UNITED STATES DEPARTMENT OF EDUCATION OFFICE OF INSPECTOR GENERAL AUDIT SERVICES March 31, 2015 Control Number ED-OIG/A06O0002 James W. Runcie Chief Operating Officer Federal Student Aid U.S. Department of Education 830 First Street, NE Washington, DC 20202 Dear Mr. Runcie: This Final Audit Report, entitled Pell Grant Lifetime Eligibility Limit, presents the results of our audit. The objective of the audit was to determine if Federal Student Aid (FSA) has controls in place to ensure that students who have met or exceeded the Pell Grant lifetime eligibility limit do not receive additional Pell Grants. Our review covered the audit period July 1, 2012, to June 30, 2014. 1 We concluded that FSA had controls in place to reasonably ensure that students who have met or exceeded the Pell Grant lifetime eligibility limit do not receive additional Pell Grant funds. Therefore, this audit report does not include any findings or recommendations for corrective actions. We provided the draft audit results to FSA for technical comment. FSA informed us it had no comments. BACKGROUND The Consolidated Appropriations Act, 2012, Public Law No. 112-74, enacted December 23, 2011, limits a student’s Pell Grant eligibility to 12 semesters (or the equivalent). The change became effective with award year 2012-2013 and is referred to as the Pell Grant lifetime eligibility limit. Pell Grant eligibility is established on an award year basis. In award year 2013-2014, the maximum Pell Grant was $5,645. Whether or not a student is eligible to receive the maximum Pell Grant amount is based on the student’s Expected Family Contribution (EFC), cost of attendance, and enrollment status. The EFC is determined by a methodology defined by Congress in the Higher Education Act of 1965, as amended (HEA), and is used to establish the 1 The Pell Grant award year is July 1 through June 30. The Department of Education's mission is to promote student achievement and preparation for global competitiveness by fostering educational excellence and ensuring equal access. Final Report ED-OIG/A06O0002 Page 2 of 7 financial need of the student. The scheduled award is the maximum Pell Grant amount a student can receive during an award year, if he or she attends full-time and for a full academic year. In implementing the Pell Grant lifetime eligibility limitation, FSA looks at the percentage of a scheduled award a student used in any award year the student received a Pell Grant disbursement from the inception of the Pell Grant program, award year 1973-1974, forward. 2 This is referred to as the Pell Grant Lifetime Eligibility Used (LEU). A full-time student in a semester program could typically have two semesters in an award year. Under the Pell Grant lifetime eligibility limit, a full-time student in a semester program could maintain Pell Grant eligibility for six award years, which is the equivalent of 12 semesters. A full-time student could receive 100 percent of the scheduled award in each of the six award years resulting in a Pell Grant LEU of 600 percent before becoming ineligible for additional Pell Grant disbursements. A student’s scheduled award for an award year is prorated based on the student’s enrollment status. For example, if a student is half-time, he or she would only receive half of the scheduled award so that only 50 percent of the Pell Grant lifetime eligibility had been used in that award year. A half-time student, who had never received a prior Pell Grant disbursement, would use 50 percent of his or her lifetime eligibility for that award year, resulting in a remaining Pell Grant lifetime eligibility of 550 percent. The Pell Grant LEU in this case would be 50 percent. FSA uses the Common Origination & Disbursement (COD) system to calculate the Pell Grant LEU. The COD system is the system of record for the Pell Grant LEU. FSA uses two warning edits to notify schools that a student (1) is approaching the Pell Grant LEU limit of 600 percent (between 450 and 600 percent) or (2) has exceeded the limit. FSA has also implemented a rejecting/correcting edit that prevents a school from disbursing Pell Grant funds to a student unless the school can show that the student is actually eligible. AUDIT RESULTS FSA has controls in place to reasonably ensure that students, who have met or exceeded the Pell Grant lifetime eligibility limit, do not receive additional Pell Grant funds. Specifically, we found no errors when testing the accuracy of the LEU calculation for a statistical random sample of Pell Grant awards. Further, we tested controls for a judgmental sample containing different scenarios for awards having LEUs calculated at or above 600 percent and found no significant issues. FSA relies heavily on edits within the COD system to ensure that students who have met or exceeded the Pell Grant limit of 600 percent LEU do not receive additional Pell Grants. FSA uses two warning edits to