This is the accessible text file for GAO report number GAO-02-818 entitled 'Peace Corps: Initiatives for Addressing Safety and Security Challenges Hold Promise, but Progress Should Be Assessed' which was released on July 25, 2002. This text file was formatted by the U.S. General Accounting Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products’ accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. Report to Congressional Requesters: July 2002: Peace Corps: Initiatives for Addressing Safety and Security Challenges Hold Promise, but Progress Should Be Assessed: GAO-02-818: Contents: Letter: Results in Brief: Background: Reported Crime Incidents Have Increased, but the Full Extent of Crime Against Volunteers Is Unknown Due to Underreporting: Peace Corps Provides Broad Guidance and Support to Posts, Relying on Them to Develop and Implement Effective Practices: Peace Corps’ Safety and Security Framework Is Unevenly Implemented: Peace Corps Initiatives May Enhance Volunteer Safety and Security: Conclusions: Recommendations: Agency Comments: Appendixes: Appendix I: Objectives, Scope, and Methodology: Appendix II: Other Organizations’ Safety and Security Approaches Vary According to Work and Goal: Appendix III: Peace Corps’ Crime Data Analysis System and Crime Trends: Peace Corps Has Two Systems for Collecting Crime Information: Trends Differ for Aggravated Assault, Rape, and Nonassault Crimes: Appendix IV: Peace Corps Provisions for Responding to Criminal Incidents: Appendix V: Comments from the Peace Corps: Appendix VI: GAO Contact and Staff Acknowledgements: GAO Contact: Staff Acknowledgments: Table: Table 1: Number of Aggravated Assaults and Rapes Reported by Volunteers, 1990-2001: Figures: Figure 1: Incidence of Reported Physical Assaults against Volunteers, 1991-2000: Figure 2: Incidence of Reported Sexual Assaults against Female Volunteers, 1991-2000: Figure 3: Reported Assaults against Volunteers in Relation to Length of Volunteer Service, 1993-2001: Figure 4: Percentage of Post EAPs Missing Selected Elements from Agency Checklist: Figure 5: Post Reports of Volunteer Contact Time in 2001 EAP Tests: Figure 6: Summary of Peace Corps Safety and Security Initiatives of May 2002: Figure 7: Incidence Rates of Reported Aggravated Assaults, 1990-2001: Figure 8: Incidence Rates of Reported Rape, 1990-2001: Figure 9: Incidence Rates of Reported Robberies and Burglaries, 1993- 2000: Figure 10: Incidence Rates of Reported Thefts, 1993-2000: Abbreviations: EAP: emergency action plan: Letter July 25, 2002: The Honorable Cynthia A. McKinney The Honorable Martin T. Meehan House of Representatives: Approximately 7,000 Peace Corps volunteers currently serve in about 70 countries, where they face a variety of safety and security risks. Volunteers often live in areas with limited access to reliable communications, police, or medical services and, as Americans, may be viewed as relatively wealthy and hence good targets for criminal activity. Incidents such as the 1998 bombings of the U.S. embassies in Kenya and Tanzania and the events of September 11, 2001, have heightened awareness about the risks facing all Americans, especially those living abroad. In his State of the Union address, President Bush proposed doubling the number of volunteers worldwide and increasing the number of Peace Corps posts. You asked us to evaluate Peace Corps safety and security policies and practices. In this report, we (1) describe rates and trends in crime against volunteers and review the agency’s system for generating such information, (2) describe the agency’s framework for maintaining volunteer safety and security, (3) evaluate the Peace Corps’ implementation of this framework, and (4) review agency initiatives to improve current practices. We also describe practices employed by other organizations, such as the Japanese Overseas Cooperation Volunteers, that face safety and security challenges similar to the Peace Corps. To meet our objectives, we reviewed and analyzed Peace Corps documents and data, including safety and security guidance; data on crime against volunteers since 1990; the results of worldwide volunteer satisfaction surveys in 1998 and 1999; and reports on agency safety and security practices prepared by the Peace Corps’ Office of the Inspector General and by the agency’s safety and security staff. We conducted fieldwork in five countries (Bulgaria, El Salvador, Kenya, Senegal, and Ukraine) and met with Peace Corps staff, volunteers, and representatives of local organizations that host volunteers. Finally, we interviewed agency officials and representatives of other organizations that face similar safety challenges. Appendix I provides a more detailed description of our scope and methodology. Results in Brief: The Peace Corps has reported increased numbers of assaults against its volunteers since it established a data collection system in 1990. The reported incidence rate for major physical assaults nearly doubled from about an average of 9 per 1,000 volunteer years[Footnote 1] in 1991-93 to an average of about 17 per 1,000 volunteer years in 1998-2000. The agency is not certain of the reasons for the higher occurrence rate, but officials have stated that its efforts to improve its system for collecting crime data may be a factor that has led to higher reported crimes. Yet it is unclear what the full extent of incidents may be because volunteer surveys have indicated that there is significant underreporting of crime. The Peace Corps has initiated efforts to encourage reporting and collect additional data, but there are also other unrealized opportunities for additional examination of data. For example, our analysis suggests that newer volunteers may be more likely to become victims of crime than their more experienced colleagues. Additional data analysis by the Peace Corps could enhance the agency’s ability to refine its intervention and prevention strategies. The Peace Corps designates maintenance of volunteers’ health, safety, and security as the agency’s highest priority. To reduce the risks facing its volunteers, the Peace Corps has adopted policies that address, in broad terms, monitoring and disseminating information on the security environment; volunteer training; development of safe and secure housing and work sites for volunteers; monitoring volunteers and responding to incidents and concerns; and planning for emergencies such as evacuations. In addition to establishing agency policies, Peace Corps headquarters is responsible for providing guidance and training on how to implement these policies and supervision and oversight. The agency relies on its country directors--the heads of agency posts in foreign capitols--to develop and implement procedures that suit conditions in individual countries. Volunteers also play a role in ensuring their own safety by complying with agency policies and exercising good judgment and common sense. The Peace Corps’ efforts to ensure effective implementation of its safety and security policies have produced varying results. Volunteer surveys and our visits to five overseas posts indicate that volunteers appear to be generally satisfied with agency training programs and other efforts designed to emphasize safety and security awareness. However, there is mixed performance in key areas, such as developing safe and secure housing and work sites for volunteers, monitoring volunteers and responding when they express security concerns or experience criminal incidents, and preparing for emergencies. For example, while many volunteers are provided with adequate housing and clearly defined assignments, some experience safety problems resulting from housing that has not been inspected or does not meet post standards. A number of factors, including unclear guidance, staff training that is sometimes inadequate, uneven application of supervision and oversight mechanisms, and staff turnover, hamper Peace Corps efforts to ensure high-quality performance for the agency as a whole. For example, the Peace Corps has reported that high staff turnover, caused in part by the agency’s statutorily imposed 5-year limit on employment for U.S. direct hire staff, has resulted in a lack of institutional memory, producing a situation in which agency staff are continually “reinventing the wheel.”: In May 2002, the Peace Corps informed us of a number of initiatives the agency intended to pursue to improve current safety and security practices. These initiatives are directed at many of the obstacles to improved performance that we identified, though they do not address turnover in agency staff. The Peace Corps has implemented some of these initiatives but many have yet to be integrated into agency operations. We recommend that the Director of the Peace Corps develop indicators to assess the effectiveness of the agency’s new initiatives and include the results of these initiatives in its annual reports under the Government Performance and Results Act. We also recommend that the Peace Corps develop a strategy to address staff turnover as it implements its initiatives. In written comments on a draft of this report, reprinted in appendix V, the Peace Corps concurred with our findings and provided additional information on the agency’s safety and security initiatives and technical comments that we incorporated as appropriate. The Peace Corps agreed to report on the results of its initiatives in its annual reports under the Government Performance and Results Act, as we recommended. However, the agency stated it could not effectively address the issue of staff turnover as we recommended because of its statutorily imposed 5-year limit on employment for U.S. direct hires. We modified our recommendation to suggest that the Peace Corps submit a proposal to Congress for changes in the law that would facilitate agency efforts to improve its safety and security practices. Background: Created in 1961, the Peace Corps is mandated by statute to help meet developing countries’ need for trained manpower while promoting mutual understanding between Americans and other peoples.[Footnote 2] Volunteers commit to 2-year assignments in host communities where they work on projects such as teaching English, strengthening farmer cooperatives, or building sanitation systems. By developing relationships with members of the communities in which they live and work, volunteers contribute to greater intercultural understanding between Americans and host country nationals. Volunteers are expected to maintain a standard of living similar to that of their host community colleagues and coworkers. They are provided with stipends that are based on local living costs and housing similar to their hosts. Volunteers are not supplied with vehicles. Although the Peace Corps accepts older volunteers and has made a conscious effort to recruit minorities, the current volunteer population has a median age of 25 years and is 85 percent white. More than 60 percent of the volunteers are women. The Peace Corps emphasizes community acceptance as the key to maintaining volunteer safety and security. The agency has found that volunteer safety is best ensured when volunteers are well-integrated into their host communities and treated as extended family members and contributors to development. While emphasizing protection measures such as locks and window bars, the Peace Corps generally avoids measures such as housing volunteers in walled compounds, which would reduce volunteer integration into the community. The agency also typically withdraws from countries in which breakdowns in civil authority require strong protection or deterrence measures to protect volunteers.[Footnote 3] To the extent that they share the Peace Corps’ commitment to advancing intercultural understanding, other organizations that face similar security and safety challenges also tend to emphasize community acceptance as an underlying principle. Appendix II presents in greater detail the safety and security practices of some of these organizations. Reported Crime Incidents Have Increased, but the Full Extent of Crime Against Volunteers Is Unknown Due to Underreporting: The Peace Corps’ Office of Medical Services created and operates a system for recording and analyzing crime information that focuses primarily on assault crimes. Peace Corps reports show that reported rates of assault nearly doubled from the early 1990s to the latter part of the decade. Agency officials note that the reason for this may be attributable to a number of factors, including agency efforts to improve data collection and volunteer reporting. The Peace Corps has used its data analyses to gain insight into the characteristics of assaults against volunteers and to shape volunteer training programs. However, the full extent of crime against volunteers is unclear because recent volunteer surveys show that volunteers may significantly underreport crime. Additional analyses would enhance the agency’s ability to understand trends in crime and apply this understanding to its crime prevention and intervention strategies. Medical Staff Operates Data Collection and Analysis System: Since 1990, the Office of Medical Services has collected information on assaults from post medical staff around the world and has produced analyses of incidence rates and characteristics of assaults, such as time and place of occurrence, weapons employed, and injuries sustained. Medical staff also collect summary information on the number of nonassault crimes, such as burglaries and thefts, occurring at posts each month. The office periodically publishes reports containing its analytical results and distributes these reports to senior staff, country directors, and post medical officers.