred from TSA to the Bureau of Immigration and Customs Enforcement (ICE), another law enforcement agency within the new department. Following this transfer, which was supposed to take place on November 2, 2003, the Secretary said that the Bureau’s three law enforcement workforces—air marshals, immigration agents, and customs agents—would be cross- trained to create a “surge capacity” for responding to security threats. 1 The exact number of federal air marshals is classified. 2 Public Law 107-71, November 19, 2001. Page 1 GAO-04-242 Federal Air Marshal Service The rapid expansion of the Service led to a number of operational and management control issues, which surfaced in national media reports of allegations that the Service conducted inadequate background checks on newly hired air marshals, slashed its training program to expedite the air marshals’ deployment, and failed to meet the needs of its air marshal workforce. For example, the Service allegedly over- or underscheduled air marshals for flight duty and reneged on promises of transfers to alternative locations, thereby creating dissatisfaction with the Service that, according to some reports, led to a flood of air marshal resignations. Within the context of these allegations, our objective was to look at operational and management control issues related to the Service’s expansion. We also considered implications of the Service’s organizational realignment. Specifically, as agreed with your offices, we addressed the following questions: • What procedures for obtaining background checks and providing training did the Service use to expedite the deployment of its expanded workforce to meet the Deputy Secretary’s July 2002 deadline? • To what extent has the Service developed management information and policies and procedures to support its expanded mission and workforce? • What challenges is the Service likely to face as a result of its recent mergers into DHS and ICE? To answer these questions, we analyzed program data; interviewed Service and TSA officials; and reviewed documentation from the Service and TSA on background checks and training; scheduling, mission incidents, employee misconduct, and separations; and reviewed several workforce policies and procedures. We also visited several facilities to look at the Service’s operational and management control practices and documents, including the Federal Air Marshal training facility and Human Resource Center in New Jersey, the Federal Air Marshal headquarters office in Virginia, the Federal Law Enforcement Training Center in New Mexico, and the Federal Air Marshal field office in Texas. To guide our assessment of the Service’s training, management information, and policies and procedures, we reviewed key GAO documents on internal controls and human capital management. These include our Standards for Internal Control in the Federal Government (GAO/AIMD-00-21.3.1, November 1999), Internal Control Management and Evaluation Tool (GAO-01-1008G, August 2001), Human Capital: A Guide for Assessing Strategic Training and Development Efforts in the Federal Government (GAO-03-893G, July Page 2 GAO-04-242 Federal Air Marshal Service 2003), and Model of Strategic Human Capital Management (GAO-02-373SP, March 2002). In addition, we used our report, Results-Oriented Cultures: Implementation Steps to Assist Mergers and Organizational Transformations (GAO-03-669, July 2003), to provide a framework for evaluating the Service’s challenges in merging into DHS. We also reviewed a March 2003 report by the DOT Inspector General (IG), which evaluated the Service’s selection and hiring process, training program, and scheduling process. Finally, we discussed the governmentwide background investigation process with the Office of Personnel Management (OPM). Our analysis of the operational and management control issues related to the Service’s expansion focuses primarily on the period from November 2001 through September 2003, when the Service was part of TSA; our assessment of the challenges related to the Service’s mergers is, in part, prospective. We conducted our review from September 2002 through October 2003 in accordance with generally accepted government auditing standards. See appendix I for a more detailed discussion of our scope and methodology. The Service used expedited procedures for obtaining background checks Results in Brief and abbreviated the training for air marshals so that it could deploy its expanded workforce by the Deputy Secretary’s July 2002 deadline. Under the expedited background check procedures, which other federal agencies also often use, candidates who pass preliminary background checks are able, within about 24 hours, to obtain interim secret security clearances that allow them to work until their full background checks are completed. Thousands of candidates underwent preliminary background checks, and the majority of them obtained interim security clearances. Obtaining final top-secret clearances has taken longer—sometimes up to a year—and as of July 2003, nearly a quarter of the active air marshals were still operating under interim clearances. By October 2003, about 3 percent of active air marshals were still awaiting their top-secret clearances. OPM attributed the delays to the governmentwide demand for security clearances after September 11, 2001. To deploy the necessary number of air marshal candidates by the Deputy Secretary’s deadline, the Service identified the skills critical for initial deployment and incrementally revised and abbreviated its training curriculum between October 2001 and July 2002. Then, to ensure that all newly hired air marshals were provided with training in advanced skills, the Service established an additional 4-week course and required all candidates hired after October 2001 to complete the training by mid-2004. It is unclear how the Service’s transfer to the Bureau of Immigration and Customs Enforcement will affect this requirement. Page 3 GAO-04-242 Federal Air Marshal Service The Service has begun to develop management information, policies, and procedures to support its expanded mission and workforce, but it has not yet completed this major effort. For example, it has replaced the manual system that it formerly used to schedule fewer than 50 air marshals for flight duty with an automated system that it can use to schedule thousands of air marshals on thousands of flights; however, it has not yet developed an automated means to monitor the effectiveness of controls designed to prevent overscheduling so that air marshals do not become fatigued. Preventing fatigue is important because, if air marshals are not alert, their ability to carry out their mission may be diminished. The Service has made effective use of some management information—by, for example, establishing a liaison with the airlines after air marshals’ mission reports indicated problems with coordination and communication. However, supervisors’ summaries of air marshals’ reasons for separating—as of July 2003, about 10 percent of newly hired marshals had separated—are not detailed enough for management to identify and respond to specific problems, such as dissatisfaction with the Service’s transfer policy. The Service initially lacked a means of obtaining input from its employees for use in improving its operations and management, but it has started to implement such processes. Finally, the Service has implemented policies and procedures needed to support its expansion from a single office with a budget of about $4.4 million in fiscal year 2001 to an organization with 21 field offices and a budget of $545 million in fiscal year 2003. For example, it implemented a policy on transfers between field offices and issued a written dress code policy. We are recommending that the Service automate the comparison of actual hours worked with scheduled hours to monitor the effectiveness of its scheduling controls and develop improved information on air marshals’ reasons for separation. The Department of Homeland Security agreed that information on actual hours worked should be automated and acknowledged a need to improve the quality of the information the Service collects from departing air marshals. The Service is likely to face challenges in implementing changes resulting from its mergers into the new Department of Homeland Security in March 2003 and into the Bureau of Immigration and Customs Enforcement in November 2003. While the new department, within the context of guidance from the Congress and the administration, is primarily responsible for determining what changes will occur, the Service is responsible for implementing them. At this time, changes are likely in the roles, responsibilities, and training of air marshals, assuming that the Secretary of Homeland Security’s plan to cross-train the Bureau’s three law enforcement workforces is implemented so that each group can perform the others’ responsibilities. Developing procedures for coordinating with Page 4 GAO-04-242 Federal Air Marshal Service TSA will also be important to help ensure a comprehensive, unified approach to aviation security, now that the Service is separated organizationally from the other groups with aviation security functions in TSA. Finally, changes will be needed to resolve differences in the pay systems and compensation of air marshals, immigration agents, and customs agents. Our recent work on mergers and organizational transformations proposes several key practices—setting implementation goals and a time line to build momentum and show progress from day one, establishing a communication strategy to create shared expectations and report related progress, and involving employees to obtain their ideas and gain their ownership for the transformation—and associated implementation steps that can assist the Service as it addresses the challenges in merging into the department and the Bureau. In an earlier report, we recommended these and other key practices to the department.3 FAA’s Federal Air Marshal program expanded the Sky Marshal program, Background which was established as part of the Customs Service in the 1970s to deter hijackings to and from Cuba. Shortly after TWA Flight 847 was hijacked in Athens, Greece, in June 1985, then President Ronald Reagan called for an expansion of the Sky Marshal program. On August 8, 1985, the Congress enacted the International Security and Development Cooperation Act,4 which established the statutory basis for the program within DOT, which further delegated the responsibility to FAA.5 Since then, the Federal Air Marshal program has provided specially trained, armed teams of civil aviation security specialists for deployment worldwide on antihijacking missions. As a result of the events of September 11, 2001, the President and the Congress decided to rapidly expand the Service. On September 17, 2001, FAA began to develop a plan to recruit federal air marshals in unprecedented numbers. Accordingly, FAA designed a process and put together a team of specialists, incorporating resources from its Human Resource Management, Aviation Medical, Civil Aviation Security, and Federal Air Marshal Training organizations to implement the recruitment 3 U.S. General Accounting Office, Homeland Security: Management Challenges Facing Federal Leadership, GAO-03-260 (Washington, D.C.: Dec. 20, 2002). 4 Public Law 99-83, August 8, 1985. 