notify schools that a student (1) is approaching the Pell Grant LEU limit (between 450 and 600 percent) or (2) has exceeded the Pell Grant LEU limit. FSA also implemented a rejecting/correcting edit code that prevents a school from disbursing Pell Grant 2 Prior to 1980, the Pell Grant was called the Basic Educational Opportunity Grant; later renamed the Federal Pell Grant after Senator Claiborne Pell, to honor his efforts in creating the program. Final Report ED-OIG/A06O0002 Page 3 of 7 funds to a student with an LEU of 600 percent unless the school can show that the student is actually eligible. In award year 2012-2013, FSA only used the warning edits because FSA was not sure of the quality of the historical Pell Grant data. Because the warning edits did not require action on the part of schools, FSA monitored Pell Grant LEUs that exceeded 600 percent, using weekly reports, and identified Pell Grant overpayments of $13.7 million for award year 2012-2013. FSA has been able to collect most of the overpayments. As of October 22, 2014, only $10,179 in overpayments (the majority of which ranged from $1 to $100) remained uncollected, representing 2,205 students at 805 schools. FSA implemented the rejecting/correcting 3 edit for award year 2013-2014. According to FSA’s Director of Grants and Campus Based Programs, there were few disputes regarding the accuracy of historical data during the first year of operation of the warning edits. As a result, FSA determined that the data was sufficiently reliable to proceed with implementation of the rejecting/correcting edit in award year 2013-2014. We tested the rejecting/correcting edit and the warning edits by reviewing the edits on Pell Grant awards from award year 2013-2014 for a judgmental sample of 53 awards out of a total of 44,169 awards which had LEU’s of 600 percent or higher. Our sample focused testing on awards where we perceived higher risk. We identified 11 out of the universe of over 10 million Pell Grant awards, in excess of the 600 percent Pell Grant limit, which we considered to be immaterial. Based on the results of our sample, we concluded that the two warning edits and the rejecting/correcting edit were generally working as intended. Initially, FSA only had Pell Grant disbursement data available in the COD system to calculate the Pell Grant LEU from award year 1999-2000 forward. Although not required to maintain the historical data, FSA had archived data from award years 1973-1974 to 1998-1999, with the exception of award year 1974-1975. FSA uploaded the archived data to the COD system to calculate the Pell Grant LEU. We reviewed the LEU calculations for 147 randomly selected Pell Grant awards out of a total of more than 10 million awards made for award year 2013-2014. Our testing of the accuracy of the Pell Grant award LEU calculations for the award year 2013-2014 consisted of reviewing calculations for a stratified random sample of Pell Grant awards and reviewing the calculations and confirming the presence of the previously archived Pell Grant data. We found no LEU calculation errors in our sample. 4 Further, during the course of testing the LEU calculation, we found evidence that archived Pell Grant data from as early as award year 1975-1976 was included in the calculations. Based on the results of the sample, we conclude that the LEU calculations for award year 2013-2014 were accurate. 3 The rejecting/correcting edit will either reject or correct the disbursement, depending on the school’s processing options set within the COD system. For a rejecting school, the COD system will reject the Pell Grant disbursement for students that go over the 600 percent LEU. For a correcting school, the COD System will reduce the disbursement so that the recipient’s LEU does not exceed 600 percent LEU. 4 This sample size was determined so that if no errors were encountered we would have at least 95 percent confidence that the LEU calculation error rate did not exceed 2.5 percent. Final Report ED-OIG/A06O0002 Page 4 of 7 OBJECTIVE, SCOPE, AND METHODOLOGY The objective of our audit was to determine if FSA has controls in place to ensure that students who have met or exceeded the Pell Grant lifetime eligibility limit do not receive additional Pell Grants. Our review covered award years July 1, 2012, to June 30, 2014. For Pell award year 2012-2013, we reviewed controls in place for the warning edits FSA established. Our sample testing of awards covered the Pell Grant award year 2013-2014 when the three edits, two warning and one correcting/rejecting, were in place. To achieve our objectives, we performed the following procedures: 1. Reviewed relevant laws and guidance, including the Consolidated Appropriations Act, 2012; section 401(c)(5) of the HEA; FSA’s Dear Colleague Letters GEN-12-01 and GEN-13-14; and various FSA Electronic Announcements which implemented the Pell Grant lifetime eligibility limit. 