This is the accessible text file for GAO report number GAO-05-38 entitled 'Federal Thrift Savings Plan: Customer Service Practices Adopted by Private Sector Plan Managers Should Be Considered' which was released on February 18, 2005. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to the Chairman, Committee on Homeland Security and Governmental Affairs, U.S. Senate: United States Government Accountability Office: GAO: January 2005: Federal Thrift Savings Plan: Customer Service Practices Adopted by Private Sector Plan Managers Should Be Considered: GAO-05-38: GAO Highlights: Highlights of GAO-05-38, a report to the Chairman, Committee on Homeland Security and Governmental Affairs, U.S. Senate Why GAO Did This Study: Intended to resemble private sector 401(k) pension plans, the federal government’s Thrift Savings Plan (TSP) held more than $128 billion in retirement assets for over 3 million participants at the end of 2003. Customer service-related difficulties during the Federal Retirement Thrift Investment Board ‘s (TSP’s governing body) record-keeping system conversion in 2003 led the Chairman of a Senate Committee to ask GAO to examine the customer service provided to TSP participants. This review describes (1) customer service provisions within TSP and those offered by private sector managers and (2) customer service practices used by private sector plan managers that could be considered for use in TSP. What GAO Found: TSP managers and private managers (servicing multiple pension plans) enable participants to select their preferred means of customer service from a similar range of options--such as telephone, Web sites, and on- site representatives--but each emphasizes different approaches. Both TSP and private plan managers provide customer service through automated telephone assistance as well as live representatives located at call centers. Both TSP and private managers also use standards to measure the efficiency and effectiveness of their call centers. However, TSP managers emphasize the efficiency of call centers based on quantifiable standards, such as the time it takes to respond to incoming calls, while private plan managers place a greater emphasis on the policy of satisfying each customer’s needs in one call. Both TSP and private sector plan managers also use Web sites to deliver plan information and allow participants to conduct personal transactions, and private plan managers emphasize the use of their Web sites as the primary vehicles for delivering retirement education and information to participants. Finally, while TSP managers said that agency representatives serve as the initial contact points for TSP employees to learn about TSP and receive counseling, private plan managers use on- site representatives less to supplement services provided by call center representatives and Web-based resources. Private sector plan managers we contacted have adopted various other practices that are not featured within TSP, such as regularly assessing customer satisfaction and using regularly updated technology to improve customer service. These managers gather participant feedback on their voice response system via short, automated surveys at the end of participants’ calls and use short, on-the-spot surveys to gather information on participants’ experience with their Web site. These plan managers emphasized the importance of incorporating participant feedback into their customer service delivery model in order to better meet the needs of their participants. Although TSP managers have surveyed participants in the past, they do not have a systematic approach to assess whether their customer service meets participants’ needs. TSP managers rely largely on indirect feedback from customer service staff, agency coordinators, and others who respond to complaints or requests for assistance from participants. The privately managed plans we studied also appear to utilize more up-to-date technologies to provide customer service, such as allowing participants to create account statements for any period of time or offering seminars over the Web on different plan topics that participants can access anytime. The TSP Web site provides fewer options and relies more on basic features. What GAO Recommends: GAO recommended that the Board (1) develop a systematic effort to assess TSP participants’ overall satisfaction with the services provided and (2) institutionalize the routine collection of information and systematic assessment of industry trends and innovations. The Board disagreed with our recommendations because it had stated its intentions to conduct a participant survey and has visited several large private plan managers. However, we continue to believe that ongoing efforts to assess customer satisfaction and practices used in the customer service industry should be institutionalized as a regular aspect of TSP’s operations. www.gao.gov/cgi-bin/getrpt?GAO-05-38. To view the full product, including the scope and methodology, click on the link above. For more information, contact Barbara Bovbjerg at (202) 512-7215 or bovbjergb@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: TSP and Private Sector Plan Managers Provide a Similar Range of Customer Service Options but Emphasize Different Approaches: Private Sector Plan Managers Have Adopted Various Practices That TSP Managers Could Consider to Improve Customer Service: Conclusions: Recommendations for Executive Action: Agency Comments and Our Evaluation: Appendix I: Scope and Methodology: Appendix II: Comments from the Federal Retirement Thrift Investment Board: Figure: Figure 1: Plan Service Provision Options: Abbreviations: ETAC: Employee Thrift Advisory Council: FERSA: Federal Employees' Retirement System Act: NFC: National Finance Center: OPM: Office of Personnel Management: PSR: Participant Service Representative: TSP: Thrift Savings Plan: United States Government Accountability Office: Washington, DC 20548: January 18, 2005: The Honorable Susan M. Collins: Chairman: Committee on Homeland Security and Governmental Affairs: U. S. Senate: Dear Chairman Collins: Since its inception in 1986, the federal government's Thrift Savings Plan (TSP) has become a central component of federal employees' retirement savings. Intended to resemble private sector 401(k) pension plans, TSP held more than $128 billion in retirement assets for about 3.2 million participants and has become one of the largest retirement plans in the United States.[Footnote 1] TSP resembles private sector 401(k) pension plans in that both allow employees to contribute a portion of their current compensation through payroll salary deductions and invest their account balances among a menu of investment options selected by the employer.[Footnote 2] Under pressure to provide plan features and customer service that are more consistent with those of large private sector plan managers,[Footnote 3] in 2003 TSP managers launched a new record-keeping system designed to allow improved service to plan participants.[Footnote 4] After experiencing problems with system access, customer assistance, and transaction processing during implementation of this new system, TSP managers have reported that the plan now processes more transactions more efficiently and with more participant control than in the past.[Footnote 5] However, the difficulties of implementing the new record-keeping system and the problems that ensued led plan participants and members of Congress to raise questions about the overall level of customer service at TSP. In light of these questions, the Chairman of the Senate Committee on Homeland Security and Governmental Affairs asked GAO to examine the customer service provided to TSP participants. This review describes (1) customer service provisions within TSP and those offered by private sector managers and (2) customer service practices used by private sector plan managers that could be used to improve TSP's service. To describe customer service in TSP, we interviewed officials from TSP and the National Finance Center (NFC), reviewed documentation governing TSP's customer service operations, and observed TSP operations. [Footnote 6] To describe how private sector plan managers provide customer service and identify practices that might prove useful to TSP, we interviewed five managers of plans representing an average of 4.9 million participants and an average of $169.8 billion in plan assets,[Footnote 7] observed the managers' customer service operations, and obtained documentation of their operations where possible.[Footnote 8] We conducted our work between December 2003 and December 2004 in accordance with generally accepted government auditing standards. Appendix I describes the scope and methodology of our work in greater detail. Results in Brief: TSP and private sector plan managers enable participants to select their preferred means of customer service from a similar range of options--such as telephone, Web sites, and on-site representatives--but each emphasizes different approaches. Both TSP and private plan managers use the telephone to provide customer service through automated telephone assistance as well as live representatives located at call centers. TSP and private plan managers also both use standards to measure the efficiency and effectiveness of their call centers. However, TSP managers emphasize the efficiency of call centers based on quantifiable standards, such as the time it takes to respond to incoming calls, while private plan managers place a greater emphasis on the policy of satisfying each customer's needs in one call. Both TSP and private sector plan managers also use Web sites to deliver plan information and allow participants to conduct personal transactions, and private plan managers emphasize the use of their Web sites as the primary vehicles for delivering retirement education and information to participants.