Corporation For National And Community Service: Better Internal Control and Revised Practices Would Improve the Management of AmeriCorps and the National Service Trust

In 2008, the Corporation began phasing out its primary reporting system--WBRS--and moving its functions to a new system--the My AmeriCorps portal. All transactions that have implications for the Trust will be performed in this portal once it is fully functional. The Corporation opted to make this conversion because WBRS would require a number of updates to reflect policy changes in the AmeriCorps programs and the portal is designed in a manner that allows consolidating data from multiple systems. In January 2008, the Corporation's OIG reported that the Corporation had adequate controls to: (1) allow only the refilling of slots vacated by members who served less than 30 percent of their term who did not win and education award; (2) allow refilling of a slot only once; and (3) provide the Corporation with visibility, although not in real time, of when refills have reached 97 percent of awarded slots or when the number of refilled slots reaches 5 percent of awarded slots. The report recommended that the Corporation incorporate automated controls of refills in the My AmeriCorps portal in order to provided real-time visibility of when the 97 and 5 percent thresholds are reached in order to prevent over-enrollment.

The Corporation reported that it submitted nearly 85,000 SSNs (45,605 for FY 2002 and 38,967 for FY 2003) to the Social Security Administration (SSA) for verification in April 2004. The results of the verification were received back from SSA in May 2004 and more than 96% of the SSNs matched without any errors. The files containing the errors were sorted and distributed to the field offices to research and verify the information by comparing to original documents, such as the social security cards of the members, and checking for typographical or data entry errors. The Corporation completed several requests for documents and received approximately an 85% response rate. In December 2004, the Corporation reported that it reinforced its policy and provided guidance to grantees to ensure that member files properly document eligibility. The Corporation submitted about 109,000 records to SSA for the next two fiscal years (53,424 for FY 2004 and 55,633 for FY 2005). Of these, 96% matched without any errors. Most errors were related to last name changes. According to agency officials, the Corporation renewed its Interagency Agreement with SSA to match SSNs of new members enrolled through the end of FY 2006 and planned to submit the data for matching in October 2006.

According to Corporation officials, the Corporation has strengthened the documentation within Oracle Designer by implementing more rigorous style guidelines and adding additional details to the current data dictionary to enable it to be more user-friendly. Several projects have further enhanced the documentation of all aspects of the entire eSPAN system. These include: (1) a documentation style guide; (2) the incorporation of RoboHelp for all online help documentation; and (3) a Report Library, which will serve as repository for all system report requests. To further increase the accuracy of the system documentation, a Business Intelligence project was undertaken to scrutinize the data dictionary and provide an inventory of the entire system and its interconnections. Corporation officials reported that while the initial phase has concluded, the project continues to expand the data being placed in the data warehouse. The Corporation anticipates the system will continue to expand and further refinements will be made to the data dictionary for the foreseeable future and enhancements will continue to be made to some features as the eSPAN system increases in functionality.

The Corporation contracted with Econometrica, Inc. to assess the trust model. Econometrica issued a report in June 2005 documenting testing it performed for six external factors. It recommended incorporating one into the model--undergraduate enrollment. In August 2006, Econometrica issued a second report describing changes it made to the model for the Corporation such as including anticipated trends in undergraduate enrollment.

The Corporation revised its procedures and included steps to monitor Trust activity and deobligate funds for AmeriCorps positions not utilized by grantees. As of September 2004, Corporation officials used an eSPAN query to analyze Trust obligations and determine whether funds could be deobligated. In FY 2006, the Trust reserve had a balance of $40 million and the Corporation requested an additional $7 million for FY 2007. As part of GAO's annual budget justification review of the Corporation's FY 2007 budget request, we advised Congress not to provide the requested $7 million. However, the Corporation responded that the additional $7 million for the reserve would be offset by $9.7 million it expected to deobligate from expired, unused AmeriCorps member slots awarded in prior years. In it's FY 2008 budget justification, the Corporation included a schedule of estimated funds available for deobligation in FY 2007 and FY 2008. It also opted not to request funds for the Trust Reserve. It continued this practice in its FY 2009 budget request.

The Corporation reported that it regularly monitors Trust activity. Per Congressional direction, the Corporation prepares monthly Trust financial reports, which include updated enrollment, earning, and usage data. As of October 2004, the Corporation continued to use conservative assumptions for enrollments to calculate Trust obligations and budgetary needs. In April 2005 and in early 2006, Corporation officials said that its estimating process has a short history and the National Service Trust provides a buffer in the event its estimates are incorrect. The Corporation added that it lacks sufficient data to justify a change to its policy of growing the National Service Trust and when more experiential data becomes available, it will analyze reserve requirements and make adjustments as appropriate. In FY 2006, the Trust reserve had a balance of $40 million and the Corporation requested an additional $7 million for FY 2007. According to the Corporation's FY 2008 and FY 2009 budget justifications, it continues to use the same conservative assumptions in its funding model.

After evaluating the impact of the refill policy on grantees, Corporation officials revised the policy for program year 2004-2005. The revised policy allows programs that have fully enrolled their member slots to replace any member who terminates service before completing 15 percent of his or her term. The policy includes a fail-safe mechanism--refilling will be suspended if total enrollment reaches 97 percent of awarded slots or the number of refills reaches 5 percent of awarded slots. According to the policy announcement, the Corporation planned to modify its data systems to ensure that only eligible programs were able to refill slots and to track refill activity. The Corporation further revised this policy for the 2007-2008 program year to allow grantess to replace members that had completed 30 percent of his/her term, provided the terminating member is not eligible for and does not receive a pro-rated education award.