2. Reviewed FSA’s contract documents pertaining to the COD system, including Change Requests, General Design Documents, and Production Readiness Reviews that were relevant to the implementation of the Pell Grant lifetime eligibility limit, specifically the LEU calculation, LEU Detail Reports, warning edits, and the correcting/rejecting edit. 3. Interviewed FSA officials within the Business Operations, Grants and Campus Based Division; the Deputy Chief Compliance Officer; the Policy Liaison and Implementation Director; and the Division Manager for Change Management. 4. Interviewed at Accenture, FSA’s contractor for the COD system, the Contract Manager, Client Service Delivery Director, Solution Manager, and the Testing Director. In order to answer our audit objective, we performed two different forms of tests. The first type of testing we performed was to assess the accuracy of the LEU calculations in the COD system. The second type of testing was to assess whether the COD warning edits, put in place for the Pell Grant award year 2012-2013, and correcting/rejecting edit, put in place in the Pell Grant award year 2013-2014, prohibited students from receiving Pell Grants in excess of 600 percent LEU. We performed these tests on two separate samples derived from a universe of 10,093,209 Pell Grant awards from award year 2013-2014. LEU Calculation Accuracy Testing In order to assess the accuracy of the LEU calculation for Pell Grant awards, we tested a statistical random sample from the universe of Pell Grant awards. We tested selected awards by reviewing calculations for accuracy and we also reviewed selected awards to determine if calculations included archived data. Final Report ED-OIG/A06O0002 Page 5 of 7 We obtained the universe of Pell Grant awards for award year 2013-2014 from FSA using data from the COD system. The universe consisted of 10,093,209 Pell Grant award records. We split the universe into two strata based on date of birth of the Pell Grant recipient to increase the chance that our sample would contain award records with an associated LEU based, in part, on the archived data for award years 1973-1974 through 1998-1999. We selected a stratified random sample of 147 Pell Grant awards as shown in Table 1. We established the sample size so that if we encountered LEU calculation errors (the calculation was incorrect or did not include all Pell Grant award and disbursement data) we could project an error rate and if we did not encounter errors we would have at least a 95 percent confidence rate that the incorrect LEU calculations in the 2013-2014 award universe did not exceed 2.5 percent. For each Pell Grant recipient included in our sample, we obtained the historical Pell Grant award data from the COD system. Using this data, we recalculated a Pell Grant LEU percentage for each student. To determine that the COD system’s Pell Grant LEU was accurate, for each student, we compared the Pell Grant LEU percentage that we calculated to the percentage contained in the COD system. We also reviewed award history of sampled students to confirm the presence of archived data in the LEU calculation. Table 1. Universe and Sample Size Assessment of Pell Grant Awards for LEU Accuracy Testing Stratum 5 Universe Size Sample Size Birthdate of award applicant on or before 993,930 35 January 1, 1974 Birthdate of award applicant after 9,099,279 112 January 1, 1974 Total Pell Grant award universe 10,093,209 147 LEU Edit Control Testing Warning and Correcting/Rejecting Edit We tested a judgmental sample of Pell Grant awards to determine whether the LEU edits were functioning as intended. We limited the universe identified above to the 44,169 award records where the LEU percentage was greater than or equal to 600 percent. We narrowed our scope to this group because those award records should have had at least one LEU edit associated with the record. Based on data analysis, we categorized the Pell Grant awards into one of four risk-based groups based on, in order of risk: disbursements made and the LEU limit was exceeded, existence of disbursements with no edit status, existence of disbursements with warning edits, and the absence of disbursements. Within each risk category, we randomly selected Pell Grant awards for testing, as shown in Table 2. Both the risk categories and sample sizes selected from each category were intended to identify awards we perceived to be at higher risk of recipients receiving Pell Grants in excess of 600 percent or having inaccurate LEU edits. 