[Footnote 9] Finally, while both TSP and private sector plan managers use on-site coordinators to provide plan information, this customer service function is more heavily emphasized in TSP. TSP managers said that agency representatives serve as the initial contact points for actively employed TSP participants to learn about TSP and receive counseling. Private sector plan managers use on-site representatives less and do so to supplement services provided by call center representatives and Web-based resources. Private sector plan managers we contacted have adopted various other practices that are not featured within TSP, such as regularly assessing customer satisfaction and using regularly updated technology to improve customer service. These managers gather participant feedback on their voice response systems via short, automated surveys at the end of participants' calls and provide links on their Web sites to gather information on participants' experience with their Web site. These plan managers emphasized the importance of incorporating participant feedback into their customer service delivery models in order to better meet the needs of their participants. Although TSP managers have surveyed participants, they have not done so since the early 1990s and do not have a systematic approach to assess whether their customer service meets participants' needs. TSP managers rely largely on indirect feedback from NFC and TSP staff and others who respond to complaints or requests for assistance from participants, such as agency coordinators. The privately managed plans we studied also appear to utilize more up-to-date technologies to provide customer service, such as allowing participants to create account statements for any period of time or offering seminars over the Web on different plan topics that participants can access anytime. The TSP Web site provides fewer options and relies more on basic features. This report contains recommendations to the Federal Retirement Thrift Investment Board (the Board) concerning actions needed to identify potential areas of improvement in the customer service provided to TSP participants by developing and implementing an evaluation effort to assess the level of customer satisfaction and routinely surveying the largest private sector plan managers to keep up with current industry trends. We recommended that the Board (1) develop a systematic evaluation effort, including broad survey methods that are supplemented by other efforts to randomly survey TSP participants, to assess their overall level of satisfaction with the services provided and (2) institutionalize the routine collection of information to regularly and systematically assess trends and innovations in the defined contribution industry. The Board disagreed with our first recommendation because it had previously announced its intention to survey TSP participants on a variety of topics, including client satisfaction. However, in its February 2004 monthly meeting, the Board said that such a survey would not be conducted for at least another 2 years. As we state in this report, the private sector plan managers that we spoke with believe that direct, ongoing participant feedback is needed to respond to the changing needs of plan participants. Without obtaining more frequent feedback from participants, TSP managers cannot determine what improvements would best satisfy participants' needs. Therefore, we continue to believe that the Board needs to develop and implement an ongoing evaluation effort that includes different types of surveys to systematically assess the level of customer satisfaction. The Board also disagreed with our second recommendation because it has conducted and will continue to conduct on-site reviews of the largest private plan managers to determine the services they provide and the technological capabilities they plan for the future. While the Board's recent visits to such plan managers are a positive step, we continue to believe that a routine and systematic effort to collect information on the practices used in the private sector customer service industry and assess their potential for use in TSP should be institutionalized as a regular aspect of TSP's operations. Background: The Federal Employees' Retirement System Act of 1986 (FERSA)[Footnote 10] created the Thrift Savings Plan as one of the basic elements of a new retirement system for federal workers to, among other purposes, provide options for retirement planning and encourage personal retirement savings among the federal workforce.[Footnote 11] Most federal workers are allowed to participate in TSP, which is available to federal and postal employees, members of Congress and congressional employees, members of the uniformed services, and members of the judicial branch. Eligibility depends upon coverage under the Federal Employees' Retirement System or the Civil Service Retirement System and civilian or military status.[Footnote 12] Eligible federal employees are able to contribute up to a fixed percentage of their annual base pay or a flat amount subject to IRS limits. Additionally, certain participants are eligible for automatic 1 percent contributions and limited matching contributions from the employing federal agency. TSP provides federal (and in certain cases, state) income tax deferral on employee contributions and their related earnings, similar to those offered by many private sector 401(k)-type pension plans.[Footnote 13] As with a 401(k) plan, participants are able to contribute a portion of their basic salary into an individual tax-deferred account. Participants have the ability to manage their accounts and conduct a variety of transactions similar to those available to 401(k) participants, such as reallocating contributions, borrowing from the account, making withdrawals, or purchasing annuities. Administration of TSP falls under the purview of the Federal Retirement Thrift Investment Board, an independent agency in the executive branch established by Congress under FERSA. This five-member, presidentially appointed Board's primary responsibilities include establishing policies for the investment and management of TSP. In addition, the Board must manage the Thrift Savings Fund solely in the interest of the participants and beneficiaries of TSP, as well as create administrative policies for the TSP. In addition to assigning these broad duties, FERSA also charges the Board with appointing an Executive Director and an Employee Thrift Advisory Council (ETAC).[Footnote 14] The Executive Director and staff are responsible for implementing the Board's policies and managing the day-to-day operations of TSP, prescribing regulations to administer FERSA, and other duties. The Employee Thrift Advisory Council advises the Board and Executive Director on matters relating to the investment and administration of TSP.[Footnote 15] While the Executive Director has responsibility for establishing and maintaining TSP participant accounts, these account record-keeping services are currently performed primarily by the U.S. Department of Agriculture's National Finance Center under a contractual agreement with the Board. Through this agreement, the TSP service office of NFC is responsible for updating participants' accounts based on data provided by agency payroll offices, processing participant account transactions, and providing customer service support related to these functions. Federal agency payroll and personnel offices also play a role in the administration and customer service activities of TSP. These offices are responsible for helping new employees enroll in TSP when they are hired and assisting existing employees make changes during designated semiannual seasons called open enrollment periods. They also retain control of administrative functions affecting employee payroll deductions, such as the election and alteration of contribution percentages. Additionally, FERSA requires TSP to provide information to employees participating in TSP and requires the Office of Personnel Management to establish a training program for all retirement counselors of federal agencies.[Footnote 16] TSP and Private Sector Plan Managers Provide a Similar Range of Customer Service Options but Emphasize Different Approaches: Though they emphasize different approaches, TSP and private sector plan managers enable customers to select their preferred means of service from a similar range of service options--including telephone, Internet, and on-site assistance (fig. 1). Both TSP and private plan managers give customers the option of obtaining assistance through automated telephone systems as well as live representatives located at call centers, although they emphasize different standards when evaluating the assistance provided. In addition, both TSP and private managers provide Web sites that deliver plan information and allow participants to conduct personal transactions, but private plan managers emphasize the use of their Web sites as the primary vehicles for delivering retirement education and information to participants. Finally, while both TSP and private managers use on-site coordinators to provide plan information, this customer service function is more heavily used by TSP. Whereas TSP managers said that agency representatives serve as the initial contact points for actively employed TSP participants to learn about TSP, private managers use on-site representatives less and do so to supplement call center representatives and Web-based resources. Figure 1: Plan Service Provision Options: [See PDF for image] [A] Only actively employed Thrift Savings Plan participants have access to on-site agency representatives. Also, not all private plan managers offer on-site representatives. [End of figure] Call Centers Are Used to Provide Customer Service by Both TSP and Private Sector Plan Managers: TSP managers provide an automated, toll-free telephone system and call center staff that help answer participants' questions, and TSP managers measure the efficiency of the call centers based on quantifiable standards, such as the time it takes to respond to incoming calls. TSP's automated telephone system, known as the ThriftLine and operated by the National Finance Center, allows participants to access general plan information, including share prices, rates of return, current loan interest rates, current annuity interest rates, and plan news.[Footnote 17] In addition to obtaining general plan information, participants can gain access to the ThriftLine system to get personal account information, such as their account balance and how they are allocating their contributions, or to manage their account, such as by making interfund transfers or withdrawals. Participants can also use the ThriftLine system to reach a TSP call center representative. TSP managers also maintain a call center where participants can reach service representatives; these representatives answer participants' questions and can also help them make changes to their accounts. About 155 call center representatives field incoming telephone calls and answer participants' questions about loans, allocation changes, interfund transfers, and withdrawals.