[Footnote 4] Appendix III provides additional information on the processes employed to gather information and produce these reports. Apparent Increase in Assaults Is Difficult to Interpret: Reported incidence rates for most types of assaults have been higher in recent years, as shown in figures 1 and 2. For example, the reported incidence rate for major physical assaults nearly doubled from an average of about 9 per 1,000 volunteer years in 1991 to 1993 to an average of about 17 per 1,000 volunteer years in 1998 to 2000. Reported incidence rates also increased for minor physical assaults and, to a lesser extent, for minor sexual assaults. The reported rate of major sexual assaults decreased from about 10 incidents per 1,000 female volunteer years at the beginning of the 1990s to an average of slightly more than 8 per 1,000 female employees at the end of the decade. According to agency officials, the decreasing incidence of major sexual assaults in the face of increases in minor sexual assault suggests that the decline in major sexual assaults is a true decline rather than a reporting artifact. Appendix III provides more information on crime rates and trends. Figure 1: Incidence of Reported Physical Assaults against Volunteers, 1991-2000: [See PDF for image] Legend: V/T year = 1 full year of service by a volunteer or trainee. Note: Major physical assaults include those in which a weapon was used, substantial injuries were sustained, or the victim had to use significant force to disengage the assailant. Minor physical assaults include those committed without a weapon or those in which minor injuries were sustained. Source: Peace Corps analysis. [End of figure] Figure 2: Incidence of Reported Sexual Assaults against Female Volunteers, 1991-2000: [See PDF for image] Legend: V/T year = 1 full year of service by a volunteer or trainee. Note: Major sexual assaults include rape, attempted rape, assaults in which a weapon was used, physical injury resulted, or substantial force was necessary to disengage the assailant. Minor sexual assaults are all other assaults of a sexual nature. Source: Peace Corps analysis. [End of figure] According to Peace Corps officials, the general increase in reported assaults may reflect an actual increase in the number of such incidents suffered by volunteers, better efforts by the agency to ensure that all medical officers report all assault events, or an increased willingness among volunteers to report incidents. Through its volunteer satisfaction surveys, the agency is aware that the level of underreporting is significant. For example, according to the 1998 survey, volunteers did not report 60 percent of rapes and 20 percent of nonrape sexual assaults. Underreporting reduces the Peace Corps’ ability to state crime rates with certainty and to develop well- informed plans for addressing crime problems.[Footnote 5] The agency has taken steps to encourage volunteers to report incidents. For example, the coordinator for volunteer safety and security stated that he is developing training materials for medical officers to ensure that they transmit clear messages to volunteers about incident reporting. The Peace Corps is also including questions about underreporting in its current volunteer satisfaction survey.[Footnote 6] Volunteers may not report criminal incidents for a variety of reasons, including embarrassment, fear of repercussions, concern about confidentiality, and the belief that Peace Corps staff could not help. Volunteers may decline to report minor incidents when, aside from offering counseling, it is unclear what Peace Corps staff can do for the volunteer. In addition, volunteers are sometimes unclear about what to report, and staff observed that definitions for reportable nonassault crimes, in particular, need clarification. Crime Data Analysis System Has Produced Useful Insights but Could Be Enhanced: The Peace Corps’ system for gathering and analyzing data on crime against volunteers has produced useful insights, but opportunities for additional analyses may help the agency develop better-informed intervention and prevention strategies. In addition, the results of agency analyses could be more broadly shared. Some post medical officers we interviewed stated that they use headquarters analyses of crime data during volunteer training to illustrate the risks volunteers face. These analyses also have influenced the content of the Peace Corps’ volunteer training programs. For example, agency analyses of the circumstances surrounding rape incidents have shown that nearly 60 percent of such crimes from 1993 to 1999 were perpetrated by volunteers’ friends, coworkers, or acquaintances, and that more than 50 percent occurred in a home environment. The Peace Corps’ coordinator for volunteer safety and security stated that the agency was revising volunteer rape awareness training materials to reflect these insights. In recent years, the Peace Corps has made a number of improvements in its crime data collection and analysis system. In 1999, the agency revised its assault reporting form to include information on victim and assailant alcohol use and on whether victims were alone when incidents occurred. Additional analyses would enhance the Peace Corps’ ability to identify other characteristics of crimes and crime risk factors and develop better-informed prevention and intervention strategies. For example, as shown in figure 3, we found that the number of reported assaults is highest among volunteers in their first few months of service. Nearly a third of all reported assaults after 1993 occurred in the volunteer’s 4TH to 8TH months of service--immediately after the volunteers have completed training and taken up residence at their assigned sites. This finding could be explored and the results considered in developing volunteer training materials. Medical staff and safety and security staff at the Peace Corps agreed that the agency could benefit from additional research on crime against volunteers but observed that neither the medical office nor the coordinator for volunteer safety and security had staff available to perform such research. Among the new initiatives the Peace Corps has stated that it will implement is the hiring of a statistician to perform additional analyses on crime data. Figure 3: Reported Assaults against Volunteers in Relation to Length of Volunteer Service, 1993-2001: [See PDF for image] Note: The Peace Corps did not provide information on months in service prior to 1993. Source: GAO analysis of Peace Corps assault and administrative data. [End of figure] The Peace Corps distributes its crime data analyses to agency officials but does not provide access to this information for potential volunteers. For example, it does not post the results, or a summary thereof, on the agency’s Web site. Most volunteers in the field that we interviewed stated that they had been provided little or no specific information on crime incidents before their arrival in the country for preservice training. The Peace Corps’ safety and security initiatives include efforts to more fully inform applicants and recruits of the safety and security challenges they are likely to face as volunteers. Peace Corps Provides Broad Guidance and Support to Posts, Relying on Them to Develop and Implement Effective Practices: Volunteer health, safety, and security is the Peace Corps’ highest priority, according to the agency. To address this commitment, the agency has adopted policies for monitoring and disseminating information on the security environments in which the agency operates, training volunteers, developing safe and secure volunteer housing and work sites, monitoring volunteers, and planning for emergencies such as evacuations. Headquarters is responsible for providing guidance, supervision, and oversight to ensure that agency policies are implemented effectively. The Peace Corps relies heavily on country directors--the heads of agency posts in foreign capitols--to develop and implement practices that are appropriate for specific countries. Country directors, in turn, rely on program managers to develop and oversee volunteer programs.[Footnote 7] Volunteers are expected to follow agency policies and exercise some responsibility for their own safety and security. Headquarters Establishes Policies and Provides Support for Their Implementation: Peace Corps headquarters is responsible for establishing the agency’s safety and security policy framework and supports posts in implementing these policies through (1) guidance and training and (2) supervision and oversight. Headquarters Has Established Agencywide Policies: According to agency officials, the Peace Corps has long regarded volunteer safety and security as its highest priority. The agency maintains this focus in its current strategic planning documents, prepared under the provisions: of the Government Performance and Results Act.[Footnote 8] In 1999, the Peace Corps established a policy framework that outlines the agency’s principles for maintaining volunteer safety and security.[Footnote 9] These agencywide policies are broadly phrased to give country directors flexibility in developing procedures that suit conditions in countries as diverse as Belize and Kazakhstan. Peace Corps policies cover the following: * Monitoring and disseminating information on the security environment in Peace Corps countries. Volunteers should be provided with a clear understanding of the risks they face (including an overall assessment of the risks facing volunteers and information on country-specific conditions) so that they can make informed decisions about their own safety. * Training volunteers. Volunteers should be provided with training that prepares them to “adopt culturally appropriate lifestyles and exercise judgment that promotes safety and reduces risk in their home, at work, and while traveling.”: * Developing volunteer housing and work sites. Volunteers should be placed in “appropriate, safe, and secure housing and work sites.” Criteria for selecting sites include the potential for volunteers’ obtaining and maintaining “acceptance” in the communities where they will live and work. * Monitoring sites and volunteers and responding to safety concerns and criminal incidents. Post staff should make periodic visits to volunteer sites and respond to volunteer safety and security concerns and incidents, including crimes against volunteers. * Planning for emergencies. Posts must maintain accurate contact information on all volunteers and develop and annually test emergency action plans (EAP) to guide staff and volunteers in the event of a natural disaster, political unrest, or other emergency. Headquarters is to review the EAPs and the EAP test results. Headquarters Supports Posts in Implementing Safety and Security Policies: Headquarters has developed written guidance and training for headquarters and field staff to support implementation of safety and security policies. In collaboration with other agency officials, the coordinator for volunteer safety and security has developed a variety of guidance materials for posts, including information on “best practices” in safety and security operations from posts around the world, crisis management and rape response handbooks, and training modules that posts can apply in preparing volunteer safety and security training programs. These materials are generally nonprescriptive and can be adapted to country-specific conditions. Peace Corps staff, including country directors and program managers at posts, are given training in safety and security procedures as part of their introduction to their positions. For example, all new program managers attend a 4-week overseas staff training session in Washington, D.C., that addresses safety and security issues and other aspects of their work. Agency staff also attend periodic in-service training events that may include safety and security matters. Headquarters also provides supervision and oversight. Three regional directors,[Footnote 10] each assisted by a small staff of country desk officers, supervise Peace Corps posts abroad. Agency policies state that these regional directors are to ensure that country directors establish effective volunteer safety and security support systems. The regional directors, with their country desk officers, monitor post operations in all areas--including safety and security--by E-mail, telephone, and occasional country visits. This informal dialogue is supplemented by formal submission and review of post EAPs and EAP test results. In addition to these regional directors, Peace Corps’ Office of Volunteer Safety and Overseas Security (headed by a coordinator for volunteer safety and security) and the Office of the Inspector General contribute to headquarters’ supervision and oversight of post practices. A field-based regional safety and security officer works in each of the three regions. At the request of regional or country directors, these officers review and provide advisory reports on post safety and security practices. The Office of the Inspector General, among other things, reviews safety and security operations at posts and issues formal recommendations that require an official post response. Country Directors at Individual Posts Are Responsible for Implementing Policies: Peace Corps country directors are responsible for developing procedures to ensure the effective implementation in specific countries of the agency’s broadly phrased policies, as previously mentioned. For example, country directors develop safety and security criteria for prospective volunteer sites and procedures for ensuring that sites meet these criteria before volunteers arrive. They also develop and provide volunteer safety and security training programs in accordance with agency policies. Volunteers Play a Role in Ensuring Their Own Safety: Volunteers are expected to exercise responsibility for their own safety and security. They are expected to reduce the level of risk they face at their sites and while traveling by complying with post policies and exercising good judgment. They do this in part through the relationships they build with sponsoring organizations and elements of the local community. Peace Corps posts employ a number of program managers who work with local organizations to develop programs in areas such as education and health and to identify housing and work assignments for volunteers. After 3 months of incountry training, volunteers move to diverse sites, often far from Peace Corps posts, where they live in a community and work with local counterpart organizations such as schools and municipal governments. Program managers are expected to monitor volunteers once they arrive at their sites and to provide support when needed. Volunteers do not work directly for or have daily contact with agency staff, however. They are not considered U.S. government employees for most purposes,[Footnote 11] nor do they not have diplomatic immunity. Peace Corps’ Safety and Security Framework Is Unevenly Implemented: Peace Corps’ efforts to implement its safety and security policies have produced varying results. We found mixed performance in key areas, which may expose some volunteers to risk. Volunteers are generally satisfied with the safety and security information and training they receive. We identified a number of instances of uneven performance in developing safe and secure housing and work sites and responding to volunteers’ safety concerns. In addition, while all posts have developed an EAP that they test at least annually, the plans and tests vary in quality and comprehensiveness, and the Peace Corps does not have information about how long it would take to reach its volunteers in case of an emergency. A number of factors, including unclear guidance, inadequate staff training, uneven application of supervisory and oversight mechanisms, and staff turnover, hamper Peace Corps efforts to ensure high-quality performance for the agency as a whole. Volunteers Are Generally Satisfied with Post Efforts to Inform Them about the Security Environment: Posts are responsible for monitoring the host country’s safety and security environment and for keeping volunteers informed about safety and security issues. Numerous volunteers we met with were generally satisfied with post efforts in this area. The Peace Corps does not require country directors to prepare formal assessments of the security environment. In general, country directors stay informed about the security environment through regular discussions with local Department of State security officials, information on crime reported by volunteers, and other means.