5 49 C.F.R. Sec. 1.47(p)(1). Page 5 GAO-04-242 Federal Air Marshal Service process. The process was designed to ensure that each air marshal candidate met the medical entry standards, passed DOT’s drug-testing program, and was preliminarily judged suitable to obtain a top-secret clearance, which is required for permanent employment with the Service. As part of the assessment, each candidate was required to participate in a security interview with an investigator from FAA, OPM, or the U.S. Investigative Services (an OPM contractor6), as well as interviews with representatives of FAA’s Office of Human Resource Management and the Service. In October 2001, FAA implemented this recruitment process, and the Deputy Secretary of Transportation also set July 1, 2002, as the deadline for recruiting, hiring, and training enough federal air marshals to provide coverage on flights that posed high security risks. In November 2001, after the Aviation and Transportation Security Act was passed, TSA assumed FAA’s responsibilities for aviation security and supported FAA’s recruitment effort through July 2002. Between October 2001 and July 2002, TSA received nearly 200,000 applications for federal air marshal positions. Thousands of applicants were assessed for employment, and TSA, through OPM, initiated full background investigations for top-secret clearances. Other federal agencies also made law enforcement officers available to augment the Service until TSA could hire, train, and deploy the first few classes of new air marshals. See appendix II for a demographic profile of the Service’s expanded workforce. With expansion, the Service’s annual budget grew from $4.4 million for fiscal year 2001 to $545 million for fiscal year 2003. Currently, the Service operates a headquarters office in Virginia, 21 field offices, and a specialized air marshal training and human resource facility in Atlantic City, New Jersey. Basic law enforcement training takes place at the Federal Law Enforcement Training Center in Artesia, New Mexico. See appendix III for a map of these facilities and appendix IV for a time line of the major organizational events affecting the Service since September 11, 2001. DHS brings together some 23 federal agencies comprising over 100 organizations, including the Federal Air Marshal Service, in what the department describes as the most significant transformation of the U.S. 6 OPM contracts primarily with U.S. Investigative Services to check the applicants’ personal records and often to conduct face-to-face interview with friends, colleagues, and family. Page 6 GAO-04-242 Federal Air Marshal Service government since the merger in 1947 of the various branches of the armed forces into the Department of Defense. DHS is divided into five directorates, one of which, Border and Transportation Security, includes both TSA and ICE. Among other organizations, ICE includes a portion of the former Immigration and Naturalization Service, now called the Bureau of Citizenship and Immigration Services; the U.S. Customs Service now called Customs and Border Protection; and, as of November 2, 2003, the Federal Air Marshal Service. To expedite the deployment of thousands of air marshals, the Service Expediting obtained preliminary background checks and provided abbreviated Background Checks training before deploying air marshal candidates on flights. As a result, the Service was able to meet the Deputy Secretary’s deployment deadline and and Training Enabled carry out its mission. the Service to Meet the Deputy Secretary’s Deployment Deadline Initial Deployment Was To deploy its expanded workforce as quickly as possible between October Swift, but Completion of 2001 and June 2002, the Service followed the same expedited background Final Background check procedures that federal agencies have used since 1995, when Executive Order 12968 authorized the temporary use of interim security Investigations Has Been clearances.7 Under these procedures, candidates who require security Slow clearances and pass preliminary background checks may, within about 24 hours, obtain interim security clearances that allow them to work until their full background checks have been completed and they obtain their final clearances. A preliminary background check consists of an interview with a security specialist; a review of an applicant’s responses to a standard questionnaire for national security positions; a criminal history check, based on fingerprints and a review of biographical data from 7 Executive Order 12968, dated August 2, 1995, authorizes agencies to grant employees temporary eligibility for access to classified information while the initial investigation is under way. When such eligibility is granted, the initial investigation shall be expedited. Page 7 GAO-04-242 Federal Air Marshal Service National Crime Information Center files;8 and credit reports. An interim security clearance may be revoked at any time if unfavorable information is identified during an investigation. Between October 2001 and July 2002, thousands of candidates were assessed for employment, and TSA, through OPM, initiated full background investigations for top-secret clearances. According to TSA management, the majority of the candidates passed the preliminary background checks and obtained interim security clearances that allowed them to work while their full background checks were being completed. Less than a quarter of the candidates did not pass the preliminary checks because of bankruptcy, bad credit, or other problems. TSA placed these candidates on a “pending/ready” list and did not allow them to work as air marshals, but it pursued full background investigations for them because many of the issues identified during preliminary background checks are minor and are favorably resolved during full background investigations. Full background checks for thousands of candidates identified a small number as unsuitable. In June 2003, the Service placed 80 air marshal candidates on administrative leave while TSA resolved issues that surfaced during full background investigations. By August 2003, 47 of these candidates had received their top-secret clearances and have since been returned to flight status. Of the 33 remaining candidates, 19 have been denied clearances, and the Service is taking steps to terminate their employment; 4 have been approved for, but have not yet received, top- secret clearances; 7 have resigned; and the remaining 3 are awaiting TSA’s approval of their top-secret clearances. The Service said it has continued to identify some candidates as unsuitable, and as of October 2003, 14 air marshals were on administrative leave because of issues that surfaced during full background checks. When definitive information for each of these cases is obtained, the Service said, the air marshal would be returned to flight status or steps would be taken to terminate the air marshal’s employment. During our review, we found that the background investigations used to grant top-secret clearances for air marshals were not being expedited as 8 The National Crime Information Center is a computerized index of criminal justice information (i.e., information on criminal record histories, fugitives, stolen properties, missing persons) located at Federal Bureau of Investigation’s (FBI) Criminal Justice Information Services Division in Clarksburg, West Virginia. It is available to federal, state, and local law enforcement and other criminal justice agencies 24 hours a day, 365 days a year. Page 8 GAO-04-242 Federal Air Marshal Service requested. According to TSA, an expedited background investigation costs $3,195 and should be completed within 75 days, whereas a regular background check costs $2,700 and should be completed within 120 days. Consequently, for every 1,000 background investigations, the Service paid a premium of about $495,000. TSA paid the expedited fees to OPM up front, as required,9 but as of July 2003, about 23 percent of the air marshals were still operating under interim security clearances.10 Some candidates had been awaiting final clearances for up to a year. The Service told us in April 2003 that it had, on numerous occasions, raised concerns about the delays in processing final security clearances but had met with little success. Additionally, the Service said that its efforts to reclaim the difference in cost were unsuccessful. DHS said that TSA’s Credentialing Office had taken steps since June 2003 to ensure that every active air marshal was operating under a top-secret clearance; and as of October 2003, about 3 percent of the active air marshals were operating under interim security clearances. According to OPM, the primary reason for these clearance-processing delays is that the agency has received an unprecedented number of requests for background investigations governmentwide since September 2001.11 For fiscal year 2002, OPM’s data indicated that the average processing time for 75-day expedited background checks was 96 days. OPM said that the expedited requests received higher-priority processing than the regular (120-day) background checks, resulting in faster turnaround for services related to the expedited requests. OPM added that its contractor charges premiums for expedited requests because the costs for these requests are higher. Consequently, according to OPM, no price adjustments are made when overall deadlines are missed. While the Service is not responsible for the delays in completing air marshals’ full background investigations, we found that it could have 9 OPM bills an agency for the full amount of an investigation at the time the investigation is scheduled. 10 TSA’s Credentialing Program Office is responsible for adjudicating the results of air marshals’ background investigations. This function was formerly under TSA’s Office of Security. 11 In his statement on June 3, 2003, before the Subcommittee on Homeland Security, House Committee on Appropriations, the Associate Director for Human Resources Products and Services, said that OPM was working to increase its capacity to provide background investigations for all federal customers and had streamlined its internal processes to make as much use as possible of automated systems. Page 9 GAO-04-242 Federal Air Marshal Service provided OPM with information for scheduling the investigations more efficiently. As candidates applied for positions between October 2001 and June 2002 and their preliminary background checks were completed, the Service offered conditional employment to some of the candidates and, as discussed, placed others on a “pending/ready” list. However, the Service did not make this information available to OPM. As a result, some potential employees received their top-secret clearances ahead of other candidates who were being trained or deployed on flights.12 We brought this issue to the attention of the Service in March 2003; and in May, the Service sent OPM a list of candidates and asked OPM to give highest priority to investigations of those who were already deployed on flights. In addition, the Service has asked OPM to schedule the investigations for senior managers first and then to schedule investigations for other applicants on a first-in, first-out basis. On May 28, 2003, the Service also detailed a liaison from its Office of Field Operations to assist TSA’s Office of Security in setting priorities for reviewing and adjudicating the backlog of background investigations. Changes to the Training To deploy the requisite number of air marshals by the Deputy Secretary’s Curriculum Helped July 2002 deadline, the Service revised and abbreviated its training Expedite Deployment program. From October 2001 through July 2002, it modified the air marshal curriculum incrementally, eventually reducing the original 14-week program to about 5 weeks for candidates without prior law enforcement experience and about 1 week for candidates with such experience. The revised curriculum was designed to provide candidates with the basic law enforcement knowledge, skills, and abilities needed to perform their duties as air marshals, including knowledge of the Service’s rules and regulations, physical skills, and basic and advanced marksmanship. The curriculum no longer included certain elements