[Footnote 18] In addition to answering participants' questions, these call center staff also register requests for assistance and complaints in a record-keeping system, which allows supervisors or lead analysts to review such comments and respond to participants to help them resolve problems. Call center staff said participants most commonly call to ask questions about loans or withdrawals. The staff also indicated that in the past participants with loans were more likely than others to call back seeking more help. Recently, the Board has taken a number of steps to improve telephone service to TSP participants. These included establishing an additional call center to reduce service interruptions and providing toll-free telephone service to facilitate participant access to service representatives. The new call center, located in Cumberland, Maryland, opened in July 2004 to complement the center located in New Orleans, Louisiana, during normal operations and provide backup during weather- related or other local events that could otherwise interrupt service. Also in July 2004, the Board began providing toll-free telephone service to TSP participants where participants will be able to obtain TSP account or transaction information via the automated telephone service 24 hours a day, 7 days a week. TSP managers emphasized the efficiency of call centers based on quantifiable standards--such as the percentage of calls that are answered within a specified time frame--and pointed to their performance on these measures as evidence of call center productivity. For example, the TSP call center tries to answer at least 90 percent of calls within 20 seconds of a participant calling. TSP was meeting this standard in 2003 prior to the implementation of the new record-keeping system and has since begun meeting it again.[Footnote 19] However, TSP failed to meet this standard for about a 10-month period (May of 2003 through February 2004), when it was experiencing an increase in the volume of calls received prompted by the implementation of the new record-keeping system. Finally, TSP also closely monitors the percentage of callers who hang up before receiving service (known as the call abandonment rate) and the average call length as a measure of service, and each call center representative is required to help an average of at least 50 participants per day.[Footnote 20] TSP managers said that resolving the participant's concern on the first call is also a goal, and they have enlisted the help of a private sector contractor to make changes to improve the quality of their call center processes and are developing new standards for call center representatives. Private sector plan managers also provide caller assistance through automated toll-free telephone systems and live service representatives. Some of these managers have more than one defined contribution plan under their management and provide unique toll-free numbers for each plan. Participants complete a log-in process that routes them to an automated menu unique to their particular plan--a system similar to that provided by the ThriftLine. From this menu, participants can then check their account balances, get price quotes, receive plan information, and process account transactions. These automated customer service lines also give participants the option of speaking to a live call center representative who can provide assistance with transactions or answer their questions. In addition to providing basic assistance, service representatives use the call as an opportunity to direct participants to educational and plan materials available on their Web site. While private managers used similar means as TSP to provide caller assistance, they emphasized different types of standards than TSP when evaluating the success of their call centers. Private managers told us that they downplayed the importance of quantitative measures and instead focused on ensuring that service representatives fully satisfied each customer's needs efficiently and politely in one call. Specifically, private sector managers emphasized this policy of first- call resolution as an overriding goal and said they viewed quantitative measures, such as those emphasized by TSP, as secondary or, in some cases, an inaccurate measure of call center performance. For example, while TSP keeps close watch on the call abandonment rate, private managers we spoke with placed little emphasis on this measure. In fact, one plan manager told us he believed the call abandonment rate to be an inaccurate measure of productivity, as some participants terminate calls after opting to visit the Web site in response to the telephone recording they hear while on hold. Furthermore, unlike TSP managers, who evaluate service representative utilization by representatives' ability to answer a minimum number of calls each shift, private managers we spoke with did not have required call minimums, instead stressing that service representatives should fully resolve each participant's needs, regardless of the time required. Finally, although private managers had requirements--such as answering 90 percent of calls within 20 seconds--specified in the contracts with their clients, they did not use such measures as the primary standard for evaluating service representatives or call center effectiveness.However, we did not attempt to assess the possible impact of these differences on the quality of customer service. Web Sites Are Provided by Both TSP and Private Plans, but Private Plan Managers Tend to Use Them for Broader Purposes: TSP managers also provide customer service through their Web-based transaction system. TSP's Web site is maintained and operated by TSP staff and according to TSP managers has received an average of about 15.7 million hits per month and processed an average of 216,696 transactions per month during the first 8 months of 2004. Participants can use the Web site for functions such as accessing plan information, making loan requests, downloading forms, transferring funds, changing allocations between funds, and accessing account balances, among other things. Participants can also conduct many of these transactions over the phone, but by using the Web site, participants can also make a loan request and can submit the form online for quicker processing. According to TSP managers, the Web site has to a great extent replaced paper forms for processing transactions and providing information. For example, 93 percent of interfund transfers were conducted through the Web site during the first 8 months of 2004. TSP managers also said that they have eliminated paper versions of forms, statements, and bulletins that have traditionally been provided through the mail or sent to agency coordinators because most of these documents can be obtained through the Web site. TSP managers stated that their goal in doing so is to improve their level of service while decreasing costs.[Footnote 21] Private sector plan managers that we spoke with said that in addition to providing a means for participants to process transactions, they also intend their Web sites to be the primary vehicles for delivering retirement education and information to their participants. Like TSP managers, private managers said they allow participants to conduct different transactions, check account balances, and receive plan information electronically. One plan manager that we spoke with said that his Web site processed approximately 9 million transactions per month in 2003. In addition, we found that private sector Web sites provided a range of resources for participants, from Web-broadcast seminars to downloadable brochures and general investment information, and most private plan managers underscored the importance of the Web site as a tool that allows participants to access information on their own. One manager told us that his goal is not only to educate participants but also to influence participant behavior by strategically placing educational information on the Web site so as to link such educational materials to the topic or transaction a participant is pursuing. For example, the plan manager said that research has found that most participants visited the site for information on their account balance or personal rate of return, not to seek out educational materials. In response to findings about participant Web usage, the plan manager redesigned the site so that educational materials are embedded in relevant pages of the site, rather than isolated on a separate page. In contrast, TSP separates all plan and retirement information from account and transaction access. Agency Coordinators Serve as the Front Line of Customer Service Operations for TSP, but Managers of Similar Private Sector Plans Rely on Such On-site Representatives Less Frequently: Designated by their federal agencies as liaisons with TSP managers, agency coordinators provide customer service to participants within their particular agencies. Agency coordinators, who are typically located in their agencies' personnel or benefits offices, are the primary contacts for actively employed TSP participants. Generally, agency coordinators are expected to inform eligible employees of TSP options and benefits; maintain supplies of TSP forms and informational materials; collect, process, and submit TSP election forms to agency payroll offices; and respond to inquiries from active employees.[Footnote 22] For some agencies, the coordinator's responsibilities are combined with those of the agency retirement counselor.[Footnote 23] FERSA authorizes agencies to designate retirement counselors who are responsible for providing employees with benefits information and mandates that OPM establish a training program for these retirement counselors.