[Footnote 12] Posts use various mechanisms, such as newsletters, E-mail, and memorandums to disseminate safety information to volunteers. Although posts vary in how and when they disseminate such information, volunteers at the posts we visited said they were fairly satisfied with the level of information they receive about safety and security. According to the 1998 and 1999 volunteer satisfaction surveys, over 80 percent of volunteers found that the Peace Corps kept them adequately or well-informed regarding safety and security, while around 14 percent said that they were not at all informed or poorly informed. Volunteers Are Generally Satisfied with Safety Training: Training is central to the Peace Corps’ approach to volunteer safety. Volunteers are generally satisfied with the safety training that the agency provides. Posts have considerable latitude in the design of their safety training programs, but all provide volunteers with 3 months of preservice training that includes information on safety and security. Posts also provide periodic in-service training sessions that cover technical issues. Many of the volunteers we interviewed said that the safety training they received before they began service was useful and cited testimonials by current volunteers as one of the more valuable instructional methods. In both the 1998 and 1999 volunteer satisfaction surveys, over 90 percent of volunteers rated safety and security training as adequate or better; only about 5 percent said that the training was not effective. Some regional safety and security officer reports have found that improvements were needed in post training practices.[Footnote 13] The inspector general has reported that volunteers at some posts said cross-cultural training and presentations by the U.S. embassy’s security officer did not prepare them adequately for safety-related challenges they faced during service. Some volunteers stated that the Peace Corps did not fully prepare them for the racial and sexual harassment[Footnote 14] they experienced during their service. Some female volunteers at posts we visited stated that they would like to receive self-protection training. Peace Corps Showed Mixed Performance in Developing Safe and Secure Housing and Work Sites for Volunteers: Although many volunteers are provided with housing that meets Peace Corps standards and well-defined work assignments, some volunteers do not have this experience. We found that volunteer housing is not always inspected before the volunteer arrives, some housing does not meet posts’ standards, and some posts have unclear or nonexistent guidance for selecting volunteer housing. In addition, vaguely defined work assignments and unsupportive counterparts may also increase volunteers’ risk by limiting their ability to build a support network in their host communities. We also found that documentation recording information and problems, by site location, was not maintained, which affects the ability of Peace Corps staff to make informed decisions about future placements and could lead to placing volunteers at sites that have previously experienced safety problems. Peace Corps policies call for posts to ensure that housing is inspected and meets post safety and security criteria before the volunteers arrive to take up residence.[Footnote 15] Nonetheless, some volunteers arrive at their sites to find that their housing is not ready, has not been inspected, or does not meet post standards. At all of the posts we visited, we found instances of volunteers who began their service in housing that had not been inspected and had various shortcomings. For example, one volunteer spent her first 3 weeks at her site living in her counterpart’s office. She later found her own house; however, post staff had not inspected this house even though she had lived in it for several months. In other cases, volunteers and staff said that housing was approved despite deficiencies, with the understanding that the community would rectify the problems before the volunteer arrived. The community failed to comply, however, and staff did not revisit the sites to ensure that problems had been resolved. Several inspector general safety assessments reported instances where the Peace Corps’ failure to inspect housing resulted in volunteers’ not having appropriate housing when they arrived at their sites. According to recent Peace Corps reports, some posts have unclear or nonexistent criteria for selecting a house, which can result in volunteers living in inappropriate housing. For example, the Peace Corps’ review of one post found that unclear housing standards led to multiple instances of volunteers’ living in inadequate housing. In one case, a volunteer lived in a one-room apartment with her counterpart and the counterpart’s boyfriend. Poorly defined assignments and unsupportive counterparts may also increase volunteers’ risk by limiting their ability to build a support network in their host communities. Our previous work in this area has shown that the Peace Corps has had difficulty providing volunteers with well-structured assignments.[Footnote 16] At the posts we visited, we met volunteers whose counterparts had no plans for them when they arrived at their sites, and only after several months and much frustration did the volunteers find productive activities. Several inspector general reports support this finding. For example, at one post volunteers reported that their coworkers were not at all or were poorly prepared for their arrival. Some volunteers had no real job to do or had not been assigned a counterpart. Senior Peace Corps officials agreed that poorly defined assignments pose a safety risk because volunteers who lack the routine a job provides may spend time away from their sites and have difficulty integrating into their communities. While 76 percent of volunteers in the 1999 volunteer satisfaction survey said that their assignment responsibilities were moderately or mostly clear, 24 percent said these responsibilities were somewhat or not at all clear. Peace Corps policy requires posts to maintain site history files documenting the placement of volunteers at specific sites. Staff thus should have a record of the safety and security environment at volunteer placement sites to help ensure that other volunteers are not placed at sites with significant problems. Four of the five posts we visited did not fully comply with Peace Corps requirements--most of these kept records of safety and security problems in the volunteers’ personal files, thereby making it difficult for program managers to access information about specific sites. Inadequate or nonexistent site history files can affect staff’s ability to make informed decisions about future placements and could lead to placements in areas where volunteers have previously experienced safety problems. For example, at one post we visited, two female volunteers who experienced severe sexual harassment at their site were reassigned to new sites. Records of the incident were kept in their personal files, but the post had no file organized geographically to track after the volunteers were moved. A female volunteer from another program area was later placed in a nearby assignment that required her to travel regularly through the site where the difficulties had occurred. Reports by Peace Corps’ inspector general and regional safety and security officers have also cited problems with posts’ site history files. Peace Corps Showed Variation in Monitoring and Responding to Volunteer Concerns: Peace Corps guidance does not specify how its posts should monitor volunteers. Peace Corps policy allows each post flexibility in establishing the frequency of required staff visits to volunteer sites. Posts conduct site visits to assist volunteers and monitor their activities. We found that there is variation in the frequency of staff contact with volunteers. In addition, volunteers have mixed views on staff responsiveness to safety and security concerns and criminal incidents. We reviewed about 25 percent of all site visit policies established by posts and found that the required frequency of staff visits to volunteer sites ranged from once per year to four times during the first year of service. Volunteers may have more frequent contact with Peace Corps staff if they wish. At the five posts we visited, we found that staff made regular site visits to most volunteers, in accordance with each post’s policies. In the 1998 volunteer satisfaction survey, 68 percent of volunteers reported that the frequency of site visits was adequate or better; 21 percent said that the frequency of site visits was inadequate. Many volunteers at the posts we visited were satisfied with the frequency of site visits. Many Peace Corps staff told us that it is sometimes difficult for them to stay abreast of volunteers’ whereabouts when volunteers are away from their sites. Some staff also said that volunteers face safety risks when they are away from their sites because the volunteers are outside their supportive network and because public transportation may be unsafe. The posts we visited have policies to keep track of volunteers who leave their sites, but we found that volunteers’ compliance with these policies was uneven. Many volunteers we interviewed said that they do not always inform the Peace Corps when they leave their sites, but they may inform other people such as neighbors. One reason volunteers may not report their whereabouts is that Peace Corps policy states that volunteers are “on duty” 7 days per week. Although posts may not follow this policy in practice, some volunteers said they are reluctant to inform the post when they plan to leave their sites because they worry that the post may deduct vacation days. This practice may make it difficult if the Peace Corps needs to contact volunteers in an emergency. Volunteers had mixed views about the Peace Corps’ responsiveness to safety and security concerns and criminal incidents. (Appendix IV describes Peace Corps provisions for responding to criminal incidents.) The few volunteers we spoke with who said that they were victims of assault expressed satisfaction with staff response when they reported the incidents. However, at four of the five posts we visited, some volunteers described instances in which staff were unsupportive when the volunteers reported non-assault safety concerns. For example, one volunteer we interviewed informed Peace Corps staff several times that she needed a new housing arrangement because her doorman repeatedly locked her in or out of her dormitory. The volunteer said staff were unresponsive, and she had to find new housing without the Peace Corps’ assistance. In the 1998 and 1999 volunteer satisfaction surveys, 60 percent of volunteers stated that they were satisfied with safety and security support provided by Peace Corps staff, and about 35 percent reported that they were unsatisfied or only somewhat satisfied with this support. According to the 1998 survey, 64 percent of volunteers said that staff response to issues raised during site visits was adequate or better, but 26 percent of volunteers said staff response was inadequate. Senior Peace Corps officials recognize the importance of responding to volunteer safety concerns, and one acknowledged the need to improve staff responsiveness, particularly to nonassault incident reports. At two posts we visited, country directors attributed unsupportive responses to poor communications between volunteers and staff and to staff attitudes toward volunteers. Comprehensiveness and Quality of Emergency Action Plans Are Uneven: Posts must be well prepared in case an evacuation becomes necessary-- Peace Corps evacuated more than 1,600 volunteers from 26 posts from 1993 to 2001. Peace Corps policy requires that all posts develop an EAP, test it annually, and submit it and the test results to headquarters. We found that posts complied with these requirements. However, we also found that some posts’ EAPs lacked key information, and none of the EAPs contained all of the dimensions listed in the EAP guidance for developing effective emergency plans. Moreover, the Peace Corps has not defined the criteria for a successful EAP test nor is there a standard format for reporting test results. Both factors contribute to making the Peace Corps’ assessment of posts’ emergency drills