5 We stratified the population and sample using student birthdate of January 1, 1974, in order to increase the chances of selecting a student whose LEU calculation would have included archived data. Final Report ED-OIG/A06O0002 Page 6 of 7 The category with the highest risk included records that showed a Pell Grant disbursement for 2013-2014 and had a Pell Grant LEU greater than or equal to 600 percent. This category would be the riskiest group since no LEU should have been greater than 600 percent. The category with the second most risk included records that showed a Pell Grant disbursement for 2013-2014 and no warning edits in the universe file. Since this category had a 2013-2014 Pell Grant award amount disbursed and was currently showing no warning edits, there was some risk. The category with the third most risk included records that showed a Pell Grant disbursement for 2013-2014 and had a warning edit shown in the universe file. This category would have some risk since there was a disbursement but the risk would be reduced due to the warning edits shown in the universe file. Lastly, the category at least risk included records that did not have a Pell Grant disbursement for 2013-2014 and many records included a warning edit in the universe file. This category had a small amount of risk since there was no Pell Grant disbursement for 2013-2014. For each of the 53 awards sampled, we reviewed the COD system to determine that the applicable warning edit and correcting/rejecting edit functioned as intended and that the student’s Pell Grant LEU did not exceed 600 percent. As shown in Table 2, we identified 11 Pell Grant awards in excess of the 600 percent Pell Grant limit. We followed-up with FSA as to the cause of each disbursement where the Pell Grant LEU exceeded 600 percent. However, we considered the 11 Pell Grant awards out of over 10 million Pell Grant awards to be immaterial. Because our results are based, in part, on a judgmental sample, they cannot be projected to the entire universe of Pell Grant awards. Table 2. Universe and Sample Size by Level of Risk for LEU Edit Control Testing Risk Categories Category Universe Size Sample Size Highest Risk LEU in excess of 600 percent with 11 6 11 disbursement 2nd Highest Risk LEU equal to 600 percent with a 1,749 27 disbursement and unknown edit status 3rd Highest Risk LEU equal to 600 percent with a 38,079 10 disbursement and known warning edits made Least Risk LEU equal to 600 percent with no 4,330 5 disbursement made Total of Pell Grant awards with LEU equal 44,169 53 to or in excess of 600 percent Data Reliability We relied on computer-processed data from the FSA’s COD system, which was provided by FSA, to perform our detail testing of the COD system’s Pell Grant LEU calculation, the warning edits, and the correcting/rejecting edit. The data from the COD system consisted of the Pell Grant universe for award year 2013-2014 and Pell Grant award history. To evaluate the reliability of this data, we gained an understanding of the design and implementation of the COD system’s Pell Grant LEU calculation, the warning edits, and the correcting/rejecting edit through 6 Of the 11 records, there were 10 distinct students. One student had Pell Grant disbursements at two different schools for the award year 2013-2014. Final Report ED-OIG/A06O0002 Page 7 of 7 interviews with FSA and Accenture officials (FSA’s contractor), and reviewed other relevant documents and reports. We also compared Pell Grant disbursement data in the award year 2013-2014 universe file to Pell Grant disbursement data obtained from the National Student Loan Data System, another data system maintained by the U. S. Department of Education, to assure the data was reasonably complete. At the time of our audit, the Pell Grant program had been in existence for 41 award years, since award year 1973-1974. FSA was missing data for the second year of the Pell Grant program, award year 1974-1975. Within the context of our audit objectives, this missing data was considered minor and did not affect the reliability of the data. We concluded that the computer- processed data provided by FSA was sufficiently reliable for the purposes of our audit. We performed audit field work from March 3, 2014, through October 2, 2014, at FSA’s offices in Washington, D.C., and our offices in Dallas, Texas, Pittsburgh, Pennsylvania and New York, New York. We held an exit conference with FSA officials to discuss the results of the audit on December 10, 2014. We conducted this performance audit in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. In accordance with the Freedom of Information Act (5 U.S.C. §552), reports issued by the Office of Inspector General are available to members of the press and general public to the extent information contained therein is not subject to exemptions in the Act. We appreciate the cooperation given us during this review. If you have any questions, please contact Daniel P. Schultz, Regional Inspector General for Audit, at 646-428-3888. Sincerely, /s/ Patrick J. Howard Assistant Inspector General for Audit
Pell Grant Lifetime Eligibility Limit
Published by the Department of Education, Office of Inspector General on 2015-03-31.
Below is a raw (and likely hideous) rendition of the original report. (PDF)