of the original training program, such as driving skills and cockpit familiarization, because these were not deemed critical for air marshals to perform their duties. The curriculum also eliminated a 1-week’s visit to an airline and some instruction in the Service’s policies and procedures, which was to be provided on the job. Moreover, although the curriculum retained instruction in both basic and advanced marksmanship, air marshal candidates no longer had to pass an advanced marksmanship test to 12 OPM told us in March that it had not received a prioritized list of clearances from the Service, but noted that it had occasionally received requests to expedite or check the status of particular investigations or to discontinue investigations that were no longer needed. Page 10 GAO-04-242 Federal Air Marshal Service qualify for employment. Candidates were still required to pass a basic test13 with a minimum score of 255 out of a possible 300—the highest qualification standard for any federal law enforcement agency, according to the Service. To provide all the newly hired air marshals with needed skills, beyond the basic abilities the Service determined were critical for immediate deployment, the Service instituted a new 4-week advanced training course in October 2002. All air marshals hired from October 2001 through July 2002, regardless of their previous law enforcement experience, were required to complete the course by January 2004. This course includes some elements, such as emergency evacuation and flight simulator training, that the Service did not include in the 5-week course because, although it considered the elements important for air marshals to carry out their mission, it did not consider them critical for immediate deployment. In addition, the course provides further training in advanced marksmanship skills. Air marshals hired after August 2002 attend this advanced training course after completing their basic training. The Service has developed a centralized tracking system to ensure that all air marshals take this course. Although the Service is now providing additional marksmanship training, its decision not to restore the advanced marksmanship test14 as a qualification standard for employment has proved controversial. Passing this test would require candidates to demonstrate their speed and accuracy in a confined environment similar to the environment on board an aircraft. The DOT IG’s report suggested that the Service needed to adopt a firearms qualification standard that was more stringent and comprehensive than the basic firearms qualifying test. The Service disagreed, emphasizing that its minimum score is the most stringent in federal law enforcement and adding that its 4-week course provides further training in advanced firearms skills. Our review of the Service’s documentation confirmed that instruction in advanced marksmanship is a critical part of this training, even though passing this element is no longer a condition of employment. In August 2003, the Service reported that proposed cutbacks in its training funds would require it to extend the date for all air marshals hired from 13 The federal law enforcement Practical Pistol Course (PPC). 14 The Aircraft Tactical Pistol Course (ATPC). Page 11 GAO-04-242 Federal Air Marshal Service October 2001 through July 2002 to complete the 4-week advanced course from January 2004 to mid-2004. According to DHS, the Service’s transfer to ICE will not adversely affect either the funding for air marshals’ training or the schedule for newly hired air marshals to complete the 4-week training course, since a total of $626.4 million is being transferred from TSA to ICE. While this funding exceeds the $545 million that the Service received for fiscal year 2003, it is not clear how much of the funding will be allocated for training. Given the importance of training to ensure that air marshals are prepared to carry out their mission, we believe that maintaining adequate funding for training should remain a priority. Additionally, should reductions in the funding for training be required, our recent work on strategic training and development efforts provides alternatives that an agency can consider to across-the-board cuts—such as evaluating training needs, setting training priorities, developing alternative training requirement scenarios, and determining how much funding each of these scenarios would require.15 Our work further suggests that it is important for agencies to ensure that their training and development efforts are cost effective, given the anticipated benefits and to incorporate measures that can be used to demonstrate contributions that training and development programs make to improve results. These principles are applicable at all times, but especially when funds are limited. Determining whether air marshals with prior law enforcement experience have the same training needs as those without such experience could help set cost-effective training priorities. We found that a cornerstone of human capital management is the ability to successfully acquire, develop, and retain talent. Investing in and enhancing the value of employees through training and development is a crucial part of addressing this challenge. This investment can include not only formal and on-the-job-training but also other opportunities, such as rotational assignments. Our work further specifies that agencies should link their training curriculum to the competencies needed for them to accomplish their mission. The Service has begun developing a formal training curriculum beyond the basic and advanced training courses described above. This curriculum requires air marshals to participate in 5 days of recurrent training each quarter that, in addition to the quarterly weapons qualification, includes training in advanced firearms, operational tactics, 15 U.S. General Accounting Office, Human Capital: A Guide for Assessing Strategic Training and Development Efforts in the Federal Government, GAO-03-893G (Washington, D.C.: July 2003). Page 12 GAO-04-242 Federal Air Marshal Service defensive tactics, surveillance detection, emergency medicine, physical fitness, and legal and administrative elements. Additionally, the Service is developing rotational assignments for air marshals that allow them to participate in law enforcement task forces, as well as fill a variety of operational and training positions in headquarters and the field. The Service recognizes that such opportunities can not only enhance professional development but also help to prevent problems such as boredom and burnout. According to the Secretary of Homeland Security, one of the advantages of the Service’s transfer to ICE is that it will enhance air marshals’ professional development opportunities. As the Service grew from a small, centralized organization to an Management organization with 21 field offices and thousands of employees, its need for Information, Policies, information, policies, and procedures to manage its expanded workforce and operations also grew. The Service collects several types of and Procedures Have information that it can use to continually improve its operations and Not Kept Pace with oversight and, in some instances, it has used the information to do so. In other instances, however, the Service lacks sufficiently detailed Expanded Operations information for effective monitoring and oversight. The new, decentralized organization has also required new or written policies and procedures to cover new situations and ensure that the same guidance is available to air marshals in all locations. According to DHS, it recognized that the Service would need to revise its existing policies16 or draft new ones, and it has been working to do so since March 2002. Nonetheless, its policy- development efforts sometimes responded to problems, rather than anticipating and preventing them. DHS told us that it is committed to proactively addressing policy issues and developing procedures. Management Information The Service collects information on air marshals’ work schedules and Is Not Sufficiently Detailed other issues, including potential security incidents documented in reports or Comprehensive for filed by air marshals after completing their missions, allegations of misconduct by air marshals, and reasons provided by air marshals for Effective Monitoring and leaving the Service. Such information can be useful to managers in Oversight monitoring mission operations and retention. According to our Standards for Internal Control in the Federal Government,17 the information should 16 Before being transfered to TSA in March 2002, the Service continued to follow standard operating procedures designed for a small organization with one facility. 17 U.S. General Accounting Office, Standards for Internal Control in the Federal Government, GAO/AIMD-00-21.3.1 (Washington, D.C.: November 1999). Page 13 GAO-04-242 Federal Air Marshal Service be recorded and communicated to management and others within the agency who need it, and it should be provided in a form and within a time frame that enables them to carry out their responsibilities. For example, one way to do this would be to ensure that pertinent information is captured in sufficient detail to help management identify specific actions that need to be taken. Moreover, according to our human capital model, a fact-based, performance-oriented approach to human capital management is a critical success factor for maximizing the value of human capital. In addition, high-performing organizations use data to determine key performance objectives and goals, which enable them to evaluate the success of their human capital approaches. For example, obtaining employee input and suggestions can provide management with firsthand knowledge of the organization’s operations, which management can use to ensure ongoing effectiveness and continuous improvement. The Service has analyzed and made effective use of its mission reports and conduct data, but other management information that it currently collects is not sufficiently well defined or detailed for monitoring and managing the workforce. Although the Service initially had no systematic means of obtaining regular input from its employees, it has recently put processes in place to solicit air marshals’ opinions and suggestions. In addition, the Service is participating in an Office of Management and Budget program assessment project. As part of this effort, DHS said it has identified annual and long-term performance measures and related performance outcome targets to evaluate the Service’s organizational effectiveness along key strategic goals and objectives. Through this project and other strategic planning initiatives, DHS says it expects to systematically measure and analyze the Service’s organizational performance along human capital, mission scheduling, professional development, and quality of work-life dimensions. Automated System Improved When the Service was first directed to expand its mission and operations, Scheduling, but More it was using a manual system to schedule air marshals for flight duty. This Information Is Needed for system was quickly overwhelmed as the number of air marshals and flights Monitoring grew, leading to the concern that air marshals were being scheduled inconsistently for flight duty. The Service acknowledged that during this period, some air marshals were overworked while others were underutilized. In June 2002, the Service replaced the manual system with an automated system, which, according to Service officials, improved the agency’s ability to schedule and deploy its workforce. While the automated system expanded the Service’s scheduling capability, it did not provide the Service with all of the information it needed for effective monitoring. For example, it