difficult. The Peace Corps’ EAP policy requires posts to develop an EAP tailored to the conditions at that post and to test the EAP annually. We found that all posts had developed EAPs and had tested them annually. To guide the post through the development of an EAP, the Peace Corps has created a suggested format designed to assist the posts in formulating effective emergency plans. This format, a checklist of about 25 dimensions, includes items such as providing alternate transportation plans; maps demarcating assembly points; a description of the embassy warden system; a host government collaboration agreement that lists other government offices that could be used as a resource during an emergency; and methods for emergency communications. In our review of 65 EAPs (over 90 percent of total EAPs), we found that none of the EAPs we examined contained all of the dimensions listed in the EAP checklist, and, as illustrated in figure 4, many lacked key information. Recent Peace Corps reviews and inspector general evaluations have also identified numerous deficiencies in post EAPs, including inadequate emergency contact information, undeveloped emergency communication networks, and insufficient or nonexistent collaborative arrangements with the host country government--items called for in the EAP checklist. A Peace Corps official stated that some of the checklist items were not included in the EAP because they were not applicable. However, we found that these submitted EAPs did not explain why this information was not relevant. Figure 4: Percentage of Post EAPs Missing Selected Elements from Agency Checklist: [See PDF for image] Source: GAO analysis of post EAPs. [End of figure] The Peace Corps’ policy requires that all posts test their EAPs but does not establish detailed criteria for evaluating the results of the tests or for recording the results uniformly. The agency allows country directors discretion in making decisions in these areas. In the EAP guidelines, making contact with volunteers is one of the first steps in responding to a crisis. In some cases, posts set goals on time frames for reaching volunteers, either through communication technology or by travel to the volunteer, as benchmarks for measuring the test’s successfulness. For example, of the five country directors we interviewed, two had set targets for reaching at least 90 percent of their volunteers within 24 hours or less; both country directors achieved their goals. Our review of EAP test results showed that most tests are limited to sending a message to all volunteers during business hours and requesting that volunteers respond when they receive the message. According to a senior Peace Corps official, this does not indicate how the plan would work in a real emergency. As shown in figure 5, in our analysis of 63 EAP test results (over 90 percent of all results) submitted to headquarters, we found that 40 percent of posts did not provide information to headquarters on the length of time it took them to contact volunteers. Figure 5: Post Reports of Volunteer Contact Time in 2001 EAP Tests: [See PDF for image] Note: Of the EAP test results we received from the Peace Corps, 60 were conducted in 2001 and 3 were conducted in 2000. Source: GAO analysis of Peace Corps data. [End of figure] Underlying Factors Contribute to Uneven Implementation: Several factors contribute to the uneven implementation of Peace Corps’ safety and security policies. These factors include unclear guidance and weaknesses in safety and security training for staff and volunteer leaders, uneven application of supervision and oversight mechanisms, and turnover among U.S. direct hire staff. Guidance Is Not Always Clear and Staff Training Is Sometimes Inadequate: The Peace Corps’ safety and security framework outlines general requirements that posts are expected to comply with but does not often specify required activities, documentation, or criteria for judging actual practices. This may make it difficult for staff to understand what is expected of them. Many posts have not developed clear reporting and response procedures for incidents such as responding to sexual harassment. The agency’s coordinator for volunteer safety and security said that unclear procedures make it difficult for senior staff, including regional directors, to establish a basis for judging the quality of post practices. The coordinator also observed that, at some posts, regional safety and security officers had found that staff members did not understand what had to be done to ensure compliance with agency policies. Although the Peace Corps provides new staff with training on safety and security procedures, evidence suggests that staff training may not always be adequate. In addition, volunteer leaders and wardens who are assigned safety and security responsibilities are not always provided with relevant training. Program managers with whom we spoke found their initial 4-week overseas staff training useful. However, some country directors said that provisions could be strengthened for training lower level staff with significant safety and security responsibilities and for continuing the education of long-time program managers.[Footnote 17] A senior Peace Corps official agreed with the latter observation, noting that assessment of staff members’ long-term training experience was warranted. Peace Corps reports have also found that some volunteer leaders[Footnote 18] who assist in site selection and volunteer monitoring and who act as contact points in the event of an emergency do not receive adequate training and are not prepared to discharge their safety-related duties. Our interviews with volunteer leaders and wardens at five posts support this finding. For example, we visited one post where staff members relied on six volunteer leaders to play a significant role in developing sites and responding to volunteer concerns. Four of these volunteer leaders had held the position for several months, but the Peace Corps had not yet trained them for their duties. All of them expressed concern that post staff expected them to take the lead in site development even though they had not been trained to do this. At another post, we visited a volunteer warden whose site is a consolidation point in the event of volunteer evacuation to a neighboring country. She said that the Peace Corps had provided her with no training on her responsibilities in case of an emergency. Uneven Application of Supervision and Oversight Mechanisms: Informal supervisory mechanisms and a limited number of staff hamper Peace Corps efforts to ensure effective supervision and oversight. The agency has some formal mechanisms for documenting and assessing post practices, including the annual evaluation and testing of post EAPs and regional safety and security officer reports on post practices. Nonetheless, regional directors and country directors rely primarily on informal supervisory mechanisms, such as staff meetings, conversations with volunteers, and E-mail, to ensure that staff is doing an adequate job of implementing the safety and security framework. Several country directors and a former regional director stated that overreliance on informal communications can hinder adequate oversight of staff performance in key areas. One country director observed, for example, that it is difficult to oversee program managers’ site development or monitoring activities because the post does not have a formal system for overseeing. The Peace Corps’ limited use of written or computerized records compounds difficulties in supervising staff at posts and in identifying implementation problems, including noncompliance records that are kept but not always updated.[Footnote 19] Officials from the Inspector General’s office noted that their work revealed important disparities among posts in their ability to maintain computerized records, especially site histories and volunteer files. For example, one post we visited had created computerized record-keeping systems that permitted easy access to information on site visits and volunteer concerns, greatly facilitating effective supervisory review of the quality of staff support for individual volunteers over time. Another post was in the initial stages of creating such a system. Other posts, however, had no such systems and did not require staff to complete site visit reports to be filed by volunteer or location. Some posts we visited did not formally document nonassault crimes unless the volunteer reported the incident to the medical office. The Peace Corps’ regional safety and security officers and staff from the Inspector General’s office play an important role in helping posts implement the agency’s security framework. However, the number of staff in these offices limits their ability to provide input to posts. Staff at headquarters and at the posts where the agency’s three regional safety and security officers have provided assistance view these officers as a resource for enhancing volunteer safety and security. Officers’ visits to posts can include activities such as leading workshops with volunteers and post staff to assess security practices; training post staff and volunteers on safety and security issues; assisting posts in testing their EAPs and providing feedback on the results; and helping posts respond to specific safety and security challenges, such as preparing for national elections or reevaluating the security situation in light of changing country conditions. However, according to the Peace Corps, the officers provided input to only about one-third of the agency’s posts between October 2000 and May 2002.[Footnote 20] Oversight by the inspector general’s staff is also limited because of staffing levels. From December 1999 through December 2001, the inspector general issued reports containing findings on safety and security practices at 12 posts.[Footnote 21] In addition, the Peace Corps has no system to track post compliance with inspector general recommendations to ensure that they are properly implemented. However, according to agency officials, they are working to develop such a system. Turnover Hinders Implementation of Effective Practices: One factor that may contribute to the Peace Corps’ difficulty in implementing its safety and security policies is turnover among key managers. According to a June 2001 Peace Corps workforce analysis, turnover among U.S. direct hires was extremely high, ranging from 25 to 37 percent in recent years. This report found that the average tenure of these employees was 2 years, that the agency spent an inordinate amount of time selecting and orienting new employees, and that frequent turnover produced a situation in which agency staff are continually “reinventing the wheel.” The report attributed much of the problem to the 5-year employment rule, which statutorily restricts the tenure of U.S. direct hires, including regional directors, country desk officers, country directors and assistant country directors, and inspector general and safety and security staff. Several Peace Corps officials said that turnover affects the agency’s ability to maintain continuity in oversight of post operations. In addition, the lack of documentation described above, combined with high turnover, means that the agency is losing opportunities to apply lessons learned from previous staff tenures. Peace Corps Initiatives May Enhance Volunteer Safety and Security: In May 2002, the Peace Corps informed us of a number of initiatives that the agency had already taken or intended to take to improve its current safety and security practices. Peace Corps officials noted that these initiatives were generated through an agencywide safety and security review that began in fall 2001. The agency’s initiatives are intended to address many of the issues we identified and may lead to improved safety and security practices. However, the Peace Corps faces important challenges in implementing these initiatives, and their impact on agency practices remains to be seen. Peace Corps’ Initiatives Aim to Improve Performance: The Peace Corps’ initiatives are intended to improve the agency’s safety and security practices and make them more uniform. (See figure 6 for an overview of the Peace Corps’ initiatives announced in May 2002.) For example, they are intended to clarify guidance, strengthen supervision and oversight mechanisms, and provide human resources to help maintain documentation and perform research into patterns and trends in crime against volunteers. Figure 6: Summary of Peace Corps Safety and Security Initiatives of May 2002: [See PDF for image] [A] Volunteer staging occurs before volunteers leave the United States for their assigned countries. Source: GAO analysis of Peace Corps information. [End of figure] To support country director efforts, the agency plans to hire additional safety and security staff at all levels. At headquarters, the agency has stated that it will hire an associate director for safety and security who will have responsibility for overseeing all agency safety and security activities. To assist the new associate director, the Peace Corps increased its staff of field-based regional safety and security officers from three to seven in June 2002. The agency plans to add five more officers in 2003. To strengthen the agency’s ability to analyze and apply information on crime against volunteers, the Peace Corps has stated that it will provide the new associate director with a safety and security data manager/analyst who will research crime trends and related issues, in collaboration with the Office of Medical Services. To assist regional directors in supervising country director activities, the Peace Corps plans to provide each of the regional directorates with a headquarters-based security officer who will work with the country desk units to monitor and assist post efforts to ensure that their safety and security systems meet agency expectations. To provide full-time assistance at the country level, all posts have been authorized to hire safety and security administrative associates. The agency expects at least 35 posts to create such positions within a year. Among other things, these new staff members will assume responsibility for ensuring that posts maintain accurate and complete records on site histories, site visits, and criminal incident reports.