[Footnote 24] Agencies, however, have primary responsibility for designing and implementing their programs according to agency-specific needs. TSP managers said that the coordinators are the front line of TSP's customer service operations. However, they also said that they lack the authority to monitor how frequently, at what point, and for what purpose coordinators contact participants. TSP managers view their role in providing customer service through agency coordinators as limited primarily to assisting these coordinators by distributing information and answering questions as they perform their TSP-related duties. The agency coordinators we spoke with said they were responsible for providing information on TSP to new employees, providing employees information on open season enrollment, conducting retirement seminars, helping with payroll issues, and answering any questions that employees may have about TSP. We found that the information and assistance provided to employees by coordinators in different agencies varied. For example, when we spoke with nine agency coordinators from large and small agencies, three coordinators told us that they had requested TSP staff to conduct seminars at their agencies to provide information on TSP and retirement planning. Of the six other coordinators that we spoke with, all but one said they conducted their own seminars or used an outside contractor. The agency coordinators said they also distribute information electronically or through posted bulletins, informational meetings, orientation sessions, and fliers in office mail. Several coordinators indicated that they refer participants to the TSP Web site or the ThriftLine to obtain plan information or to conduct transactions. The coordinators themselves receive guidance and new information on TSP from TSP managers by attending optional quarterly meetings and coordinator training sessions that are offered monthly and reading TSP bulletins--monthly publications that provide recent information on the TSP and are available to agency coordinators through the TSP Web site and the mail. The agency coordinators we spoke with indicated that they had other responsibilities in addition to their TSP-related duties. Private sector plan managers rely on on-site representatives less and tend to deliver educational and transaction resources directly to participants. As a result, plan managers view call center representatives, automated telephone services, and Web sites as their front line of customer service operations. This emphasis on self- service results in these managers providing all available plan information and retirement education through one or more of these service delivery methods and using on-site representatives only as a backup measure or not at all. Private sector companies have been moving away from face-to-face service delivery approaches and toward live telephone assistance to meet consumers' expectations for fast and convenient service and to provide this service more cost-effectively. Additionally, by allowing participants to conduct all transactions-- including making changes to contribution percentages--through the Web site or telephone services, plan managers also rely less on employers to provide services. One plan manager we spoke with described the use of on-site representatives as flawed, because information provided by different on-site representatives may vary. By providing all information through a Web site or centralized contact center, the manager told us, the plan could ensure consistency of service. Despite this, some plan managers we spoke with did use on-site representatives in a limited capacity. One plan manager provided limited on-site support for its state government and school district clients, but the arrangement was specific to the plan sponsor and the plan manager did not have a companywide practice for utilizing such representatives. Specifically, the number, location, and function of representatives provided to a client varied from plan sponsor to plan sponsor. For its state government and school district clients, the plan manager only provided on-site service when the plan sponsor specifically requested a representative or during specified times, such as open enrollment periods. Private Sector Plan Managers Have Adopted Various Practices That TSP Managers Could Consider to Improve Customer Service: Private sector plan managers we contacted have adopted various other practices that are not featured within TSP, such as regularly assessing customer satisfaction and using regularly updated technology to improve customer service. Private managers frequently assess their performance by gathering participant feedback on the services provided and use this information to improve their customer service delivery. Although TSP managers have surveyed participants, they have not done so since the early 1990s and currently have no systematic approach to assess whether the plans customer service meets participants' needs. In addition, private managers that we spoke with appear to utilize more up-to-date technologies to provide customer service than do TSP managers. Private Sector Plan Managers Frequently Assess Their Performance by Gathering Participant Feedback on the Services Provided: Private sector plan managers regularly gather participant feedback and use the feedback received to improve their customer service. Since most private plan managers deliver customer service through multiple methods, they use a number of mechanisms to gather participant feedback. These mechanisms are customized to gather information specific to the service method the participant used. For example, some plan managers we spoke with provide a link on their Web site to a survey that allows participants to share their impressions of and experiences with the service provided through the site. Some managers also use short on-the-spot surveys to gather feedback about a participant's experience on the plan manager's Web site.[Footnote 25] These plan managers also survey participants after they have interacted with the plan's contact center and voice response telephone system through short, automated surveys at the end of a call. Managing officials told us that these surveys include questions regarding the service method's ease of use, whether or not the participant was able to complete the intended transaction, and sometimes ask whether the participant has suggestions as to how the service method or transaction process could be improved. Traditional survey methods are also used, such as an annual mail or telephone survey to randomly sample participants and assess their overall level of satisfaction with the services provided. However, plan managers said that these surveys take longer and are more costly to administer than other methods. All of the plan managers we spoke with emphasized the importance of incorporating participant feedback into their customer service delivery models in order to better meet the needs of their participants. They told us that it is important to determine what is working and what is not by regularly gathering participant feedback. For example, one plan manager told us that his plan had been mailing participant statements to the employer for distribution for several years. He thought that this method of distribution would be more convenient and cost-effective for the plan and that since no participants were complaining, it must be working efficiently. However, when participants were surveyed, they made it known that the statements were usually not distributed by their employer in a timely manner, if at all, and as a result, statements are now mailed directly to participants. Another plan manager said that his plan had redesigned its entire Web site and added features based primarily on feedback from their participants. TSP managers said that they have not surveyed participants since the early 1990s because they were waiting for a new record-keeping system to be developed, and they said it would not have made sense to survey participants until after the transition issues were resolved and participants had some experience under the new system. Rather than survey participants shortly after the new system was implemented, in 2003, TSP managers have since decided to wait until a new investment option is added to the existing selection of five funds. The Board anticipates implementing the new option in mid-2005. Although the Board has a system for addressing participant complaints, TSP has no systematic mechanism in place for soliciting participant views about the quality of the service they are receiving. Instead, TSP managers rely largely on indirect feedback from NFC and TSP staff and others, such as agency coordinators, who respond to complaints or requests for assistance from participants. Agency coordinators and the call center representatives at NFC provide some feedback to TSP managers regarding areas of potential improvement. TSP managers are also responsible for gathering participant feedback through the Employee Thrift Advisory Council, as required by FERSA. However, while some ETAC representatives provide TSP managers with feedback on draft TSP publications, legislative initiatives, and other issues, ETAC representatives do not systemically solicit feedback from their constituents. Some ETAC representatives may receive sporadic feedback from participants, but ETAC does not conduct surveys of plan participants. As a result, TSP managers are dependent on call center representatives or agency coordinators to forward any feedback they receive from participants. Also, the executive director of ETAC said that TSP participants might be more likely to raise customer service issues with their local representatives, such as union representatives, rather than elevate issues to the national level. Therefore, the extent to which participants within the represented agencies and employee organizations provide feedback to their ETAC representative is unclear. Because TSP relies on customer complaints as an indicator of participant satisfaction, its managers do not have the information necessary to determine the degree to which participants are satisfied with the services. A TSP official said that because participant complaints have decreased significantly and leveled off since the record-keeping system conversion in June 2003, participants are probably satisfied with the services that TSP is providing. However, TSP managers' reliance on complaints does not take into account participants who are dissatisfied and have not complained or do not know where to complain about the services they received. In other instances, participants have to send letters to TSP managers or the TSP call center, or contact their member of Congress with problems or concerns they have had with the services they received. Privately Managed Plans Use More Up-to-date Technology in Delivering Customer Service than TSP Does: Private sector plan managers that we spoke with also utilize more up- to-date technologies to improve customer service than TSP does. We found that the privately managed plans' Web sites provided participants more flexibility and options for managing and learning about their retirement accounts than did the TSP Web site. Private plan managers also told us that one part of providing the best possible customer service is allowing participants to serve themselves in an easy and convenient manner, and they have found that the more up-to-date technology that they have incorporated into their customer service delivery models helps facilitate participant self-service. For example, one plan manager's Web site allowed participants to instantly create and print account statements for any period of time. This allows participants to easily determine how much they earned, lost, or contributed for any given period of time. Plan managers said that such information helps participants make decisions about how to allocate their funds and about retirement planning. Conversely, the TSP Web site only makes statements available for calendar year quarters, and the participant must wait approximately 2 weeks after the close of the quarter to obtain the statement. Privately managed plans also use their Web sites to help participants understand their retirement plan and the options available