did not initially break down data on Page 14 GAO-04-242 Federal Air Marshal Service air marshals’ use of leave into enough categories for the Service to assess whether some air marshals were abusing sick leave in order to get a day off. Specifically, an article in USA Today reported that about 1,250 air marshals called in sick over an 18-day period. Eventually, the Service determined that the article was based on a report generated by the automated scheduling system that overrepresented the number of air marshals who were on sick leave. Although the report was labeled “Sick Leave,” it included data on all air marshals who were unavailable for flight duty, not only for sickness but also for other reasons such as administrative leave, and it listed each day of unavailability for flight duty as a separate incident, although the same air marshal might have been unavailable for several days in a row for the same reason. In analyzing data from the scheduling system, we found that because the system reported all leave charges—sick, administrative, military, or other—as sick leave, the Service could not distinguish air marshals who were unavailable to fly because they were out sick from air marshals who were unavailable to fly because of injuries but were available for light field office duty. For example, an air marshal with an injured ankle might not be able to fly, but could perform administrative work in the field office. The Service has since modified the scheduling system to obtain better information on the type of leave—sick, military, or administrative— charged by air marshals who are unavailable to fly. The DOT IG also investigated cases concerning sick leave abuse and likewise found that it was based on a misunderstanding of the report’s contents stemming from the report’s label. Although the automated scheduling system provides information for managers to monitor how many hours air marshals are scheduled for work, automated information is not available for comparing the number of hours actually worked with the number of hours initially scheduled. These numbers can differ when flights are delayed or cancelled because of bad weather or mechanical problems. Information on these differences is important for Service managers to consider because of their implications for both the Service’s mission and air marshals’ quality of life. For example, if air marshals work too many hours, they may become too tired to concentrate on their mission, or if they spend too much time away from home, they may become dissatisfied with their jobs. Information on the number of hours flown will also be important for the Service to carry out a new long-term study, initiated by the Director in the summer of 2002, on the medical and physiological effects of flying. To date, the Service, in collaboration with FAA’s Civil Aviation Medical Page 15 GAO-04-242 Federal Air Marshal Service Institute and the Air Force, has identified a methodology and objectives for the study and completed a literature review to identify trends, possible risks, and other pertinent information. As part of the study, the Service plans to collect and analyze data from recurrent air marshal physical examinations and to compare these data with physiological data from the Civil Aviation Medical Institute. Although the Service is still awaiting funding approval to conduct the physical examinations and develop the database, Service officials plan to begin both efforts in the first quarter of fiscal year 2004. The study team has also developed a training course on human physiology as it relates to the aviation environment. The Service expects this course to be available early in fiscal year 2004. On the basis of some early findings from the study team’s literature search, the Service set limits in its automated flight-scheduling system to address mission, quality-of-life, and health concerns. The system limits scheduled “duty time” to 10 hours a day or 50 hours a week.18 Our analysis of schedules from the automated system for 37 weeks found that about 92 percent of the schedules were consistent with these controls.19 The Service added that further guidance has been developed that results in scheduling air marshals to fly an average of 4.2 hours per day, 18 days per month. Thus, air marshals should fly about 75 hours per month, which the Service said was within the aviation and military standards for pilots—90 and 100 hours per month, respectively. As part of implementing this guidance, the Service is conducting a detailed analysis of individual flight schedules to determine if the goals are being met. The Service reported on the basis of this analysis that, as of September 2003, scheduled flight time averaged 76.5 hours per month. The Service’s analysis, however, focuses on flight schedules and not on actual hours worked by the air marshals. Information on the hours air marshals actually work is not available for automated comparison with the hours they are scheduled to work because the actual work hours are recorded manually on time and attendance sheets and are not transferred to the automated system. Without an automated way to compare actual hours worked with scheduled hours, the 18 According to an analysis done by the Service, air marshals spend an average of 4 hours and 25 minutes per workday in flight and use the remainder of the workday to prepare for flights or layovers. Air marshals must be at the airport 1.5 hours before their first flight and stay there 15 minutes after arrival. The combination of flight time and the aforementioned 1.5 hours and 15 minutes is referred to as “duty time.” 19 The remaining 8 percent, Service officials explained, could be due to inconsistencies that resulted when the Service overrode the controls to meet mission needs—to, for example, provide sufficient coverage for Super Bowl weekend. Page 16 GAO-04-242 Federal Air Marshal Service Service lacks a tool needed to determine if the automated flight-scheduling system is meeting its objectives related to mission, quality-of life, and health concerns. DHS agreed that the information on actual hours should be automated and said that the Service intended to incorporate this capability via personal digital assistants (PDA) issued to all air marshals. The Service Has Used Mission Between September 2001 and September 2003, air marshals submitted Reports to Improve Operations, reports of almost 2,100 incidents that occurred during their missions. A but Some Coordination Issues little over 40 percent of these mission reports describe passengers that Remain exhibited suspicious behavior to the air marshals. About 18 percent of the reports discuss disagreements or conflicts between air marshals and airline or airport personnel over airport or airline procedures. The remaining mission reports cover a wide variety of incidents that the Service grouped into 17 other categories, as shown in appendix V. The Service has taken some action to follow up on the air marshals’ mission reports, but it has not addressed all of the issues the reports raise. For example, the Service established a liaison with the airlines in response to reports of disagreements and conflicts with the airlines. According to an official with the Air Transport Association,20 this action has improved relations between the air marshals and the airlines. Nevertheless, some coordination and communication issues remain. In October 2002, for instance, the Service purchased PDAs for distribution to all air marshals. Service officials told us that before making the purchase, they contacted FAA about obtaining approval to use the feature that would allow the air marshals to communicate with one another aboard aircraft. In August 2002, FAA advised the Service that it planned to approve this PDA feature for use by air marshals during flight. However, FAA’s approval was never finalized, and the airlines have not allowed the air marshals to use the PDAs for this purpose because of concerns about interference with flight control or navigational signals. According to Service officials, air marshals have stopped using their PDAs’ communication feature in flight until FAA approves its use, and the Service continues to work with FAA to obtain such approval. The Service reports that air marshals continue to use other features of the PDAs, such as their cell phone, pager, e-mail, surveillance, and photo-display capabilities. 20 The Air Transport Association is a trade association for 22 major U.S. airlines and five foreign carriers. Page 17 GAO-04-242 Federal Air Marshal Service The Service Has Used Reports Between October 2001 and July 2003, the Service collected data on almost of Misconduct to Improve 600 reports of misconduct by air marshals, which it classified into over 40 Management Controls categories. Among the categories with large numbers of reported cases were “insubordination or failure to follow orders,” “loss of government property,” and “abuse of government credit cards.” According to Service officials, they have used the misconduct database to identify issues such as abuse of government credit cards and cell phones that need to be investigated.21 For example, during the Service’s rapid expansion, management noted an unacceptable number of unauthorized charges and late payments associated with air marshals’ use of the government-issued travel card. Further investigation revealed that the process of claiming reimbursement for travel was slow and burdensome and there were misunderstandings about what charges were proper. After corrective action, the delinquency rate dropped dramatically. Similarly, an analysis of the misconduct data indicated that a number of air marshals were accused of being abusive to airline personnel during the boarding process. A detailed review of the data pointed to differences in the Service’s and the airlines procedures for boarding aircraft. Subsequently, the Service negotiated a mutually agreeable solution with the airlines to resolve these differences. In these instances, the Service used misconduct reports to effectively refine its management controls. Information on Air Marshals’ The Service maintains data on the number of air marshals who leave the Reasons for Leaving the Service and categorizes their reason for leaving.22 However, these data are Service Is Not Detailed Enough not detailed enough for management to identify and follow up on issues to Target Retention Efforts that could affect retention. Retention is important both to ensure the continued deployment of experienced personnel who can carry out the Service’s security mission and to avoid the costs to recruit, train, and deploy new personnel, which, according to the Service, total about $40,275 per person. Our analysis of the Service’s data on separations indicates that from September 2001 through July 2003, about 10 percent of the thousands of newly hired air marshals left the Service. However, during August 2002, when the media reported a “flood” of resignations from the Service, our 21 Generally, Service staff members in headquarters investigate reports of misconduct, but for more serious cases, the Service has been coordinating its investigations with TSA’s Office of Internal Affairs. 