[Footnote 22] To improve staff understanding of agency safety and security policies and requirements, a 2-year cycle of safety and security training has been authorized. This training will be delivered through an ongoing series of subregional workshops with six attendees from each post and led by field-based regional safety and security officers. A series of training sessions for country desk officers and other headquarters staff will be led by headquarters-based regional security officers. In addition, the agency has provided easier access to its safety and security guidance by placing all relevant materials in a single location on its agencywide intranet. Posts that do not have easy access to the Internet were provided with these materials on a compact disc, produced in February 2002. Implementation Process Remains to Be Fully Clarified: As the Peace Corps begins to implement its recently announced initiatives, it will face a number of important challenges. The agency has yet to fully clarify the criteria to be applied in evaluating the adequacy of agency practices or the mechanisms to be used in documenting and sharing information on its progress in attaining compliance with agency policies. The agency’s response to these challenges will have a major impact on its ability to ensure that its initiatives have their desired effect. The key to the Peace Corps’ developing a safety and security framework that achieves its desired goals is the effective implementation of the agency’s safety and security initiatives. Criteria for assessing whether the revised policies are being adequately implemented have yet to be fully defined. The Peace Corps has taken steps to clarify its policies and has improved and provided easier access to its guidance on implementing these policies. However, greater clarity could be provided without imposing detailed requirements that may be impractical or inappropriate in some countries. For example, revised agency guidance requires posts to include formal risk assessments in their EAPs. The agency has guidance available on preparing such risk assessments but does not have models available for posts to use. Similarly, the initiatives include authorization for posts to hire administrative associates who will be assigned various safety and security support tasks, including ensuring that the posts’ filing systems provide ready and complete access to relevant records. However, the agency has not developed criteria or examples for judging the adequacy of these filing systems. Conclusions: The Peace Corps is embarking on a major expansion of its volunteer workforce during a time of heightened risk for Americans living abroad. Providing safety and security for its volunteers is the Peace Corps’ highest priority. Our review of the agency’s efforts to ensure compliance with its basic safety and security policies and guidelines shows that there are cases of uneven implementation of key elements of the safety and security framework that could pose risks to volunteers. These include uneven performance in developing safe and secure housing and work sites, responding to volunteer concerns, and planning for emergencies. The Peace Corps has recently announced several new initiatives to improve overall compliance with its safety and security policies. We believe that, if effectively implemented, the new initiatives can reduce potential risks facing volunteers. However, it is not yet clear how the Peace Corps will document its progress in achieving compliance or will share information about better practices. While the Peace Corps does generate reports on practices at individual posts, the agency does not currently have a means to (1) document the overall quality of its safety and security practices or (2) assess changes in the quality of these practices over time. The initiatives do not contain provisions for formal assessments or for documenting progress in implementing them so that this information can be shared with staff. Moreover, the Peace Corps has not indicated what action, if any, it intends to take in addressing the issue of staff turnover. We believe that the Peace Corps will need to address the implications of staff turnover if it is to effectively implement its new initiatives designed to ensure the safety and security of its volunteers. Recommendations: To help ensure that the Peace Corps’ initiatives have their intended effect, we recommend that the Director develop indicators to assess the effectiveness of the initiatives and include the results of these initiatives in the agency’s annual reports under the Government Performance and Results Act. We also recommend that the director develop a strategy to address staff turnover as it implements its initiatives. Among other things, this strategy could include proposals to Congress to change the law concerning the 5-year limit on employment of U.S. direct hire staff. Agency Comments: In written comments on a draft of this report, reprinted in appendix V, the Peace Corps concurred with our findings and provided additional information on the agency’s safety and security initiatives and technical comments that we incorporated as appropriate. In response to our first recommendation, the Peace Corps agreed to report on the results of its safety and security initiatives in its annual reports under the Government Performance and Results Act. In response to our second recommendation, the Peace Corps stated that it had developed a strategy for mitigating the effects of high staff turnover as it implements its safety and security initiatives, but that unless the law concerning the 5-year rule is changed the agency cannot effectively address the difficulties presented by staff turnover. Given the agency’s position on this matter, we modified our recommendation to suggest that the Peace Corps submit a proposal to Congress for changes in the 5-year rule that would facilitate agency efforts to improve its safety and security practices. We are sending this report to interested congressional committees and the Director of the Peace Corps. We will also make copies available to other interested parties on request. In addition, this report will be available at no charge on the GAO Web site at http//www.gao.gov. Please contact me on (202) 512-4268 if you or your staff have any questions concerning this report. An additional GAO contact and staff acknowledgments are listed in appendix VI. Jess T. Ford Director, International Affairs and Trade: Signed by Jess T. Ford: [End of section] Appendixes : [End of section] Appendix I: Objectives, Scope, and Methodology: You requested that we evaluate the Peace Corps’ safety and security practices. In response, we (1) described rates and trends in crime against volunteers and reviewed the agency’s system for generating such information, (2) described the agency’s framework for maintaining volunteer safety and security, (3) evaluated the Peace Corps’ implementation of this framework and identified factors affecting this implementation, and (4) evaluated the agency’s initiatives to improve current practices. To address our first objective, we (a) examined agency reports on crime trends and characteristics of assaults from 1991 to 2001; (b) reviewed agency guidelines and interviewed medical services staff at headquarters and in the field to clarify the Peace Corps’ processes for gathering, analyzing, disseminating, and applying information; and (c) performed independent analyses of Peace Corps data to determine the extent to which agency findings accurately reflect information from the field and to explore opportunities for additional useful analyses. To perform our independent analyses, we obtained computer files containing original crime data for 1990 through 2001 and excerpts from the Peace Corps’ administrative database on the numbers of volunteers serving during this period and characteristics such as age, gender, and date of entry into service. We used these data to replicate the Peace Corps’ analyses of crime rates and characteristics of assaults, finding that our results were consistent with the Peace Corps’. We also examined the data for missing elements, mislabeled data, and related problems. We found a number of technical problems; for example, inconsistencies in coding sometimes made it difficult to distinguish between missing values and those that were incorrectly coded. However, these problems did not materially affect the Peace Corps’ analyses. To obtain information on underreporting, we reviewed relevant portions of the Peace Corps’ volunteer satisfaction surveys for 1998 and 1999 and interviewed agency staff and volunteers. We interviewed agency staff on the potential usefulness of additional analyses and explored the data made available to us to identify trends or relationships that merit further inquiry. We did not attempt to verify the accuracy or completeness of data collection among medical officers at individual posts. To present a clear description of the agency’s framework for maintaining volunteer safety and security, we reviewed agencywide policies and guidance materials that are provided to post staff, such as handbooks and examples of best practices. We also examined materials that the agency uses in training staff to carry out their safety and security responsibilities. We interviewed key headquarters staff, including regional managers, country desk officers, general counsel and medical office officials, and the agency’s coordinator for volunteer safety and security about their roles and responsibilities and the manner in which agency policies and guidance materials are applied in practice. To obtain broader perspectives on safety and security challenges in developing countries and options for responding to those challenges, we spoke with security specialists at the Department of State in Washington, D.C., and with U.S. embassy security officers in the countries we visited, listed below. We also spoke with headquarters or field-level staff, or both, from a number of organizations that face security and safety challenges similar to those faced by the Peace Corps, including the Japanese Overseas Cooperation Volunteer Program, the British Volunteer Service Organization, and the United Nations’ Volunteer Program. We attended a conference on security practices for nongovernmental organizations sponsored by the American Red Cross. To evaluate the Peace Corps’ implementation of its safety and security framework, we obtained documents from and interviewed headquarters and field-level staff and volunteers. We visited posts in Bulgaria, El Salvador, Kenya, Senegal, and Ukraine to examine safety and security practices. At these posts, we interviewed agency staff with significant safety and security responsibilities, including country directors, program managers, and medical officers, and the three regional safety and security officers employed by the Peace Corps at the time of our work. We examined post record-keeping procedures and relevant files. We spoke with more than 150 volunteers, visiting more than 30 at their sites and speaking with their local counterparts when possible. To broaden our understanding of Peace Corps practices beyond the countries we were able to visit, we consulted the results of the Peace Corps’ worldwide volunteer satisfaction surveys for 1998 and 1999, all 12 reports issued by the agency’s inspector general between December 1999 and December 2001 that contained findings on safety and security issues, and reports on relevant issues at 24 posts generated by the agency’s safety and security staff between September 2000 and November 2001. We examined nine assessments of the security environment in individual countries prepared between 1996 and 2001. In addition, we obtained and analyzed documentation on specific safety and security functions at multiple posts when it was available. For example, we examined 65 post emergency action plans (EAP) and headquarters’ feedback on these plans, and we reviewed site development criteria and procedures from 18 posts in the Peace Corps’ Inter-America/Pacific region, in addition to those from the posts we visited. To evaluate the Peace Corps’ recently announced safety and security initiatives, we obtained and reviewed documentation on the initiatives and the Peace Corps’ efforts to clarify and provide easier access to agency policies and guidance materials. We met with the Peace Corps’ Director and other senior staff to discuss the substance and intent of the proposed measures. We conducted our work from July 2001 through May 2002 in accordance with generally accepted government auditing standards. [End of section] Appendix II: Other Organizations’ Safety and Security Approaches Vary According to Work and Goals: Organizations that assign personnel to live and work abroad can draw from three basic strategies to develop safety and security procedures: * acceptance--reducing the risk level by integrating into a host community; * protection--reducing vulnerability by employing protective devices, such as walls and locks; and: * deterrence--eliminating threats by posing a counterthreat, for example, by employing armed guards. Organizations that emphasize person-to-person cultural exchange as a major goal tend to rely on the acceptance approach to safety and security; they seek to enhance safety and security primarily by ensuring that individuals are accepted as members of host communities. Nonetheless, these organizations may differ substantially in the details of their approach. As organizations become less concerned with establishing person-to-person ties within a host community and more concerned with achieving specific technical or development goals, they may place more emphasis on protection and, sometimes, deterrence measures.