to them within the plan. For example, one plan manager we spoke with provided access to prerecorded seminars on the Web sites that cover current account-related topics and features. Participants may view these seminars at any time as an alternative to the printed literature that was also available through the Web site if desired. Conversely, the TSP Web site only offers print copies of its general plan information brochure, available in hard copy or for viewing online, and the brochure on general plan information has not been updated since mid- 2001. The private plan managers that we spoke with also utilized more sophisticated calculators on their Web sites than does TSP. For example, one plan manager's Web site provided a calculator that would project a participant's account balance given certain hypothetical parameters, such as number of years before retirement, that the participant can select. The calculator is securely linked to the participant's account, and certain items, such as the participant's account balance and historical rate of return on assets, are automatically entered into the calculator, simplifying the process for the participant. TSP also offers an account projection calculator, but the calculators they offer are not on a secure server and all parameters must be specified manually, and the participant may not know what assumptions, like the estimated rate of return, are reasonable to specify. The plan managers that we spoke with told us that participants are more likely to use and benefit from tools such as the retirement income projector if they are convenient and easy to use. All of the private plan managers that we spoke with emphasized the importance of keeping abreast of the latest technology and industry trends in order to provide participants with the highest possible level of customer service. For example, two plan managers told us that they regularly and systematically review not only their competitors' Web sites for new ideas and innovations, but they also study the Web sites of other companies, such as Yahoo and Amazon, to keep abreast of the latest technological developments. Because TSP managers are not taking full advantage of such technology, TSP participants may not receive the benefits it offers. As plan managers have continually updated their Web sites to incorporate more sophisticated features, they have simultaneously experienced increased Web usage and decreased calls to service representatives by participants. Another manager stated that certain Web features could reduce mailing costs by allowing plan managers to provide some information electronically. Although TSP managers have a simple and functional Web site available for participant use, the site does not offer the flexibility and convenience of the Web sites provided by the private plan managers that we reviewed. Although TSP managers told us that they have recently taken steps to learn about industry innovations--including talking to and occasionally visiting private sector plan managers, viewing private sector Web sites, and reviewing the literature from researchers in the field, they have not yet institutionalized the routine collection of information to regularly and systematically assess trends and innovations in the defined contribution industry.[Footnote 26] Conclusions: TSP has become one of the largest retirement savings plans in the United States, and it must provide customer support and record keeping for millions of participants across many federal agencies throughout the United States and, in some instances, the world. TSP must exchange information with multiple federal agency payroll, personnel, and data- processing representatives and handle millions of participant transactions every month. In addition, TSP relies on other federal agencies to provide participants' benefit information, including educational and other information, on TSP. Because these characteristics make TSP unique, its approach to customer service is naturally somewhat different from that of other pension plans. Nonetheless, some aspects of private sector practices seem appropriate for TSP managers to consider. For example, as TSP managers continue to modernize their customer service operations, they should look for opportunities to use participant feedback to identify broader innovations to improve their services, in addition to focusing on individual participants' specific concerns. The practices of private sector plan managers suggest that direct, ongoing participant feedback is invaluable in responding to the changing needs of plan participants. Without obtaining more frequent and representative feedback from participants, TSP managers cannot determine what improvements would best satisfy participants' needs. While recognizing the recent improvements made to TSP's customer service--such as the new record-keeping system, the additional call center, and toll-free telephone service--we conclude that TSP could benefit from additional measures that could enhance customer service further. For example, the innovative, technological, and operational practices that have become commonplace in the private sector customer service industry might be means to continual customer service improvements at TSP. Also, TSP could benefit from regular and sustained examination of private sector customer service practices. As TSP participants become more familiar with online financial transactions and the services provided by private plan managers, participants may come to expect services not currently provided by TSP. However, the costs of making changes to TSP's services would have to be balanced against the potential benefits such changes could provide to participants and the plans themselves. Recommendations for Executive Action: To help ensure that federal workers have the options needed to effectively plan and to encourage them to save for retirement, TSP managers should continually seek new ways to improve their customer service operations. Both the potential costs and benefits should be weighed in making decisions about changes to service. Therefore, we recommend that the Federal Retirement Thrift Investment Board direct the Executive Director to take the following actions: * Develop and implement an evaluation effort to systematically assess the level of customer satisfaction and to identify, as needed, areas of potential improvement for the ThriftLine, caller assistance center, Web- based transaction system, and TSP coordinator program. It should consider a variety of different approaches, including traditional methods such as surveying participants annually through the mail or telephone to assess their overall level of satisfaction with the services provided, supplemented by other approaches, such as exploring the use of Web-based and automated telephone call evaluation tools to randomly survey TSP participants. * Institutionalize the routine collection of information from the largest private sector plan managers to keep up with current industry trends and assess whether the new and existing practices used by private managers would prove advantageous and cost-effective to TSP. Agency Comments and Our Evaluation: We obtained written comments on a draft of this report from the Federal Retirement Thrift Investment Board and made changes to the report as appropriate. The full text of these comments is reproduced in appendix II. In response to the Board's comments, we added information regarding its plans for surveying participants. We also added information regarding the Board's recent announcements that it will review the service provided by private plan managers. The Department of Labor was also provided a draft of this report for review and advised us that it had no comments. In its written comments, the Board disagreed with our recommendation regarding the implementation of an evaluation effort to assess the level of customer satisfaction. The Board stated that our recommendation was moot because the Board had previously announced its intention to survey TSP participants on a variety of topics, including client satisfaction. However, in its February 2004 monthly meeting, the Board said that such a survey would not be conducted for at least another 2 years. As we state in our report, the private sector plan managers that we spoke with believe that direct, ongoing participant feedback is needed to respond to the changing needs of plan participants. Without obtaining more frequent feedback from participants, TSP managers cannot determine what improvements would best satisfy participants' needs. The Board also suggested that certain aspects of plan performance indicate increased participant satisfaction. In particular, the board stated that TSP's high participation rate, increased number of transactions, and low account withdrawal rate among inactive participants, such as retirees or former employees, imply a high level of participant satisfaction. Although such performance measures provide an indication of participant satisfaction, more sophisticated efforts, such as participant surveys, are needed to provide direct information on the changing needs of plan participants and what improvements would best satisfy those needs. We recognize that these efforts may require substantial resources, which is why other means, such as exploring the use of Web-based and automated telephone call evaluation tools to randomly survey TSP participants, are needed as supplementary efforts. Without a systematic effort to obtain feedback from participants, the Board is not in the best position to determine participant satisfaction. The Board also disagreed with our recommendation that it should routinely solicit information from the largest private plan managers to assess whether their practices would be advantageous and cost-effective if adopted by TSP. The Board said that it has conducted and will continue to conduct on-site reviews of the largest private plan managers to determine the services they provide and the technological capabilities they plan for the future. While the Board's recent visits to such plan managers are a positive step, we believe that a routine and systematic effort to survey practices used in the private sector customer service industry should be institutionalized as a regular aspect of TSP's operations. Further, in considering which new practices to adopt, the Board would need to weigh the costs of making changes to TSP's services against the potential benefits to participants and the plans themselves. Conducting such cost/ benefit analyses is a management function and would be inappropriate for GAO to initiate. The Board suggested that the section of our report on customer service options questions the federal government's reliance on agency retirement counselors