22 The Service selects and records one predominant reason for an air marshal’s separation from the Service, although the air marshals may have cited more than one reason. Page 18 GAO-04-242 Federal Air Marshal Service analysis indicated that slightly more than 4 percent of the newly hired air marshals had left.23 We found that the most frequently recorded reasons for air marshals separating from the Service were to take other jobs and personal reasons.24 Such reasons are not detailed enough for management to identify and target issues that may hinder retention. To gain greater insight into the reasons for separation, we examined the Service’s documentation for 95 selected cases.25 For 37 of these cases, the departing air marshals cited multiple reasons for leaving the Service. For example, one departing air marshal cited personal reasons and going back to his previous job. Even with this additional information, we could not identify management issues that might have led to the separations because the reasons documented by the Service were too general and vague. The Service’s method of collecting data on air marshals’ reasons for separation may be responsible, in part, for the generality and vagueness of the information. Specifically, the Service uses either an open-ended exit interview with the air marshal’s first-line supervisor, the air marshal’s resignation letter, or both to collect the data.26 The supervisor conducts and writes up the exit interview and an administrative official in the field forwards the interview write-up, resignation letter, or both to human resource officials in Service headquarters. A human resource specialist then reviews the documentation and determines which of the reasons cited is the primary reason for the separation. This method of collecting information has several limitations. First, the open-ended exit interview may not prompt responses that go beyond generalities, such as taking another job or personal reasons, to determine whether management issues, such as problems in transferring to a duty station closer to home or burdensome work schedules, contributed to the air marshal’s resignation. 23 Because TSA was a newly created agency without a workforce history (including, for example, information on deaths, retirements, transfers, and resignations), we were unable to meaningfully compare attrition data for the Service to other federal agencies during this period. Therefore we are not making a value judgment on the meaning of the number of persons leaving the Service or their rate of departure. However, these data are relevant to the resources that have to be expended to maintain a specified number of marshals in the Service. 24 A small number of air marshals left because they could not pass training. 25 The details of our selection process are provided in appendix I. 26 This documentation included an exit information form that the Service began using in 2002 to gather data from separating air marshals. Page 19 GAO-04-242 Federal Air Marshal Service Second, using the first-line supervisor to conduct the interview may discourage detailed responses, either because the air marshal may not want to reveal his or her concerns or reasons or because the supervisor may not want to report specific issues. Finally, using a human resource specialist to determine the primary reason for a separation means that the reason is filtered through another party rather than provided directly by the air marshal who is resigning. Our work on human capital has determined that feedback from exit interviews can guide workplace- planning efforts. If these exit interviews are constructed to collect valid and reliable data, they allow managers to spotlight areas for attention, such as employee retention. According to the DOT report, air marshals interviewed by the IG’s office were concerned about the way the air marshal program was being managed, which contributed to low morale in the Service. The air marshals the IG interviewed expressed dissatisfaction with the Service’s work schedules, aircraft boarding procedures, and dress code policy.27 The Service Initially Lacked, During the early stages of its expansion, the Service did not have but Has Recently Begun to processes or mechanisms in place to gather input and suggestions from its Implement, Processes for employees. Such processes and mechanisms would not only allow the Obtaining Employee Input Service to monitor air marshals’ concerns about management issues, as the DOT IG’s report also noted, but it would also provide the Service with its employees’ firsthand knowledge and insights that it could use to improve operations and policies. According to our work on human capital, leaders at agencies with effective human capital management seek out the views of employees at all levels and communication flows up, down, and across the organization, facilitating continuous improvement. Tools commonly used for obtaining employee input include employee satisfaction surveys, employee advisory councils, and employee focus groups. Recently, the Service began putting processes and mechanisms in place to gather input from its employees. The Service reports that all field offices have methods, such as advisory committees, for air marshals to ask questions or express concerns to senior field office management. Additionally, question and answer sessions are held when the Director, Deputy Director, or Assistant Director visits a field office and during the 27 As discussed in appendix I, the DOT IG’s sample results cannot be projected to the universe of the Service’s air marshal workforce. Page 20 GAO-04-242 Federal Air Marshal Service basic and advanced training classes. To obtain further employee input, the Service participated in an ombudsman program that TSA sponsors to improve its operations and customer service. According to the Service, it is also developing a lessons learned and best practices intranet site that will allow substantive communication on issues of interest and concern to all air marshals. Developing and Policies and procedures that were designed to support a small, centralized Implementing Policies and Service were not designed for and could not accommodate the needs of a Procedures for an vastly expanded and decentralized workforce. According to our Standards for Internal Control in the Federal Government,28 internal control should Expanded Organization provide for an assessment of the risks an agency faces from both external Took Time and Created and internal sources. For example, when an agency expands its operations Some Confusion to new geographic areas, it needs to give special attention to the risks that the expansion presents. In attempting to hire, train, and deploy its new workforce by the Deputy Secretary’s deadline and establish a new field organization to support its new domestic mission, the Service had little time to systematically assess the risks of expansion and ensure that its policies and procedures were appropriate and adequate. Efforts to develop new policies or modify existing ones to accommodate new circumstances took time, and during the transition, some air marshals voiced concerns to the media. Delays in hiring supervisors and the discretion they were given in interpreting policies may have contributed to air marshals’ confusion. Policy on Transfers Was Not Before its expansion, the Service was a centralized organization with one Implemented until May 2003 office and fewer than 50 air marshals. Because there were no field offices, the Service had no policy on transfers between field locations. The vacancy announcement used during the hiring process stated that field offices would be located in various major metropolitan areas, and a Service official stated that air marshal applicants were allowed to express their preferences for particular field locations. According to a media report, air marshals alleged that transfers to their preferred locations were promised but that those promises were not kept. Our review of a recruiting video and other documents related to the hiring process did not find any evidence that transfers were promised; however, the recruiting video indicated that opportunities for transfer existed. Service officials said that no transfers were promised and that as the Service hired air marshals and implemented its new field office structure, it assigned the 28 GAO/AIMD-00-21.3.1. Page 21 GAO-04-242 Federal Air Marshal Service newly hired marshals to the 21 field offices as needed. Service officials later added that except in hardship cases, the air marshals were expected to remain in the originally assigned field offices for 3 to 5 years. The DOT IG also investigated this issue and interviewed air marshals who alleged that promises of transfers made during the hiring process were not kept, but the IG did not determine whether there was a legitimate basis for the air marshals’ concerns. By June 2002, the Service had received over 500 applications for transfers. Until a policy was issued, the Service tried to respond to the air marshals’ requests and to address quality-of-life issues by developing guidance that provided for transferring any air marshal (1) who owned a home within 100 miles of an established field office and (2) whose immediate family resided in that location—provided that both of these conditions existed before the air marshal’s employment with the Service. While the Service communicated this guidance orally to field managers, some air marshals were reportedly confused about why their requests for transfers were denied. In January 2003, the Service postponed further action on transfer requests, officials said, until applicable policies—on hardship and transfers—were finalized. On May 29, 2003, the Service implemented a hardship transfer policy that established processes and criteria for approving transfer requests when an air marshal or an immediate family member incurs a medical or child-custody hardship. In developing the policy, the Service said it looked into other law enforcement agencies’ transfer programs to identify best practices. Unwritten Policy on Dress During the early months of the Service’s expansion, air marshals Created Confusion expressed confusion and dissatisfaction to the media about policies covering their attire. At that time, the Service had no written dress code policy. Instead, according to Service officials, the agency carried over an unwritten FAA policy that air marshals should dress appropriately for their missions and the air marshal in charge of a mission should determine what attire was appropriate for that mission. According to the Service, some airline personnel complained to the Service that marshals were not dressed to blend in with other passengers in the location of the air marshals’ assigned seats. The Service said that the lack of a written policy might have created confusion for some newly hired air marshals whose initial training did not cover the Service’s policy on dress and whose field office supervisors had discretion in interpreting the policy. In May 2002, the Service issued a policy that directed air marshals to dress so as to present a professional image and blend into their environment. The Page 22 GAO-04-242 Federal Air Marshal Service Service believes that this policy enables air marshals to perform their duties without drawing undue attention to themselves. For example, an air marshal might wear a business suit on a morning flight to New York and a sports shirt on an afternoon flight to Orlando. To explain and ensure consistent application of the policy, the Director of the Service discussed this policy with supervisors and staff during his visits to many field offices and to the Service’s training center. Air Marshals Expressed Air marshals also discussed concerns about the Service’s workweek policy Concern about Reasons for with the media. Some air marshals complained that they had been Changes in Workweek Policy promised 4-day workweeks to compensate for the rigors of travel but were being required to work 5-day workweeks. Other air marshals reported being confused about the reasons for the change from a 4-day to a 5-day workweek and questioned whether this change was necessary. According to Service officials, the change in workweek policy occurred on July 1, 2002, when the Director of the Service brought the air marshals into compliance with the requirements of law enforcement