[Footnote 23] The following are descriptions of strategies employed by organizations that face safety and security challenges similar to those faced by the Peace Corps--the Volunteer Service Organization, the Japanese Overseas Cooperation Volunteers, the United Methodist Volunteers in Mission, the foreign mission program of the Church of Jesus Christ of Latter-day Saints, the United Nations Volunteers, and Save the Children. The Volunteer Service Organization is a British nongovernmental organization whose goals and safety and security approach are similar to the Peace Corps’, with a few key differences. The organization maintains 2,000 volunteers in 71 countries for average tours of 2 years, mostly in rural areas or provincial towns. Like the Peace Corps, the agency seeks to fight poverty and promote international understanding. In contrast to the Peace Corps, the organization advertises and recruits on a job-by-job basis in response to specific requests from counterpart organizations in developing countries. The organization thus faces less of a challenge than the Peace Corps in finding productive employment and supportive organizations for volunteers. Volunteers average 38 years of age and are often experienced. Although its approach to identifying housing and monitoring volunteers is similar to the Peace Corps, the organization provides less safety and security training. It provides general risk- awareness training before volunteers’ departure for their country of service and limited country-and placement-specific risk awareness and management training upon volunteers’ arrival in the country. In contrast to the Peace Corps, which has EAPs in all of the countries where it operates, the organization has EAPs only in countries where such plans are deemed necessary. The Japanese Overseas Cooperation Volunteers also resembles the Peace Corps in its goals and approach to safety and security, with some differences. The organization operates in more than 70 countries under the aegis of the Japan International Cooperation Agency, that country’s bilateral development agency. Similar to the Peace Corps, this program sends volunteers to spend 2 years working in agriculture, civil engineering, health, and other program areas. Unlike Peace Corps volunteers, the Japanese volunteers are considered government employees. Like the Volunteer Service Organization, the Japanese program recruits volunteers for individual jobs and therefore has fewer difficulties with finding suitable jobs for its volunteers. The program does not have a formal policies and procedures manual, although it has been consulting with the Peace Corps on the development of such a manual. The organization uses a five-step classification system to assess risks in specific countries and develops actions to take on the basis of risk level. Program officials stated that the agency provides volunteers with a 3-month training program in Japan, which includes some safety and security training, but the agency provides little, if any, in-country training. Volunteers might use cell phones, satellite phones, radios, or other communication tools; the organization strives to ensure that each volunteer can be reached within 6 hours. The Peace Corps has no such minimum standard. Program officials participate in their parent organization’s EAPs. The United Methodist Volunteers in Mission, while citing intercultural exchange and relationship building as a goal, differs significantly from the Peace Corps in that volunteers generally serve only 1 to 6 months and thus have less time to integrate into a community. This church-sponsored organization, part of the United Methodist Committee on Relief, recruits volunteers to work in areas such as education and construction. Unlike the Peace Corps, these volunteers pay a fee to the Committee on Relief to cover costs, including housing and food, while in the country where they are placed. Most Methodist-sponsored volunteers are middle-aged through retirement age. A program official indicated that the safety and security training the organization provides is not as intense as the Peace Corps’ because volunteers are generally in the country for only a short time; the organization provides some information on cultural sensitivity before volunteers’ departure and an orientation when they arrive in country. Although it is not always possible for volunteers to be in daily contact with office staff, one individual with the volunteers is responsible for them on a 24-hour basis and can contact the office whenever needed. The Church of Jesus Christ of Latter-day Saints sends volunteers to do mission work worldwide. The majority of the volunteers are male and all are young--the upper age limit is 26. A church official indicated that the church provides little training in safety and security. The church monitors volunteers frequently to ensure their safety. Unlike the Peace Corps, church volunteers always travel and live in pairs and report to the in-country mission on a weekly or daily basis, depending on the risk level of the country. Volunteers also have support from local church members in the community in which they serve. Most volunteers have telephone lines in their apartments, but they are not supposed to have cell phones or radios because officials think these items could make volunteers targets for theft and assault. United Nations Volunteers operate under the auspices of the United Nations Development Program. Volunteers generally work on a program project alongside program staff and, much like the program’s regular employees, are chosen for a specific job. United Nations Volunteers are not asked to build intercultural relationships. About 5,000 of these volunteers are currently working in about 150 countries; many are native to the country in which they work. Volunteers usually serve for 2 years, although the program uses some short-term volunteers in times of crisis. Unlike Peace Corps volunteers, these volunteers usually live in the same communities as other United Nations or government staff, often in capital cities or urban areas; many bring their families and are given the use of a vehicle. Program officials stated that they do not perform formal risk assessments, but they added that they do not place volunteers in countries or areas that are considered dangerous. Program officials indicated that they provide little safety and security training, although the United Nations provides a safety and security handbook to staff members and volunteers in the United Nations system. There is little formal monitoring of volunteers. Volunteers typically have telephones in their homes and may also have cell phones or radios for project-related reasons. Save the Children is a development-oriented, nongovernmental organization with offices in about 31 countries, with staff focused on a specific job, not on intercultural exchange. In contrast to Peace Corps volunteers, most expatriate staff have had overseas experiences and are typically in their 30s. Much of the organization’s funding is from the U.S. Agency for International Development, and staff typically work closely with agency and U.S. embassy staff. The organization has not made it a practice to conduct formal risk assessments but instead relies on other nongovernmental organizations and the U.S. embassy for information. However, headquarters is beginning to task overseas offices with responsibility for conducting risk assessments. Although program officials indicated that the organization provides little training in safety and security, they have asked the Peace Corps and other U.S. government agencies for advice on training. Unlike Peace Corps volunteers, Save the Children staff live in the expatriate community and may have radios, cell phones, or both, depending on job needs and risk. They have frequent contact with other nongovernmental organizations and U.S. government employees, who live and work in the same area. In addition, country directors prepare weekly reports on staffs’ current and future locations and vacation schedules. [End of section] Appendix III: Peace Corps’ Crime Data Analysis System and Crime Trends: The Peace Corps has established two reporting systems for collecting information on crimes against volunteers. The agency’s medical staff operates both systems. As described in this report, Peace Corps data show that, with the exception of major sexual assaults, reported rates of assault against volunteers have been higher in recent years than in the early 1990s. Historical data for aggravated assaults and rapes--the most consistent data available to Peace Corps analysts--support these overall findings. Reported rates of nonassault crimes, in contrast, have remained essentially unchanged since 1990. Peace Corps Has Two Systems for Collecting Crime Information: Post medical officers are tasked with collecting detailed information on each assault incident reported by volunteers and submitting this information to headquarters through the Peace Corps’ assault notification and surveillance system. In 1997, the medical office refined the reporting categories employed in this system. Formerly asked to differentiate among only four types of assaults, field medical staff are now asked to submit reports on five types of sexual assault and five types of physical assault. When filling out reporting forms, medical officers are asked to ascertain a variety of details on victims, assailants, and the circumstances surrounding each assault, such as time and location of the incident. Medical officers are also asked to submit monthly counts of four types of nonassault crimes through the Peace Corps’ epidemiologic surveillance system, which is a reporting system that focuses primarily on gathering statistics on volunteer injuries and illnesses.[Footnote 24] These reports do not provide any details on the reported events. Trends Differ for Aggravated Assault, Rape, and Nonassault Crimes: Aggravated assault and rape are the only two categories of crime for which reporting definitions remained unchanged when the Peace Corps revised its system for categorizing and recording crimes in 1997. Therefore, data on these crimes may be the most consistent available to the Peace Corps. As shown in figure 7, the reported rate of aggravated assault against volunteers has been consistently higher since 1996 than in earlier years. As shown in figure 8, reports of rape have varied from year to year, most recently declining from a median rate of about 4.6 per 1,000 female volunteer years in 1996-1998 to a median level of about 3 per 1,000 female volunteer years in 1999-2001.[Footnote 25] Table 1 shows the actual numbers of aggravated assaults and rapes that were reported. Since the numbers of assaults, especially sexual assaults, are small, there is some question about the practical significance of these changes. Figure 7: Incidence Rates of Reported Aggravated Assaults, 1990-2001: [See PDF for image] Legend: V/T year = 1 full year of service by a volunteer or trainee. Source: Peace Corps analysis (except for 2001, which represents a GAO estimate of Peace Corps data). [End of figure] Figure 8: Incidence Rates of Reported Rape, 1990-2001: [See PDF for image] Legend: V/T year = 1 full year of service by a volunteer or trainee. Source: Peace Corps analysis (except for 2001, which represents a GAO estimate of Peace Corps data). [End of figure] Table 1: Number of Aggravated Assaults and Rapes Reported by Volunteers, 1990-2001: Year: 1990; Aggravated assaults: 20; Rapes: 11. Year: 1991; Aggravated assaults: 41; Rapes: 13. Year: 1992; Aggravated assaults: 43; Rapes: 9. Year: 1993; Aggravated assaults: 57; Rapes: 10. Year: 1994; Aggravated assaults: 45; Rapes: 6. Year: 1995; Aggravated assaults: 62; Rapes: 9. Year: 1996; Aggravated assaults: 78; Rapes: 13. Year: 1997; Aggravated assaults: 88; Rapes: 20. Year: 1998; Aggravated assaults: 87; Rapes: 19. Year: 1999; Aggravated assaults: 102; Rapes: 11. Year: 2000; Aggravated assaults: 88; Rapes: 19. Year: 2001; Aggravated assaults: 86; Rapes: 14. Source: Peace Corps. [End of table] Rates of nonassault crimes have varied little since 1993, when the agency began collecting information on incidents of burglary, theft, and robbery. Figure 9 shows a slight decrease in reported robberies and burglaries since 1993, while figure 10 shows a slight increase in reported thefts. Figure 9: Incidence Rates of Reported Robberies and Burglaries, 1993- 2000: [See PDF for image] Legend: V/T year = 1 full year of service by a volunteer or trainee. Note: A robbery is an event that does not involve violence or threat of violence, in which property or cash is taken directly from a volunteer. A burglary is an unlawful or forcible entry or attempted entry of a volunteer’s residence. Source: GAO analysis of Peace Corps data. [End of figure] Figure 10: Incidence Rates of Reported Thefts, 1993-2000: [See PDF for image] Legend: V/T year = 1 full year of service by a volunteer or trainee. Note: A theft is the taking away of, or attempting to take away, property or cash without force or illegal entry. Source: GAO analysis of Peace Corps data. [End of figure] [End of section] Appendix IV: Peace Corps Provisions for Responding to Criminal Incidents: Peace Corps policy requires posts to develop procedures for responding to all safety and security incidents reported by volunteers. The agency has not developed clear guidance for posts to apply in responding to minor incidents. However, the Peace Corps does have well-defined notification and response protocols for major sexual assaults, and posts follow similar procedures when volunteers report major physical assaults. In addition, when a volunteer decides to prosecute, the Peace Corps’ Office of General Counsel and the Office of the Inspector General’s investigations unit may provide assistance. The Peace Corps’ Rape Response Handbook, developed in 1999, establishes a protocol to ensure timely notification of appropriate staff at posts and at headquarters and describes the roles and responsibilities of post and headquarters staff in responding to a rape or attempted rape. In addition to giving guidance for reporting the incident to agency headquarters as previously described in this report, the handbook clearly establishes that the post’s medical officer is responsible for providing medical care to the volunteer who has been assaulted and for collecting forensic evidence in case the volunteer decides to prosecute. The country director is responsible for ensuring that the victim, as well as other volunteers and trainees, is safe; preserving the option to prosecute (e.g., by