for the delivery of TSP information and service. This section of the report was descriptive, illustrating the different ways in which private sector entities and TSP managers provide customer service. While we noted that some federal agencies use on-site coordinators, we observed that, in general, the private sector tended to rely less on coordinators and more on call center representatives, automated telephone services, and Web sites to deliver informational and transactional resources to the participants. The Board stated that it rejects the suggestion that the focus of retirement education be shifted away from retirement counselors. We are not suggesting that the agencies move away from retirement counselors, but we emphasize that providing retirement information through the use of retirement counselors does not preempt TSP managers from also providing such information through call centers and Web sites. The Board questioned the feasibility and value of implementing several customer service features used as examples in the draft report. For example, we included a discussion about a private sector Web site that could display a graph of the projected growth rate of a participant's account at the current contribution rate as well as at a slightly higher hypothetical rate to demonstrate the effect of higher contributions. This example was used only to illustrate the type of educational features used by the private sector and was not necessarily intended as a specific recommendation to be implemented. However, we maintain that the private sector offers a range of Web features, particularly those concerning education and retirement information, that are not available to TSP participants. Similarly, other examples, including on-the-spot participant surveys and more flexible options for statement preparation, were intended to illustrate the types of advanced tools private sector plan managers provide to their participants rather than as specific recommendations to be implemented. The Board also stated that the draft ignored TSP's superior performance results as measured by key standards that demonstrate that TSP significantly outperforms the private sector. In response, we noted that TSP had generally met its standard for call center performance measurement, such as the percentage of all calls answered within 20 seconds, except for a period of about 10 months of increased caller volume as the Board implemented its new record-keeping system. However, as presented in the draft, private managers told us that they downplayed the importance of quantitative measures and instead focused on ensuring that service representatives fully satisfied each customer's needs efficiently and politely in one call. Specifically, private sector managers emphasized this policy of first-call resolution as an overriding goal and viewed the quantitative measures, such as those emphasized by TSP, as secondary or, in some cases, an inaccurate measure of call center performance. Finally, the Board stated that the draft failed to address three of the five areas of interest to the committee. At the conclusion of the design phase of our study, GAO and the committee staff agreed to the objectives stated at the beginning of this report. We eliminated the other areas of interest largely because they overlapped with the scope of an ongoing audit by the Department of Labor of TSP's customer service. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days after its issue date. At that time, we will send copies of this report to the Secretary of Labor, the Federal Retirement Thrift Investment Board, appropriate congressional committees, and other interested parties. In addition, we will make copies of this report available at no charge on GAO's Web site at http://www.gao.gov/. If you have any questions concerning this report, please contact me at (202) 512-7215 or Tamara Cross at (202) 512-4890. Other major contributors include Daniel Alspaugh, Amy Buck, Richard Burkard, Erin Daugherty, Malcolm Drewery, Michael Morris, Corinna Nicolaou, and Roger Thomas. Sincerely yours, Signed by: Barbara D. Bovbjerg: Director, Education, Workforce, and Income Security Issues: [End of section] Appendix I: Scope and Methodology: To describe how Thrift Savings Plan (TSP) managers provide customer service, we interviewed TSP managers at the Federal Retirement Thrift Investment Board headquarters in Washington, D.C., and officials at the Department of Agriculture's National Finance Center (NFC), which has been the major contractor for account maintenance and participant support since TSP began operations in 1987. In addition, we reviewed documentation governing TSP's customer service operations, attended monthly Board meetings, and interviewed officials at the Department of Labor and the Employee Thrift Advisory Council (ETAC). Specifically, to describe the customer service operations of the ThriftLine and call center, we also observed center operations at the NFC in New Orleans, Louisiana, including customer service representatives taking telephone calls from TSP participants; we also collected and reviewed information on the number of calls made to the ThriftLine, the percentage of calls answered within specified time frames, and the number and duration of calls referred to a customer service representative. To describe the customer service operation of the TSP Web site, we also observed and documented the information and features available on the Web site through our personal computers at GAO and obtained information from TSP managers on the number of visits made and transactions conducted over the TSP Web site. We also interviewed TSP managers and NFC officials regarding the reliability of their call center or Web site data, such as the number of calls received or the number of interfund transfers transacted through the Web site. However, they were unable to provide information about any reliability tests that might have been performed. Call center and Web site data are presented here for illustrative purposes. Because these data are not central to the findings, conclusions, or recommendations of this report we did not assess their reliability. To describe the role of the agency coordinator, we also interviewed nine TSP coordinators, selected on the basis of a nonrandom sample of federal agencies with 5,000 or more employees (U.S. Postal Service, Department of Defense, Social Security Administration, and General Services Administration) and federal agencies with fewer than 5,000 employees (Government Accountability Office, Federal Trade Commission, Office of Personnel Management, Small Business Administration, and the Railroad Retirement Board). We also reviewed the Federal Employees' Retirement System Act and Department of Labor guidance on responsibilities of the federal agencies participating in TSP. To describe how private sector plan managers provide customer service through call center representatives, automated telephone systems, Web sites, and on-site representatives, we visited managers of other large defined contribution plans, conducted interviews, observed their customer service operations, and in some cases obtained documentation on their operations. Specifically regarding private call center representatives and automated telephone systems, we obtained information regarding their customer service philosophies, use of call center benchmarks, call center staff training and evaluation policies, call center technology, and call volume management. In some cases, we obtained call center representative training manuals and call center statistics. We also listened to live calls with call center representatives and observed the software tools used by these representatives to assist callers at some of the plan managers that we visited. We also observed their Web site pages and features, such as their retirement calculators and education materials and discussed how often they update their Web sites and how they determine what to include on their sites with management and information technology specialists. To identify customer service practices used by private sector plan managers that TSP managers could consider to improve its services, we also obtained copies of both online and printed customer service- related survey instruments and, in some cases, survey results, and discussed with plan managers how they used this information to determine what changes would be made to their customer service delivery. To select managers of similar defined contribution plans for comparison, we selected 4 of the top 10 private sector plan managers and 1 of the top 3 state and local government agency plans based on total plan assets under management as of December 31, 2002, as reported by Pensions & Investments and Plan Sponsor Magazine. The five plan managers that we spoke with had an average of $169.8 billion under management and serviced an average of 4.8 million participants. They ranged in size from $9.8 billion to $282.0 billion under management and from 150,000 to 10.3 million participants. We also spoke with private and public pension industry groups such as the Profit Sharing Council of America and the National Association of Government Defined Contribution Administrators about our selection methods. [End of section] Appendix II: Comments from the Federal Retirement Thrift Investment Board: FEDERAL RETIREMENT THRIFT INVESTMENT BOARD: 1250 H Street, NW: Washington, DC 20005: THRIFT SAVINGS PLAN: GARY A. AMELIO: Executive Director: November 8, 2004: Ms. Barbara D. Bovbjerg: Director of Education, Workforce, and Income Security Issues: U.S. Government Accountability Office: Washington, D.C. 20548: Dear Ms. Bovbjerg: This responds to your request for comments on the draft report now entitled "Federal Thrift Savings Plan: Practices Adopted by Private Sector Plan Managers Could Help Improve Customer Service." The draft report recommends that the Thrift Savings Plan (TSP): 1) regularly and systematically assess the level of participant satisfaction with the customer service provided, and: 2) routinely survey the largest private sector plan managers to assess whether the new and existing practices used by private managers are applicable and if they would prove advantageous to the TSP. With regard to the first recommendation, upon my appointment as Executive Director of the TSP in June 2003, I publicly announced my intention to survey TSP participants on a variety of topics, with particular emphasis on satisfaction. However, it was clear to me that participants should not be surveyed until they had acquired some experience with the new record keeping system which was just then being introduced. Although my public statements are not reflected in the draft report, they, were discussed and explained to your staff. You