availability pay (LEAP), a pay premium for unscheduled duty equaling 25 percent of a law enforcement officer’s base salary. Under this pay computation method, air marshals are required to average 10 hours of overtime per week. LEAP became applicable to the Service with the passage of the Aviation and Transportation Security Act on November 19, 2001, but the Service initially continued to compute air marshals’ schedules according to the method it had previously used, called the “first forty” method. Under this method, the first 40 hours worked in a week constituted the basic workweek, and 4-day and even 3-day workweeks were allowed if air marshals accrued 40 hours within that time. However, Service officials determined, in consultation with TSA’s legal department and human resources office, that a change to a 5-day workweek was necessary for the Service to comply with LEAP. Accordingly, the Director ordered a 5-day basic workweek, effective July 1, 2002. The DOT IG reported that over 85 percent of the air marshals its staff interviewed expressed concern about working 5 consecutive 10-hour mission days (with 2 consecutive off-duty days), saying that it resulted in fatigue and illness.29 Service officials acknowledged that working 10-hour 29 Between November 2002 and February 2003, 112 air marshals were interviewed. However, because of the methodology employed, the results are anecdotal and may not reflect the views or experiences of all Service employees. Page 23 GAO-04-242 Federal Air Marshal Service days can create fatigue,30 but said that such days are routine in the law enforcement community. Service officials also maintained that fatigue can be managed by applying scheduling controls and monitoring air marshals’ schedules. However, as noted, the Service lacks the data to ensure that air marshals’ actual work hours are consistent with the scheduling controls. The Service is likely to face challenges in implementing changes resulting Key Practices and from its mergers into DHS in March 2003 and into ICE in November 2003. Implementation Steps While changes in the size of its workforce could eventually occur in light of the many recent improvements to aviation security and federal budget Can Help Address constraints, the plans announced to date point to changes in the roles, Merger responsibilities, and training of ICE’s workforces; the Service’s coordination with TSA and other organizations; and administrative Implementation matters. DHS reported looking forward to the opportunities accompanying Challenges the Service’s pending merger into ICE. Our recent work on mergers and organizational transformations proposes several key practices and implementation steps that could assist the Service and other departmental organizations as they face these challenges. The Service Is Likely to One challenge for the Service will be to implement any changes in the size Face Challenges in or in the roles and responsibilities of its workforce that the department Implementing Changes determines are warranted after the Service is transferred to ICE. The right size of a security organization’s workforce appears to depend, among other things, on the nature and scope of the terrorist threat and on the totality of measures in place to protect against that threat. When the Service was first directed to expand, there were fewer protective measures in place than there are today. Over the past 2 years, an entire “system of systems” has been established for aviation security alone. This system includes not only the expanded Federal Air Marshal Service, but also about 50,000 federal security screeners in the nation’s airports, 158 airport security directors, explosives detection equipment for passengers and baggage, requirements for performing background checks on about 1 million airline and airport employees, reinforced cockpit doors on all passenger aircraft, and authorization for pilots to carry guns in the airplane cockpit. Now, as the department assesses the nation’s homeland security risks, considers the constraints on federal resources, and sets 30 As previously discussed, the 10-hour workday includes the time that air marshals are required to be in the airport before and after a flight as well as the time they spend in flight. Page 24 GAO-04-242 Federal Air Marshal Service priorities, it will need to determine its appropriate size. It has already begun to make changes in the federal security screener workforce by reducing the total number of full-time screeners by 6,000 in fiscal year 2003 and by planning a further reduction of 3,000 full-time screeners in fiscal year 2004 together with the hiring of part-time screeners to meet daily and seasonal periods of higher demand. In announcing the Service’s merger into ICE, the Secretary of Homeland Security did not propose a change in the size of the Service’s or of ICE’s other two law enforcement workforces, but his statement pointed to an expansion of their roles and responsibilities that would give the department greater flexibility to adjust its law enforcement resources according to varying threats. Through cross-training, the Secretary said, thousands more law enforcement agents would be available for deployment on flights, providing a surge capacity during times of increased aviation security threats. At the same time, air marshals may be assigned to other law enforcement duties, as threat information dictates. This planned expansion of the roles and responsibilities of air marshals, immigration agents, and customs agents will pose training challenges for ICE and its component organizations. According to the Secretary’s announcement, the training will be centralized, which could eventually produce some cost efficiencies, but initially a needs assessment will have to be conducted to identify each law enforcement workforce’s additional training requirements. Cross-training requirements and curriculums will also have to be established and approved. Finally, each component organization will have to coordinate the new training requirements with its other mission requirements and schedule its officers for the cross-training. The Service is also likely to face coordination challenges following its transfer from TSA to ICE. In part, the transfer is designed to improve coordination by unifying DHS’s law enforcement functions, but it also divides aviation security responsibilities that, for about 2 years, were under TSA. According to the Secretary, the transfer will facilitate the coordination and sharing of law enforcement information, thereby enhancing aviation security. However, TSA has raised questions about how air marshals’ flights will be scheduled, and the TSA Administrator has expressed a desire to influence the scheduling. Immigration agents have reportedly also wondered how ICE would juggle air marshal deployments with the bureau’s current investigative work. Finally, the Service’s transfer to ICE poses administrative challenges for the three component organizations. For example, the planned changes in Page 25 GAO-04-242 Federal Air Marshal Service the roles and responsibilities of the federal law enforcement officers could have implications for their performance evaluations and compensation. Currently, the three groups of law enforcement officers are under different pay systems and are compensated at different rates. Efforts are under way to resolve these challenges. Key Practices and On the basis of our work on mergers and organizational transformations, Implementation Steps Can we identified nine key practices and 21 implementation steps that could Help the Service Address assist DHS in successfully merging the roles, responsibilities, and cultures of the Service and the department’s other component organizations.31 Potential Merger While these practices will ultimately be important to a successful merger Challenges and we have previously recommended them for the department,32 there are three, we believe, that are particularly applicable to the Service, given the concerns about communication and other allegations reported in the media. These three practices emphasize communicating with employees and obtaining and using their feedback to promote continuous improvement. See appendix VI for a complete listing of the practices and implementations steps. One key practice in a merger or transformation is to set implementation goals and a time line to build momentum and show progress from day one. These goals and the time line are essential to pinpoint performance shortfalls and gaps and suggest midcourse corrections. Research indicates that people are at the heart of successful mergers and transformations. Thus, seeking and monitoring employee attitudes and taking appropriate follow-up actions is an implementation step that supports this practice. Our work suggests that obtaining employee input through pulse surveys, focus groups, or confidential hotlines can serve as a quick check of how employees are feeling about large-scale changes and the new organization. As discussed in this report, the Service did not initially have such tools in place—in large part because of the enormous demands it faced in recruiting, training, and deploying thousands of air marshals by the Deputy Secretary’s deadline—and it was not monitoring employee attitudes. Furthermore, although monitoring provides good information, it is also important for agency management not only to listen to employees’ concerns but also to take action. By not taking appropriate follow-up 31 Results-Oriented Cultures: Implementation Steps to Assist Mergers and Organizational Transformations, GAO-03-669 (Washington, D.C.: July 2, 2003). 32 GAO-03-260. Page 26 GAO-04-242 Federal Air Marshal Service actions, negative attitudes may translate into actions such as employee departures—or, as was the case with the Service, complaints to the media. Identifying cultural features of merging organizations is another important step in setting implementation goals. Cultural similarities between the Service and the other organizations within ICE could facilitate the Service’s merger into ICE. As the Director of the Service and others have noted, air marshals, immigration agents, and customs agents are all law enforcement officers and share a common culture. Moreover, as a spokesperson for ICE pointed out, many air marshals came to the Service from Customs and the Immigration and Naturalization Service; and some other agents served as air marshals temporarily after September 11. Establishing a communication strategy to create shared expectations and report related progress is another key practice in implementing a merger or transformation. According to our work on transformations and mergers, communication is most effective when it occurs early, clearly, and often and when it is downward, upward, and lateral. Organizations have found that a key implementation step is to communicate information early and often to build trust among employees as well as an understanding of the purpose of planned changes. As the Service found when modifying its workweek policy to implement LEAP premium pay, the absence of ongoing communication can confuse employees. Two-way communication is also part of this strategy, facilitating a two-way honest exchange with, and allowing for feedback from, employees, customers, and stakeholders. Once this solicited employee feedback is received, it is important to consider and use it to make appropriate changes when implementing a merger or transformation. Involving employees to obtain their ideas and gain their ownership is a third key practice for a successful transformation or merger. Employee involvement strengthens the transformation process by including frontline perspectives and experiences. A key implementation step in this practice is incorporating employee feedback into new