advising the volunteer of her legal rights, preserving evidence, etc.); and notifying the security office at the U.S. embassy of the assault while protecting the volunteer’s identity unless identification is essential. Embassy security staff are expected to support the Peace Corps in any investigation or prosecution following the incident. The Peace Corps follows similar notification and response protocols when a volunteer reports a major physical assault. The medical officer reports the assault to the Office of Medical Services at headquarters and provides medical treatment to the volunteer. As with a rape incident, the medical officer notifies the country director of the assault, although in the interest of medical confidentiality the volunteer’s identity and details of the incident may not be disclosed. The country director is responsible for informing the U.S. embassy security officer of the assault and may work with the embassy if the volunteer decides to prosecute. According to Peace Corps data, 18 percent of volunteers who experienced a major sexual assault and 26 percent of volunteers who reported a major physical assault between 1997 and 1999 said that they intended to prosecute. When a volunteer decides to prosecute, the Peace Corps’ Office of General Counsel covers the cost of legal counsel in the country where the assault happened. The Office of the Inspector General’s investigations unit, in conjunction with other federal agencies, may also provide support in investigations of crimes against volunteers. For example, inspector general staff may conduct interviews with Peace Corps staff and local authorities, escort volunteers who are asked to identify suspects, or arrange for examination of forensic evidence. [End of section] Appendix V: Comments from the Peace Corps: THE DIRECTOR OF THE PEACE CORPS WASHINGTON, D.C. July 16, 2002: Mr. Jess T. Ford Director International Affairs and Trade General Accounting Office 441 G Street, N.W. Washington, D.C. 20548: Re: Agency Response to GAO Report on Peace Corps Safety and Security: Dear Mr. Ford: On behalf of all of our Volunteers and staff, I thank you for your thorough Report on Peace Corps’ safety and security program. We agree with many of the Report’s findings, and we would like to take this opportunity to provide certain additional information to make the Report more complete and accurate. For ease of review, this letter sets out our comments according to the Report’s overall organizational structure. (We have provided technical comments under separate cover.): Background: Any report on Peace Corps’ safety and security system should begin by describing the reality of the Peace Corps Volunteer experience, as this explains why the Peace Corps has based its entire safety and security approach on the community acceptance and integration model. Peace Corps Volunteers are individuals who volunteer to work directly with local schools, health clinics, community centers, or other host-country sponsoring organizations. They live in the same local communities where they volunteer, at the same socio-economic level as the people with whom they work, sometimes in remote locations. Peace Corps is committed to maintaining their safety and security by providing them with a safety and security framework and an infrastructure of support- including pre-service training in language, safety and security, and cross-cultural issues; direct medical care in the field (and medical evacuation to the United States if needed); assistance with housing; a living stipend; and other institutional support. Reported Crime Incidents: We are gratified the Report acknowledges that the Peace Corps systematically gathers and analyzes a range of information about criminal incidents against Volunteers; that it “has used its data analyses to gain insight into the characteristics of assaults against: Volunteers and to shape Volunteer training programs;” and that “in recent years, Peace Corps has made a number of improvements to its crime data collection and analysis system.” (Report pages 5, 8.): We agree with the Report’s later assertions that further analysis into this issue could yield useful information. Peace Corps Safety and Security Guidance and Support to Posts: We are pleased the Report recognizes that “Volunteer health, safety, and security is the agency’s highest priority,” and that, to advance these goals, the agency “has adopted policies for monitoring and disseminating information on the security environments in which the agency operates, training Volunteers, developing safe and secure Volunteer housing and work sites, monitoring Volunteers, and planning for emergencies such as evacuations.” (Report page 10.) We also appreciate the Report’s finding that Peace Corps headquarters supports posts in implementing these policies through written guidance, training, supervision and oversight. (Report page 12.): Peace Corps Posts’ Implementation of Safety and Security Framework: We agree with one of the Report’s conclusion that several aspects of our safety and security policies are being applied well across our numerous posts. Thus, for example, we appreciate that the Report found that numerous Volunteers are generally satisfied by the degree to which posts keep them informed about safety and security issues in their country of service and with the level of safety and security training Peace Corps provides to them. (Report page 14.): Our own safety and security review confirms the Report’s finding that, for certain other areas of the agency’s safety and security framework, full implementation is uneven across our many posts. It is for this very reason that the agency has undertaken several major new initiatives to ensure more uniformity of compliance with these policies, including the issuance of Peace Corps Manual Section 270 Implementation Procedures, which sets out standards and criteria for compliance with each element of the MS 270 framework in a manner that is verifiable for Peace Corps headquarters. We note each area of concern cited by the Report and what steps the agency has already taken to address that issue: The Report found that not all posts had properly developed housing and work sites for Volunteers. The new MS 270 Implementation Procedures require that each Country Director certify in writing, among other things, that site selection criteria, site selection procedures, and Volunteer site visits standards and procedures have all been established and applied. The Report found that some Volunteers had unsupportive counterparts, which may limit their ability to build supportive networks in their host communities. The new MS 270: Implementation Procedures require the Country Director to certify in writing that appropriate communities and counterparts are identified and provided with orientation to the Peace Corps program and to the particular Volunteer assigned to that community. The Report found that not all posts maintained proper site history files. The new MS 270 Implementation Procedures require the Country Director to certify in writing that post has established and is maintaining appropriate site history files, which contain all relevant safety and security information on a site-by-site basis. The Report points out a number of concerns relating to three additional issues: (1) frequency of site visits; (2) Volunteer absences from sites; (3) Peace Corps’ responses to safety and security concerns and criminal incidences. Regarding site visits, we were pleased that, at each of the five posts you visited, “staff made regular visits to most Volunteers, in accordance with each post’s policies,” and that many Volunteers were satisfied with the frequency of site visits. (Report page 18.) In addition, the new MS 270 Implementation Procedures require the Country Director to certify in writing that Volunteer site visit standards are established and applied. We also agree that there are safety and security challenges posed by a range of Volunteer behaviors, including the specific issue of unauthorized absence from site. In response to these challenges, Peace Corps has in place the following policies and procedures: the new MS 270 Implementation Procedures require the Country Director to certify in writing that (i) each Volunteer has demonstrated an understanding of host country cultural norms and an ability to adapt his or her daily conduct accordingly; (ii) each Volunteer has demonstrated an understanding of personal safety risk factors and a commitment to adopt a safe lifestyle and minimize risk; (iii) each Volunteer has met training competencies relating to safety and security (including leave from site, transportation, drinking and other personal behavior); and (iv) each Volunteer has completed and submitted a site locator form. In addition, all posts have clear written policies stating that unauthorized absence from site is grounds for administrative separation from Peace Corps and, pursuant to the MS 270 Implementation Procedures, counterparts are made aware of the importance of these policies. Regarding Peace Corps’ responsiveness to safety and security concerns and criminal incidents, the Report acknowledges that the Volunteers you spoke with who were victims of sexual assault “expressed satisfaction with staff response when they reported the incident.” (Report page 18.) Additionally, the new MS 270 Implementation Procedures require the Country Director to certify in writing that uniform incident reporting and incident response protocols are established and applied; that incident report data are analyzed and appropriate recommendations applied; and that procedures for sharing incident report data with the Embassy Regional Security Officer and Peace Corps/Washington are established and applied. Regarding post Emergency Action Plans (“EAPs”), the Report found that all posts complied with the MS 270 requirements that they “develop an EAP, test it annually, and submit it and the test results to headquarters.” (Page 19.) The Report also acknowledges that Peace Corps headquarters has made available to posts a suggested format designed to assist the posts in formulating their EAPs. Additionally, though the Report notes that Peace Corps has evacuated more than 1600 Volunteers from 26 posts during the years from 1993 to 2001, the Report should also note that, in all of these evacuations, Peace Corps successfully evacuated every Volunteer without a single injury or death. Peace Corps has already taken several steps to ensure greater uniformity of compliance with the recommended elements of a strong EAP: (1) the Peace Corps is adding three new full-time positions of Safety and Security Desk Officer to provide specific training to the desks on improving their posts’ EAPs and to provide additional support for the review and revision process for each post’s EAP; (2) additional materials for developing and evaluating complete EAPs have been made available to posts; and (3) pursuant to the new MS 270 Implementation Procedures, Country Directors must certify in writing that their EAPs have been submitted to and made part of the Embassy’s Emergency Action Plan, and that staff and Volunteers have received appropriate training to fulfill their respective responsibilities under the EAP. Underlying Factors Contributing to Uneven Implementation: Guidance and Staff Training. According to the Report, Peace Corps staff in the field sometimes found it difficult to understand exactly what “activities [or] documentation” was required to comply with the broad policies set out by MS 270, and the criteria by which such compliance would be determined. They also thought that training in this area could be improved. (Report page 22.) As part of the agency’s own safety and security review, headquarters and field personnel together crafted the new MS 270 Implementation Procedures, which detail precisely the “activities, documentation, or criteria” for judging posts’ compliance with the MS 270 requirements. These implementation standards and procedures are now part of all Overseas Staff Training, Country Desk Unit Training, RSSO training, and built into the compliance responsibilities of the Safety and Security Desk Officers, and the Chief Compliance Officer within the Office of Safety and Security. We also agree that the issue of training for Peace Corps Volunteer Leaders and wardens is still uneven and greater uniformity of training should be implemented. We have recently distributed to all posts a new training module for Volunteer wardens covering their emergency action plan responsibilities. Supervision and Oversight Mechanisms. We appreciate the Report’s acknowledgment that the agency has undertaken several new initiatives to improve formal supervision and oversight of posts’ compliance with our safety and security framework. Specifically, we are greatly increasing the number of safety and security-related staff and re- aligning the agency’s structure to better monitor compliance. As spelled out in more detail below, we are authorizing, and providing funding for, every one of our 70 overseas posts to employ a full-or part-time staff assistant for safety and security (classified as a “Staff Support Specialist - Security”); as noted previously, we are adding three new full-time Safety and Security Desk Officer positions, one for each region; and we are creating a new division within Headquarters called the Office of Safety and Security, to be led by a new full-time position of Associate Director for Safety and Security that reports directly to the Peace Corps Director. This division encompasses the current office of Volunteer Safety and Overseas Security, and has new full-time positions of Chief Compliance Officer, Data Management/Analyst, as well as four additional Regional Safety and Security Officers (to bring the total from three to seven). Each post’s safety and security assistant will facilitate Peace Corps/ Washington’s ability to effectively supervise and oversee posts’ compliance with MS 270 by, among other things: maintaining documentation of compliance with the MS 270 Implementation Procedures, including formal tracking systems for items such as the site locater forms, site visits, site history forms, Volunteer training completion records and Volunteer