were also advised that the hiatus in surveying participants during the late 1990s was due to the pending implementation of the new record keeping system. Surveying participants on satisfaction with the legacy system would have been a worthless exercise and a waste of participant resources. This key information is also not reflected in the draft. In effect, the first recommendation mimics a decision I announced before the GAO began its work in November 2003, and is therefore moot. The draft does note the convenient multiple routes available to participants who wish to provide feedback until the survey instrument is instituted. A most extraordinary demonstration of participant feedback is the fact that more than 600,000 TSP participants who have stopped contributing can transfer their TSP account balances to private plans with the features you describe and suggest are desirable. Instead they keep their retirement funds on account in the TSP, many until the last possible moment before the Tax Code requires their withdrawal. Moreover, we have accepted 24,000 transfers totaling $700 million from other plans since July 2001 for participants who prefer the TSP to their previous retirement plans. Complaints in written correspondence and e-mail have diminished substantially as the new system and other improvements have been implemented. Nevertheless, prudent steps to survey for participant satisfaction levels will be taken as discussed. Regarding the second recommendation, I am intimately familiar with industry practices, given my twenty two years of experience with private sector retirement plans and investments before arriving at the TSP. I and senior staff at the Agency remain current through participation in industry-sponsored events. Further, I and the senior staff have and will continue to personally visit the operational centers of the largest and most respected institutions in the industry to conduct on-site reviews of the levels of service provided and the technological capabilities planned for the future. Indeed, I announced four of such visits in a public Board meeting two months ago. In fact, a member of your staff was present at the meeting. I consider this type of hands-on interface with the industry superior and far more useful than having the Agency survey for data that is often already available from industry sources. Thus, I view this recommendation as sub-optimal and also moot. The draft report questions the Federal government's focus on retirement counsellors at the various employing agencies as a key point for the delivery of TSP information and service. I was confused by the inclusion of this concern in the report because the counsellors exist as a matter of law and are a well-established, successful practice. I subsequently learned from Agency staff that the Income Security Issues group of the Health, Education and Human Services Division of the GAO previously questioned this arrangement when it recommended that the Federal Retirement Thrift Investment Board (Board) assume educational responsibilities for retirement matters beyond the scope of the TSP (GAO/HEHS-96-1). The Board strongly objected to that recommendation (as did the then- Chairman of the Senate Committee on Governmental Affairs) and it was never implemented. The General Government Division (GGD) of the GAO issued a subsequent report (GAO/GGD-99-27) validating both the Board's view and the cooperative work performed by OPM, the Board, and the counsellors at the employing agencies. That report was aptly entitled "Federal Retirement: Key Elements Are Included in Agencies' Education Programs." Agency staff had hoped this thoughtful, fact-based work by the GGD had put this issue to rest, and we shared it with the evaluators on the current project. However, the current draft returns to this design issue, again questioning its wisdom by asserting that private sector 401 (k) plans primarily (and increasingly) use their web sites for education, and suggesting a move away from the retirement counsellors. What is truly baffling about this recurrent questioning of how retirement information is provided government-wide is the fact that the key performance measurements of the success of 401(k) type plans --voluntary participation rates and the cost of running a plan --demonstrate that the TSP substantially outperforms the private sector. The draft ignores these superior performance results and suggests shifting the focal point of education from the statutorily established retirement counsellors --who are paid for by employing agencies, tailor their activities to agency and individual needs, and provide the full range of assistance on all retirement matters throughout the employees' career --to a phone-based approach that would comport with private sector practice. (Interestingly, while the draft states that "Private sector companies have been moving away from face-to-face service delivery approaches toward live telephone assistance to meet consumers expectations for fast and convenient service and to provide this service more cost-effectively," it does not address whether those costs are shifted from the employer to the plan participants in the process. That is what would happen if the TSP were to follow this model, and I expect that it also happened in the mail delivery example cited in the draft. Indeed, vendors are often happy to take on additional tasks which increase their profit, and employers are often happy to shift those costs to the plan, since plan costs are usually paid by participants.) Although we will continue to take appropriate advantage of evolving capabilities, we again reject the suggestion that we shift the focus away from retirement counsellors. This approach --which was established by the Congress, validated by the GGD report, and fully-integrated into the Federal establishment --has produced low costs for participants and superior results that may be worthy of emulation by the private sector (rather than the other way around). Indeed, I recommend that the unsupported bias toward private sector plans evident in the title of the draft be purged since it is not substantiated by performance measures or facts. An appropriate, unbiased, and fully descriptive title would be "Comparison of some customer service provisions within the TSP with those offered by some private sector managers." I recommend that the stray suggestions in the text not supported by a recommendation be excised as well. Simply listing well-known "customer service practices used by private sector managers that could be used to improve TSP's service" has no value when GAO is unwilling to make and support a recommendation that the specified practices "should" be used. I am disturbed that this Agency expended substantial time and energy to comply with your staff's multiple requests for information, yet the draft fails to address three of the five areas of interest cited by Chairman Susan Collins in her letter of November 3, 2003, (attached) i.e.: -describe the extent of the FRTIB oversight of customer service operations; -determine to what extent TSP tracks requests for assistance and complaints from TSP participants and actions taken to respond to such concerns; -discuss steps TSP has taken to improve customer service. Agency staff provided all of the information requested by the GAO in these areas. We also offered extensive data demonstrating thorough oversight, tracking, and responsiveness, as well as documenting the extraordinary improvements in customer service resulting from the implementation of the new TSP record keeping system just over a year ago. The latter has truly revolutionized TSP service with state-of-the- art functionality including web transactions, paperless statements, daily valuation, streaming (rather than monthly) disbursements and more, in a performance environment that meets or exceeds the best practices in the industry. Although the draft notes that the TSP meets the industry call center standard of 20 seconds/'90% of calls, much more relevant and illuminating information is unreported. Other service performance which meets or exceeds industry standards, and would apparently be responsive to Chairman Collins' request, are simply not discussed. For example, the largest day to day service demands of TSP participants are for interfund transfers, loans, and withdrawals. Our goal is that the vast majority of these participants receive outstanding service without needing to speak to call center representatives to obtain these transactions. The automated system we put in place last year allows each participant to manage TSP investments and execute transactions directly. This service is convenient for participants, administratively efficient, and cost effective. It is the most sweeping improvement in the TSP's history. Interfund transfers that used to take two to six weeks are now performed daily. Loans and withdrawals that used to take four to eight weeks to complete now take one to two. The volume of call center inquiries, which was increasing, has now declined. Participant letters of complaint, which were rising, are now heading down. Yet, the draft takes no notice of the dramatic service improvement or the obvious salutary effect it has had on participant satisfaction. The draft suggests that more participant funds be invested in technology. Indeed, in my meetings with industry leaders, one discussed his plan to spend $1.5 billion each year on technology, and another reported spending 25 percent of gross revenue on the same. While such an approach might be prudent for a company needing to offer "bells and whistles" to compete with others for market share, a similar approach by the TSP is unnecessary, unwelcome by the participants who pay the freight, and in my view, inconsistent with the provision in our law requiring that the Board develop policies providing for low administrative costs. 