policies and procedures. After obtaining sufficient input from key players, the organization needs to develop clear, documented, and transparent policies and procedures. Not having such policies and procedures was an impediment to the Service as it expanded, creating confusion about issues such as transfers and dress codes. DHS said that it fully recognizes the value and importance of communicating with employees and of obtaining and using their feedback to promote continuous improvement. It further noted that as the Service merges into ICE, it is committed to involving employees to obtain their opinions and gain their ownership. Page 27 GAO-04-242 Federal Air Marshal Service The rapid expansion of the Service’s mission and workforce posed Conclusions significant challenges, many of which the Service has begun to address. In the 2 years that have elapsed since the terrorist attacks of September 11, the Service has deployed thousands of new air marshals on thousands of domestic and international flights. During this time, the Service has also established a decentralized organization and begun to integrate its operations with those of a new department. While these accomplishments initially came at some cost, as evidenced by air marshals’ concerns with the Service’s management, the Service has taken steps to provide advanced training, improve scheduling, obtain and use more detailed management information, develop and communicate policies and procedures, and obtain and respond to employee feedback. Continuing these efforts will be important for the Service as it moves forward. Developing and analyzing information on the hours air marshals actually work is key to ensuring that the Service’s scheduling controls are operating as intended. Flying for too many hours can cause fatigue, potentially diminishing air marshals’ alertness and reducing their effectiveness. Capturing detailed, firsthand information on air marshals’ reasons for separation is critical to developing cost-effective strategies for promoting retention and would also allow the Service to identify and analyze the root causes of issues and to address vulnerabilities through changes to its policies, procedures, and training. While retention has not been an issue to date, the cost of recruiting, training, and deploying air marshals is too high to risk separations that could be avoided through better understanding of and attention to air marshals’ concerns. We recommend that the Secretary of the Department of Homeland Recommendations for Security direct the Under Secretary for Border and Transportation Executive Action Security to support the Service’s continued commitment to developing into a high-performing organization by taking the following actions to improve management information and to implement key practices that contribute to successful mergers and organizational transformations: • Develop an automated method to compare actual hours worked with scheduled hours so that the Service can monitor the effectiveness of its scheduling controls and support its planned long-term study of the effects of flying on air marshals and their aviation security mission. • Seek and monitor employee attitudes by obtaining detailed, firsthand information on air marshals’ reasons for separation, using such means as confidential, structured exit surveys, that will allow management to Page 28 GAO-04-242 Federal Air Marshal Service analyze and address issues that could affect retention and take appropriate follow-up actions, such as improving training, career development opportunities, and communication. We provided a draft of this report to DHS for its review and comment. Agency Comments DHS agreed with our report’s information and recommendations and said it welcomes our proposals for practices that it believes will ultimately maximize its ability to protect the American public, contribute to the protection of the nation’s critical infrastructure, and preserve the viability of the aviation industry. DHS also expressed a commitment to continuous improvement as it moves forward, including actions designed to build on the accomplishments the Service has already achieved in expanding its mission and workforce since the terrorist attacks of September 11, 2001. According to DHS, the Service has ongoing activities in several areas, such as continuing to address policy issues and develop procedures and establishing field office mechanisms and groups to discuss employee issues and concerns. We included this information in the final report. Additionally, DHS identified references in the draft report to “overscheduling” of air marshals, with an explicit suggestion that such “overscheduling” was among air marshals’ reasons for separating from the Service. We revised the report to avoid this implication, since we had not intended to suggest that air marshals were being overscheduled. Our intent was to point out that without an automated method to compare actual hours worked with scheduled hours, the Service would not readily be able to monitor the effectiveness of its scheduling controls. We also agreed with DHS that there were no data in the Service’s separation information to suggest that “overscheduling” was among air marshals’ reasons for leaving the Service, and we modified the report accordingly. DHS agreed with our recommendation to automate air marshals’ time and attendance data to facilitate comparisons of actual hours worked with scheduled hours and said that the Service was taking steps to implement the recommendation. DHS also agreed that there was a need to improve the quality of the Service’s separation information. In its comments, DHS also emphasized its belief that the Service’s merger with ICE would have a number of significant benefits, particularly from cross-training personnel. DHS noted that after cross-training, the air marshals, as well as personnel in the other ICE components, would have far more law enforcement capability and could supplement each other’s functions during times of heightened threat. Additionally, DHS said that the aviation system would benefit from the concentration and Page 29 GAO-04-242 Federal Air Marshal Service coordination of DHS law enforcement personnel under the direction of a single Assistant Secretary. We discuss these changes in our report by examining them in the context of issues that may arise as the Service merges with other agencies. In addition, we discuss key practices and implementation steps that could be useful in dealing with the changes. We note, however, that it is too early to assess any possible benefits or repercussions of the changes. Finally, DHS provided technical clarifications to the report, which we incorporated into the report as appropriate. As agreed with your offices, unless you publicly announce its contents earlier, we plan no further distribution of this report until 5 days after the date of this letter. At that time, we will send copies of this report to the Ranking Member, Subcommittee on National Security, Emerging Threats, and International Relations, House Committee on Government Reform, other interested congressional committees, the Secretary of Homeland Security, the Undersecretary for Border and Transportation Security, the Administrator of the Transportation Security Administration, and the Acting Assistant Secretary of the Bureau of Immigration and Customs Enforcement. This report is also available on GAO's home page at http://www.gao.gov. Please contact Carol Anderson-Guthrie or me at (202) 512-2834 if you have any questions about the report. Key contributors to this report are listed in appendix VII. Gerald L. Dillingham Director, Civil Aviation Issues Page 30 GAO-04-242 Federal Air Marshal Service Appendix I: Scope and Methodology Appendix I: Scope and Methodology To address each of our study objectives and research questions, we reviewed and analyzed data and documentation provided by the Federal Air Marshal Service (The Service) on background checks and training; scheduling, mission incidents, employee misconduct, and separation; and several workforce policies and procedures. We also interviewed officials responsible for implementing and operating the Service. Additionally, we used our Standards for Internal Controls in the Federal Government, Internal Control Management and Evaluation Tool,1 Human Capital: A Guide for Assessing Strategic Training and Development Efforts in the Federal Government,2 and Model of Strategic Human Capital Management,3 to help assess the Service’s training, management information, and policies and procedures. We also reviewed an audit report by the Department of Transportation’s (DOT) Inspector General (IG) on the Federal Air Marshal program.4 To guide our examination of the Service’s future challenges, we used our Results-Oriented Cultures: Implementation Steps to Assist Mergers and Organizational Transformations.5 To compare the background check procedures for the newly hired air marshals with those used before September 2001, we obtained and reviewed Service documents that described the process and procedures used to apply for a top-secret clearance, as well as for an interim secret clearance waiver. We interviewed officials at the Service’s Human Resource Center in New Jersey who were knowledgeable about the process and were coordinating the Service’s requirements with the responsible Security Management Offices at both the Federal Aviation 1 GAO issues standards for internal control in the federal government as required by the Federal Managers’ Financial Integrity Act of 1982. See 31 U.S.C. 3512©. GAO first issued the standards in 1983. GAO revised the standards and reissued them as Standards for Internal Control in the Federal Government, GAO/AIMD-00-21.3.1(Washington, D.C.: November 1999). These standards provide the overall framework for establishing and maintaining internal control and for identifying and addressing major performance challenges and areas at greatest risk for fraud, waste, and abuse, and mismanagement. GAO issued its Internal Control Management and Evaluation Tool, GAO-01-1008G (Washington, D.C.: August 2001) to assist agencies in maintaining or implementing effective internal control and, when needed, to help determine what, where, and how improvements can be implemented. 2 GAO-03-893G. 3 GAO-02-373SP. 4 SC-2003-029. 5 GAO-03-669. Page 31 GAO-04-242 Federal Air Marshal Service Appendix I: Scope and Methodology Administration (FAA) and the Transportation Security Administration (TSA). We also analyzed data provided by the Office of Personnel Management’s (OPM) Investigative Service and had discussions with OPM personnel on the number of clearances processed and the procedures that are used. To determine what changes were made in the training curriculum for the newly hired air marshals, we analyzed documents related to the air marshal training curriculum. In order to identify the curriculum in place before the changes were made, we interviewed air marshals who had been with the Service before September 2001. To understand the Service’s curriculum from September 2001 through July 2003, we evaluated class schedules, training materials, and training data that tracked the completion of coursework and firearms qualification training. We visited the Federal Law Enforcement Training Center in Artesia, New Mexico and the Service’s training center in New Jersey, where we interviewed officials responsible for overseeing the air marshal training program. In addition, we interviewed representatives of the Air Line Pilots Association, the Air Transport Association, and current and former air marshals. To determine what management information and policies and procedures the Service had developed to support its expanded mission and workforce, we examined the Service’s automated scheduling system and management information on mission incidents, reported misconduct, and reasons for separation. We analyzed the automated scheduling system data to determine if the current system controls were operating as expected. Additionally, to determine the extent of sick leave use and to address allegations of excessive use, we analyzed the “sick calls” generated from the scheduling system between July and October 2002. We also reviewed and discussed with Service management its policies and procedures for air marshals’ transfers between offices, dress code requirements, and work schedules. To determine how many newly hired air marshals have left the Service and why, we used agency data on the number of air marshals on board, hired, and separated each month; supervisory memorandums summarizing exit interviews; resignation letters; personnel action forms; and the Service’s summary database on separations. Using the summary database, we determined the number of air marshals who separated, by reason, and calculated the percentage of total employees that separated for a specific reason. We discussed the process for collecting these data with agency officials responsible for maintaining the Service’s personnel data from the Service’s Human Resource Center in New Jersey. The Service provided Page 32 GAO-04-242 Federal Air Marshal Service Appendix I: Scope and Methodology information on the processing and maintenance of its data and on the relationships among its data systems. When we had concerns about the consistency and validity of the data, we asked agency officials to address each concern. On the basis of the information provided by the agency and our review, we determined that the required data elements were adequate for the purpose of this work. To gain a basic understanding of the issues surrounding staff decisions to leave the Service, we reviewed the agency’s separation data. For each departed staff, these data capture only one predominant reason (for leaving). To supplement this analysis, we selected 95 cases (36 percent of 264 separation cases) that had some form of documentation, had occurred at various times between January 2002 and March 2003, and had originated at various field offices. For each selected case, we reviewed any available resignation letters, exit interviews, and forms documenting personnel actions. This approach allowed us to conduct a limited quality check of the Service’s data and determine whether reasons outside of those reported by the Service provided a broader view of air marshals’ reasons for leaving the Service. To get a better understanding of the types of misconduct that air marshals have been charged with, we reviewed the electronic spreadsheets that the Service uses to track the status of each case of reported misconduct. The spreadsheets included cases reported between October 2001 and July 2003. We sorted the cases of misconduct by category to determine if a particular category was prevalent. We also spoke with Service management about the adjudication of alleged misconduct and the issues related to the completeness and definition of misconduct measures. To determine the types and frequency of the mission reports submitted by air marshals, we analyzed the database maintained by the Federal Air Marshals’ Mission Operations Control Center. This database contained approximately 1,600 incidents that were reported by air marshals between September 11, 2001, and September 16, 2003. We then sorted the incidents into broad categories, including mission-related incidents and incidents that occurred between air marshals and airport or airline personnel. We also received information on the Service’s use and dissemination of the incident data from the Special Agent in Charge of Field Operations. We reviewed the DOT IG’s report on the Federal Air Marshal program as an additional source of information about the Service. This report evaluated various aspects of the Service, including its selection and hiring process and its procedures for properly training and fully qualifying air Page 33 GAO-04-242 Federal Air Marshal Service Appendix I: Scope and Methodology marshals to respond to incidents aboard aircraft. For one aspect of the report, the IG interviewed 112 air marshals in a one-on-one format at their field office duty stations. The air marshals were not selected for interview using structured or random selection methods. Information obtained through these interviews highlights employee concerns with the Service but is anecdotal and therefore cannot be projected to the universe of the Service’s air marshal workforce. Page 34 GAO-04-242 Federal Air Marshal Service Appendix II: Demographic Profile of the Appendix II: Demographic Profile of the Federal Air Marshal Service Federal Air Marshal Service Figure 1: Federal Air Marshal Service Workforce by Gender, by Percentage, as of August 2003 Figure 2: Federal Air Marshal Service Workforce by Age, by Percentage, as of August 2003 Page 35 GAO-04-242 Federal Air Marshal Service Appendix II: Demographic Profile of the Federal Air Marshal Service Figure 3: Federal Air Marshal Service Workforce by Race, by Percentage, as of August 2003 Page 36 GAO-04-242 Federal Air Marshal Service Appendix III: Locations of the Federal Air Appendix III: Locations of the Federal Air Marshal Service’s 21 Field Offices and Training Facility Marshal Service’s 21 Field Offices and Training Facility Page 37 GAO-04-242 Federal Air Marshal Service Appendix IV: Events Affecting the Federal Air Appendix IV: Events Affecting the Federal Air Marshal Service, September 2001 through October 2002 Marshal Service, September 2001 through October 2002 a The exact number of federal air marshals is classified. Page 38 GAO-04-242 Federal Air Marshal Service Appendix V: Mission-Related Incidents Appendix V: Mission-Related Incidents Reported by Federal Air Marshals, by Broad Categories, September 15, 2001 — September 16, 2003 Reported by Federal Air Marshals, by Broad Categories, September 15, 2001 — September 16, 2003 Incident category Number of incidents reported Percentage of total Suspicious person 444 21.3 Suspicious activities by person 394 18.9 Issues with airport or airline personnel • Assigned seating and/or boarding procedures 159 7.6 • Screening and/or escort procedures 106 5.0 • Check-in procedures 59 2.8 • Flight crew briefings 57 2.7 Subtotal: (Issues with airport or airline personnel) 381 18.3 Suspect items or objects 164 7.9 Third-party information reported to air marshal 129 6.2 Undercover status compromised 113 5.4 Disruptive/disorderly person 73 3.5 Security breeches 49 2.3 Medical problems 35 1.7 Arrest/detainment by or at request of air marshal 28 1.3 Interference with flight crew by passenger 20 1 Verbal threats or threatening behavior 19 0.9 Use of nonlethal force by an air marshal 16 0.8 Searches 12 0.5 Equipment retrieval/turn-in 7 0.3 Tampering with aircraft or aircraft equipment 4 0.2 Discharge of an air marshal firearm 3 0.2 To be determined 2 .1 Not applicable 1 0.1 Other 189 9 Total 2,083 100 Source: Federal Air Marshal Incident Reports Database, September 15, 2001, through September 16, 2003. Notes: Total includes some incidents counted more than once, because multiple codes for a reportable incident might have been reported (e.g., a suspicious person incident might also have been reported as a drunk and disorderly incident). The information above represents the major categories of information on incidents that air marshals report to the Service’s Operations Center. Page 39 GAO-04-242 Federal Air Marshal Service Appendix VI: Key Practices and Appendix VI: Key Practices and Implementation Steps for Mergers and Organizational Transformations Implementation Steps for Mergers and Organizational Transformations Practice Implementation step Ensure top leadership drives the • Define and articulate a succinct and transformation. compelling reason for change. • Balance continued delivery of services with merger and transformation activities. Establish a coherent mission and • Adopt leading practices for results- integrated strategic goals to guide the oriented strategic planning and reporting. transformation. Focus on a key set of principles and • Embed core values in every aspect of the priorities at the outset of the organization to reinforce the new culture. transformation. Set implementation goals and a • Make public implementation goals and timeline to build momentum and show timeline. progress from day one. • Seek and monitor employee attitudes and take appropriate follow-up actions. • Identify cultural features of merging organizations to increase understanding of former work environments. • Attract and retain key talent. • Establish an organizationwide knowledge and skills inventory to allow knowledge exchange among merging organizations. Dedicate an implementation team to • Establish networks to support manage the transformation process. implementation team. • Select high-performing team members. Use the performance management • Adopt leading practices to implement system to define the responsibility and effective performance management assure accountability for change. systems with adequate safeguards. Establish a communication strategy to • Communicate early and often to build create shared expectations and report trust. related progress. • Ensure consistency of message. • Encourage two-way communication. • Provide information to meet specific needs of employees. Involve employees to obtain their ideas • Use employee teams. and gain ownership for the • Involve employees in planning and transformation. sharing performance information. • Incorporate employee feedback into new policies and procedures. • Delegate authority to appropriate organizational levels. Build a world-class organization. • Adopt leading practices to build a world- class organization. Source: GAO-03-699. Page 40 GAO-04-242 Federal Air Marshal Service Appendix VII: Contacts and Acknowledgments Appendix VII: Contacts and Acknowledgments Gerald Dillingham, (202) 512-2834 GAO Contacts Carol Anderson-Guthrie, (214) 777-5739 In addition to those named above, Bess Eisenstadt, David Hooper, Kevin Acknowledgments Jackson, Maren McAvoy, Minette Richardson, Laura Shumway, Rick Smith, Gladys Toro, and Alwynne Wilber made key contributions to this report. (540078) Page 41 GAO-04-242 Federal Air Marshal Service The General Accounting Office, the audit, evaluation and investigative arm of GAO’s Mission 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. The fastest and easiest way to obtain copies of GAO documents at no cost is Obtaining Copies of through the Internet. GAO’s Web site (www.gao.gov) contains abstracts and full- GAO Reports and text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents Testimony using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as “Today’s Reports,” on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select “Subscribe to e-mail alerts” under the “Order GAO Products” heading. Order by Mail or Phone The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. General Accounting Office 441 G Street NW, Room LM Washington, D.C. 20548 To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202) 512-6061 Contact: To Report Fraud, Web site: www.gao.gov/fraudnet/fraudnet.htm Waste, and Abuse in E-mail: email@example.com Federal Programs Automated answering system: (800) 424-5454 or (202) 512-7470 Jeff Nelligan, Managing Director, NelliganJ@gao.gov (202) 512-4800 Public Affairs U.S. General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C. 20548
Aviation Security: Federal Air Marshal Service Is Addressing Challenges of Its Expanded Mission and Workforce, but Additional Actions Needed
Published by the Government Accountability Office on 2003-11-19.
Below is a raw (and likely hideous) rendition of the original report. (PDF)