signed competency forms; and following up with the relevant Associate Peace Corps Director or Country Director to ensure timely submission of required documentation. It is important to note that, although these individuals will greatly facilitate the post’s record keeping responsibilities, Country Directors will continue to carry ultimate responsibility for compliance with MS 270. The three new full-time Safety and Security Desk Officers will facilitate headquarters’ ability to effectively supervise and oversee each post’s compliance with MS 270 by, among other things: (i) working with each desk unit to directly monitor MS 270 compliance, including review and revisions to posts’ EAPs, and providing reports to Regional Directors and other senior officials as needed; (ii) developing tracking systems of each post’s MS 270 compliance on a regional level; (iii) providing training to desk units and other regional staff on continually improving MS 270 compliance; and (iv) assisting in tracking and implementing safety and security recommendations from RSSO evaluations. The new headquarters Office of Safety and Security will facilitate headquarters’ ability to effectively supervise and oversee posts’ compliance with MS 270 by, among other things: (i) centralizing and improving accountability of functions related to safety and security; (ii) bringing the authority of a senior executive service level manager, who reports directly to the Peace Corps Director, on the issue of Volunteer safety; (iii) directly supervising the work of the Office of Volunteer Safety and Oversees Security; and (iv) adding the new positions of Chief Compliance Officer, Data Management/Analyst, and four new Regional Safety and Security Officers. Among other responsibilities, the Chief Compliance Officer will track safety and security recommendations generated from Inspector General reports, Regional Safety and Security Officer evaluations, and other sources, as well as the status of post implementation of recommendations, and prepare monthly compliance reports for the Director’s review. The Data Management/Analyst will, among other things, design, implement and revise surveillance and monitoring systems; conduct trend analyses for selected safety conditions or events; and provide safety-related data gathering and analysis training. And the four new Regional Safety and Security Officers will continue and expand the work of the three Regional Safety and Security Officers already in place they will conduct individualized safety and security assessments of each post, provide recommendations for improving safety and security practices, including guidance to the new safety and security staff assistants on maintaining appropriate documentation of compliance, and generally be a resource to posts in fulfilling their MS 270 obligations. The Problem of Staff Turnover. We agree with the Report’s finding that staff turnover presents a challenge to the effective functioning of all agency operations, including providing for the safety and security of our Volunteers. By law, the Peace Corps is barred from employing any U.S. direct hire (other than the Director and Deputy Director of the agency) for more than five years. See 22 U.S.C § 2506(a)(5), (6).1The Report correctly notes that this “five-year rule” has created very high rates of staff turnover. Unless and until this law is changed, the agency cannot effectively “address the issue of staff turnover,” as the Report recommends. (Report page 29.): Instead, the agency has developed a strategy for mitigating the effects of high staff turnover, to as great a degree as possible, through measures such as increased training and better tracking systems. For example, we have recently instituted a regular two-year in-service safety and security training cycle for all desk units and other regional staff, U.S. recruiting offices, and in-country staff and Volunteer Leaders. In addition, we have expanded safety and security training as part of the one-month Overseas Staff Training given to all new overseas staff members, and developed safety and security modules to be presented at annual in-service trainings for programming and training staff, Administrative Officers, and Country Directors. We also believe that our compliance tracking systems will more deeply institutionalize our safety and security program, which should help mitigate the problems caused by frequent staff turnover. Peace Corps Initiatives May Enhance Volunteer Safety and Security: We agree with the Report’s comment that the agency’s initiatives “are intended to address many of the issues we identified and may lead to improved safety and security practices.” The Report recognizes that these initiatives are intended “to clarify guidance, strengthen supervision and oversight mechanisms, and provide human resources to help maintain documentation and perform research into patterns and trends in crime against Volunteers.” (Report page 25.) We have already identified many of these initiatives in our responses. Additionally, we think that, in many respects, the initiatives described above do have criteria and standards for measuring compliance. For example, MS 270 requires that each post establish and apply site selection criteria; the MS 270 Implementation Procedures require that, before a Volunteer is placed at a site, post must complete a site assessment checklist and housing checklist, which will confirm that the established criteria have been satisfied. Likewise, MS 270 requires that each post establish and apply site selection procedures; the MS 270 Implementation Procedures require that, before a Volunteer is placed at a site, there is a site assessment report, which details whether the: [1] Under extraordinary circumstances, extensions of employment, for up to three and one-half additional years of service, may be given to a limited number of employees. site complies with established written criteria. Each “Activity” described in the MS 270 Implementation Procedures has a similar “Means of Verification” section, which spells out precisely what is required to demonstrate compliance with the relevant policy. Of course, as these initiatives become more fully implemented, additional criteria may become useful. We plan internal quarterly reports on our progress towards full compliance, including documentation of compliance with specific requirements, and we will include results in our regular communications with Congress and the annual reports under the Government Performance and Results Act. Finally, with the benefit of increased analytical resources, we will continue to monitor our safety and security program both through the prism of compliance with various processes as well as the ultimate yardstick of outcome-based performance. In closing, let me again thank you for the time and energy GAO has devoted to the issue of Volunteer safety and security. Throughout my tenure as Director, it will remain my highest priority. Sincerely: Gaddi H. Vasquez: Signed by Gaddi H. Vasquez: [End of section] Appendix VI: GAO Contact and Staff Acknowledgments: GAO Contact: Phyllis Anderson (202) 512-7364: Staff Acknowledgments: In addition to Ms. Anderson, key contributors to this report were Wendy Ahmed, Kriti Bhandari, Lynn Cothern, Suzanne Dove, Bruce Kutnick, Michael McAtee, James Strus, and Christina Werth. FOOTNOTES [1] One volunteer year is equivalent to 1 full year of service by a volunteer or trainee. [2] The Peace Corps Act, as amended (codified at 22 U.S.C. 2501 et seq.), directs the agency to “help the peoples of interested countries in meeting their needs for trained manpower, particularly in meeting the basic needs of those living in the poorest areas of such countries, and to help promote a better understanding of the American people on the part of the peoples served and a better understanding of other peoples on the part of the American people.” [3] From 1993 through 2001, the Peace Corps closed or suspended operations at 24 posts because of civil unrest or warfare. The agency occasionally sends former volunteers to provide short-term “Crisis Corps” assistance during humanitarian crises or natural disasters. In recent years, such volunteers have been sent to Bosnia-Herzegovina, Guinea, and hurricane-devastated areas of Central America. [4] See, for example, Peace Corps, Safety of the Volunteer 2000 (Washington, D.C.: 2001) Assaults Against Peace Corps Volunteers 1990- 1999 (Washington, D.C.: 2001). [5] Since relatively small numbers of criminal incidents take place in relation to the size of the volunteer population, and annual changes in the number of reported crimes are also small, annual changes in underreporting can significantly affect apparent trends. [6] Peace Corps officials stated that after completing the most recent survey, the agency will conduct surveys every other year. [7] Peace Corps program managers are often known as associate Peace Corps directors. [8] The Peace Corps’ Strategic Plan for fiscal years 2000 to 2005 states that ensuring the health, safety, and security of volunteers and staff is the agency’s highest priority. [9] Section 270 of the Peace Corps Manual contains the agency’s basic policies on volunteer safety and security. [10] The three regional directorates are responsible for agency posts in Africa; Europe, the Middle East, and Asia; and Inter-America and the Pacific. [11] Volunteers are considered U.S. government employees only for certain purposes, including federal tort claims. See 22 U.S.C. 2504(h). [12] In addition to keeping volunteers informed, country directors may apply such information in making management decisions. For example, posts sometimes restrict volunteer travel to specific areas because of criminal or political violence. [13] For example, security officers recommended developing written course outlines or lesson plans and taking steps to comply with Peace Corps policy on integrating safety and security issues into the language and cross-cultural components of training. [14] According to the 1999 survey, 51 percent of volunteers reported sexual harassment, 41 percent reported racial harassment, 23 percent reported religious harassment, and 21 percent reported political harassment. [15] Some posts require host communities to provide housing for the volunteer as a sign of commitment to the volunteer’s work in the community. In interviews, many program managers said that in practice the communities often do not select and prepare a house until just before or even after the volunteer’s arrival. This practice makes it impossible for Peace Corps staff to comply with the agency’s requirement that they inspect all housing before the volunteer’s arrival. [16] See U.S. General Accounting Office, Peace Corps: New Programs in Former Eastern Bloc Countries Should Be Strengthened, GAO/NSIAD-95-6 (Washington, D.C.: Dec. 19, 1994); Peace Corps: Meeting the Challenges of the 1990s, GAO/NSIAD-90-122 (Washington, D.C.: May 18, 1990); and Changes Needed for a Better Peace Corps, GAO/ID-78-26, (Washington, D.C.: February 1979). [17] Country directors and medical officers attend regularly scheduled conferences, while program managers are provided with opportunities to attend training events every 12 to 18 months. These sessions may or may not address safety and security concerns. The Peace Corps’ coordinator for volunteer safety and security conducted a series of safety and security workshops for staff and volunteers in 1999 and 2000, but these workshops have not been repeated. [18] Volunteer leaders conduct administrative and volunteer support tasks for Peace Corps staff. These leaders are usually volunteers in their third year of service, although at one post we met volunteer leaders who were in their second year of service, and one who had not yet completed his first year. [19] According to our internal control standards, a well-designed internal control system should include “written documentation…for all significant transactions and events.” U.S. General Accounting Office, Internal Control Management and Evaluation Tool, GAO-01-1008G (Washington, D.C.: Aug. 8, 2001). [20] During the same period, the officers also provided input on the security environment in 18 countries where the Peace Corps was considering opening (or re-opening) a post. [21] Half of these reports focused specifically on safety and security issues. [22] As part of a larger effort to revamp the agency’s computer systems, Peace Corps information technology officials told us that they are developing an information management application that would provide all posts with a framework for maintaining easily accessible records on sites and volunteers. This effort--to be completed by June 2003--should facilitate post efforts to create effective safety and security information management systems. [23] Deterrence measures are likely to be used by military-type organizations or those working in areas of significant armed conflict. [24] The Peace Corps’ epidemiologic surveillance system also requests information on the four categories of assault originally listed in the assault notification surveillance system. Since a more elaborate system has been created to gather detailed information on such crimes, assault information provided through the epidemiologic system is used primarily as a check on the accuracy of reporting through the assault notification system. [25] As previously noted, Peace Corps staff observed that since many factors affect the incidence of crime against volunteers, it is difficult to attribute such declines to particular causes. GAO’s Mission: Obtaining Copies of GAO Reports and Testimony: 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 HYPERLINK http://www.gao.gov \* MERGEFORMAT www.gao.gov and select “Subscribe to daily E-mail alert for newly released products” under the GAO Reports heading. Order by Mail or Phone: 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 To Report Fraud, Waste, and Abuse in Federal Programs: Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov Automated answering system: (800) 424-5454 or (202) 512-7470 Public Affairs: Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800: U.S. General Accounting Office, 441 G Street NW, Room 7149: Washington, D.C. 20548.