5 U.S.C. §8475. The draft also suggests the use of "pop-ups" and a web icon to obtain feedback from all users. However, many web users find pop-ups extremely annoying, and analyzing ongoing feedback from 3.3 million participants, absent evidence from the GAO of the need for such an undertaking, would be imprudent. The draft further suggests the creation of "customized account statements." I am unaware of any need or justification for such a feature in the TSP, and even the industry views this as an unnecessary and expensive frill. Had we been made aware in advance of some of the suggestions that were to appear in the draft, they could have been discussed and potentially resolved. For example, the draft suggests connecting TSP calculators to actual individual account balance information, a feature that was offered by one manager and arguably provides some marginal convenience. However, we made a deliberate decision to not connect such functionality to our secure server for both security and privacy reasons. Had this been discussed in advance, we expect the evaluators would have recognized the wisdom of our approach and not included the suggestion in the draft. There were other such examples as well. Finally, although I have attempted to address the contents of the draft report, I must emphatically note that it fails in any way to provide a cost/benefit analysis for its recommendations and other suggestions. The very last sentence of the draft, that such an analysis is necessary, adds nothing, since cost is the most critical point for the consideration of operational changes in the daily valued defined contribution sector. Anyone familiar with industry practice knows that daily plan cost is determined by expense ratio and per participant analysis. The TSP's costs are far lower, on both measures, than any industry provider or sponsor. The obvious absence of any cost/benefit data substantially diminishes the value and credibility of the draft. If the Government Accountability Office intends to publicly issue recommendations and suggestions to alter the design or operations of the TSP, it should at a minimum support its proposals with its cost/benefit analysis. This is especially important since TSP costs are borne by the participants, and ultimately reduce their retirement account balances. In an effort to correct these and other deficiencies, and produce a draft that is more responsive to the Chairman's request, I have asked my staff to remain available to review with your staff the current draft, line by line if necessary, to create a more useful product. If you would like to do so, please feel free to contact Tom Trabucco at (202) 942-1641. Very truly yours, Signed by: Gary A. Amelio: [End of section] FOOTNOTES [1] Assets and participants are as of December 31, 2003. [2] As provided under 26 U.S.C. 401(k) of the Internal Revenue Code, certain defined contribution plans allow employees to reduce a portion of their current compensation via a contribution, on a pretax basis, to a qualified retirement account. Generally, there are two types of pension plans: defined benefit plans and defined contribution plans. A defined benefit plan promises a benefit that is generally based on an employee's salary and years of service. The employer is responsible for funding the benefit, investing and managing plan assets, and bearing the investment risk. In contrast, under a defined contribution plan, benefits are based on the contributions to and investment returns on individual accounts, and the employee bears the investment risk. [3] Customer service for many private and public sector defined contribution plans is delivered by a private sector "plan manager" that is hired by the company that is offering the defined contribution plan to its employees (the plan sponsor). These private sector managers typically handle all of the plan's customer service-related activities in addition to the plan's record-keeping activities. Most private sector plan managers provide services to several different plan sponsors. [4] Record keeping includes, but is not limited to, activities such as posting contributions to participant accounts, adjusting account balances for gains and losses, and processing loans and withdrawals. [5] Federal Retirement Thrift Investment Board, TSP's governing body, still seeks to offer new features. For example, TSP managers reported that they began providing toll-free telephone service in July 2004, to enable TSP participants and beneficiaries to obtain account or transaction information via the ThriftLine's automated telephone service 24 hours a day, 7 days a week. [6] The Department of Agriculture's National Finance Center has been the major contractor for account maintenance and participant support since TSP began operations in 1987. [7] We selected 4 of the top 10 private sector plan managers and one of the top three local government agency plans based on total plan assets under management as of December 31, 2002. See appendix I for more detail. [8] Most of the private sector plan managers that we spoke with serviced from several hundred plans to several thousand different plans. We also interviewed the managers of one large self-managed public sector plan. See appendix I for more detail. [9] In the federal retirement system, the Office of Personnel Management (OPM), rather than TSP, is required to establish a training program for all retirement counselors. However, the Board has responsibility for providing information to plan participants to facilitate informed decision making with respect to what level of contribution to make and how to invest those contributions. See GAO, Thrift Savings Plan, HEHS-96-66R (Washington, D.C.: Nov. 14, 1995). [10] Pub. L. No. 99-335, June 6, 1986. [11] FERSA created the Federal Employees' Retirement System (FERS), a three-part retirement system for federal employees, including enrollment in Social Security, a defined benefit plan, and TSP, a defined contribution plan. [12] Civil Service Retirement System (CSRS) is a retirement plan for federal employees and covers employees hired prior to January 1, 1984. It is a defined benefit plan, a plan that specifies the benefit to be received at retirement by the participant. [13] Participants may allocate their contributions and any associated earnings among five investment fund options: the G Fund, F Fund, C Fund, S Fund, and I Fund. The G Fund, managed by the Board is comprised of short-term nonmarketable government securities issued exclusively for the Thrift Savings Fund. The remaining four funds are managed by Barclays Global Investors and are structured to track large index funds, which are debt or equity portfolios comprised of bonds or stocks of a large number of different companies. The first of these funds, the Fixed Income Index Investment Fund, or F fund, is a bond market fund primarily invested in the Barclays U.S. Debt Index designed to track the Lehman Brothers U.S. Aggregate Index. The second fund, the C Fund, is TSP's large-company stock fund. It is invested in the Barclays Equity Index Fund and tracks the Standard & Poor's 500. The Small Capitalization Stock Index Investment Fund, or S Fund, is invested in Barclays Extended Market Index Fund and is managed to track the Wilshire 4500. The I Fund is the TSP's international stock index fund and is invested in Barclays EAFE Index Fund (Europe, Australasia, and Far East) and holds shares of major companies and industries in the European, Australian, and Asian stock markets. [14] 5 U.S.C. 8474 (a)(1) and 5 U.S.C. 8473 (a). [15] The Employee Thrift Advisory Council is comprised of 15 federal employees and consists of four members representing the four largest labor unions; six representing labor, managerial, and supervisory organizations in the U.S. Postal Service; one representing employee organizations having the purpose of promoting the interests of women in government service; one from the organization with individuals receiving the largest number of annuities under FERS or CSRS; one from the organization representing the largest number of supervisors and management officials; one representing the agency with the largest number of members of the Senior Executive Service; and one from the uniformed services. (5 U.S.C. 8473 (b)). [16] 5 U.S.C. 8432; 5 U.S.C. 8432(b); and 5 U.S.C. 8350. [17] According to TSP managers, the ThriftLine receives an average of about 17,000 phone calls per day. [18] TSP has two call centers. The call center located in New Orleans, Louisiana, receives approximately 8,500 telephone calls per day from TSP participants and has about 100 call center representatives answering calls daily. The new call center located in Cumberland, Maryland, which is operated by Spherix Inc., has about 55 call center representatives answering calls daily. [19] We did not assess the reliability of the call center data. See the section on scope and methodology in appendix I. [20] Calls averaged about 3 minutes and 21 seconds plus about 1 minute and 45 seconds for service center representatives to record their notes. [21] In the past, TSP managers automatically mailed paper statements to 3.25 million participants four times per year. To reduce costs, in 2003, TSP managers announced that they would send paper statements only if participants requested them. Since then, only 473,000 participants have requested to continue receiving paper statements, saving $3.3 million dollars per year in administrative costs. [22] Department of Labor's audit guidelines, Nov. 1, 2001. Section 8477(g) of FERSA, directs the Secretary of Labor to establish a program to carry out audits to determine the level of compliance with the act's fiduciary standards. To guide the auditors, the department has developed a fiduciary oversight program that uses detailed guides to test for compliance. [23] Agencies generally refer to retirement counselors and other retirement education staff as benefits officers, according to 5 U.S.C. 8350. [24] 5 U.S.C. 8350. [25] Unlike pop-up advertisements that may be unwelcome to Internet users, the surveys provide participants the opportunity to give feedback through a convenient method according to plan managers, especially if participants were having difficulty with the Web site or if they had suggestions for improving the functionality of the site. The on-the-spot surveys that we observed appeared to be short and easy to fill out. [26] At the September 20, 2004, TSP Board meeting, the Executive Director announced his intention to visit the operations of a few select private sector plan managers. GAO's Mission: The Government Accountability Office, the investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through the Internet. GAO's Web site ( www.gao.gov ) contains abstracts and full-text files of current reports and testimony and an expanding archive of older products. The Web site features a search engine to help you locate documents using key words and phrases. You can print these documents in their entirety, including charts and other graphics. Each day, GAO issues a list of newly released reports, testimony, and correspondence. GAO posts this list, known as "Today's Reports," on its Web site daily. The list contains links to the full-text document files. To have GAO e-mail this list to you every afternoon, go to www.gao.gov and select "Subscribe to e-mail alerts" under the "Order GAO Products" heading. Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability 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: Contact: 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. Government Accountability Office, 441 G Street NW, Room 7149 Washington, D.C. 20548: