Peggy E. (Peg) Gustafson was sworn in as the Inspector General of the U.S. Small Business Administration on October 2, 2009. Ms. Gustafson previously served as General Counsel to Senator Claire...
Statement of Hon. Eric M. Thorson, Inspector General, U.S. Small Business Administration Before the Committee on Small Business and Entrepreneurship United States Senate July 25, 2007
Chairman Kerry, Ranking Member Snowe and members of the Committee, I appreciate the invitation to appear before you to discuss our work on the effectiveness of the efforts made by the Small Business Administration (SBA) to cope with the aftermath of Hurricanes Rita and Katrina. With me today is Debra Ritt, the Assistant Inspector General for Auditing. She and her staff have worked diligently to identify and conduct audits of SBA disaster relief procedures that provide the most value to the Agency and hurricane victims. It is a selection of these audits that is the subject of my testimony today.
Shortly after assuming the position of Inspector General, I personally visited the New Orleans area and walked on the deserted streets of the worst hit areas, surrounded by total devastation. Just being there was a life altering experience. It was dramatically apparent that our oversight work, to be of the most value, would need to begin immediately and cover nearly every aspect of the SBA disaster loan process.
Within 90 days of my visit, the Office of Inspector General had established a new regional office in New Orleans putting both investigators and auditors at the scene. Our audits of the various Agency disaster programs have since covered a wide range of issues, and have uncovered some discouraging problems.
In the Fall of 2006, SBA initiated the “90 in 45” campaign to contact every borrower, some 90,000, whose loan had not been fully disbursed, within 45 days in order to properly and rapidly disburse funds for loans approved for victims of Hurricanes Katrina, Rita, and Wilma. The audits that I will discuss today were initiated either in response to complaints from SBA employees about problems allegedly resulting from this campaign or other concerns about SBA’s efforts to expedite loan disbursement.
As a referral from this Committee, we received disturbing written allegations that indicated thousands of already approved disaster loans were cancelled in what appeared to be an effort to improve statistics that focused negative attention on a huge backlog of approved but undisbursed loans. While certainly problematic, this does not diminish the Agency's great effort that increased disbursements from only $2 billion in August of 2006 to $5.5 billion by the end of the year. We have just issued internally a report, still in draft, that evaluates whether SBA unnecessarily cancelled already approved loans in 2006, as alleged by former SBA employees. Our audit focused on SBA’s Buffalo Customer Service Center, which processed about half of the over 40,000 loan cancellations made last year, of which about 12,000 loans were cancelled during the first two weeks of September 2006.
The loans were cancelled at the direction of SBA’s Fort Worth Processing and Disbursement Center, who instructed the Buffalo Center to make three attempts to call each borrower within 24 hours. If they could not reach the borrower, they were to cancel the loan. The Director of the Fort Worth Center stated to our staff that these instructions were intended to get the loans off of SBA’s books so that SBA did not have to report a backlog of undisbursed approved loans. Our audit of SBA’s records disclosed that in most cases, SBA made only one attempt to reach the borrower before canceling the loan. A letter of cancellation was sent informing the borrower, and sometimes indicated that the cancellation was at the borrower’s request, even though no such request had been made. The normal time before an SBA disaster loan commitment expires is 60 days. In this case, the loans were past 120 days. While we recognize the effort made by the Agency in allowing an extension to 120 days, given the drastic situation in this area, cancellation within a 24 hour timeframe does not appear to be a fair or prudent course of action. In fact, as we near the 2 year anniversary of these hurricanes, I have no doubt there are many cases where it is still difficult to assemble the documents that might be required to close a disaster loan.
Of the nearly 12,000 cancelled loans, we determined that close to 8,000 were cancelled without any prior notification to the borrowers. These were loans primarily for homeowners and renters who planned to use the loans to rebuild homes and replace personal property destroyed by the hurricanes. This was contrary to SBA’s own procedure, which stated that: “Before we initiate an action to cancel all or any funds, we must mail a letter giving 14 calendar days notice of the pending cancellation. The letter must specify the action the borrower can take to prevent the cancellation.”
We also discovered that when the Buffalo staff annotated its records, the limited options in the computer system recorded it as if the borrower had requested the cancellation.
Several weeks after canceling the loans, the Buffalo Center did again attempt to contact the nearly 8,000 borrowers they were previously unable to reach. Of the 4,500 borrowers eventually contacted, about 1,200 requested that their loans be reinstated. In a sample of borrowers interviewed by our audit staff, some individuals had agreed to the cancellations, but others were upset that their loans had been cancelled and felt that they had been pressured to make a decision on the spot.
If a loan is not reinstated within 6 months from loan cancellation, the borrower is prohibited from ever again reapplying for the loan regardless of need. The inability of borrowers to close on loans within the time restrictions imposed by SBA is extremely problematic in light of the slow progress made to rebuild the infrastructure in the Gulf Coast region. Therefore, we believe SBA’s deadlines for those who were not provided the required advance notice should be extended so that borrowers who are not ready to
close on their loans may still get the assistance they need. The very program the Agency initiated to assist victims in receiving disaster assistance in a more expeditious manner became a program driven, at least for these 8,000 borrowers, more by statistics than the needs of the very people it was trying to assist. The urgency of this audit is that we recognize, unlike in many audits where the numbers portrayed indicate a level of efficiency or productivity, these numbers represent individuals and families who have faced a tragedy most of us cannot imagine.
Two Audits of the 90-in-45 Campaign—“Disbursing Funds Against Borrowers Wishes” and “Securing Collateral”
We issued two audit reports on SBA’s efforts to expedite loan disbursements during its 90-in-45 Campaign. The audits undertaken regarding this campaign were in response to an employee complaint that loans processed during the campaign were being disbursed contrary to borrowers’ wishes and without obtaining all of the documents required to secure the government’s position as a lender.
Regarding the first complaint, we found only a few instances where disbursements were made contrary to borrowers’ wishes. During this campaign, loan disbursements were made without contacting the borrowers, even when SBA had not heard from the borrower in several months, or had not received additional documents needed for disbursement. When unwanted disbursements were received, the borrowers generally returned the funds and loan balances were appropriately reduced. While the number of instances we identified were small and do not reflect a widespread problem, we believe that any disbursements made contrary to borrower wishes is inappropriate as it creates an unwanted debt to the government that can impact his/her access to credit elsewhere. We also determined that SBA had established production goals for the 90-in-45 Campaign, based on the number of loans that had to be processed to reach the 90,000 in 45-day goal. Therefore, there was a strong indication that the goals could not be met without sacrificing customer service and adherence to SBA loan requirements. Because many loan officers and support staff held temporary positions, production goals became an issue of job security. Temporary employees were fearful that if they did not disburse enough loans, they would be terminated. We determined that SBA also gave cash awards to teams that made the
largest number of disbursements. Using cash incentives based solely on the volume of disbursements can lead to hasty decisions to meet production goals, with less attention paid to quality customer service and adherence to SBA legal requirements.
We concluded that while the 90-in-45 Campaign was intended to get disaster funds in the hands of borrowers quicker, the goals and bonuses may have inappropriately influenced employees to make decisions or disburse funds that were disadvantageous to the borrower.
Our second audit of the campaign examined whether SBA took proper steps to protect its interest in collateral on secured loans. We focused on SBA’s
efforts to secure collateral because, at the time of our audit, we found that the Center had nearly 5,000 unprocessed checks received from borrowers for recording and filing fees. Since these checks had not been processed, this indicated that SBA had not recorded the liens on any of these loans. Many of these checks could not be processed because they were either too old or had incorrect information. Over half (2,949) of these checks were over 90 days old and SBA had to request new checks before the liens could be recorded. We also found instances where borrowers’ checks were either written for the wrong amount or to the wrong payee. When checks were incorrect, loan closers did not always follow-up with borrowers in a timely manner to obtain replacement checks, which created further delays in protecting SBA’s secured interests.
Through statistically relevant sampling, the audit disclosed that SBA released an estimated $368 million in loan proceeds on over 3,000 secured loans without perfecting liens on property used as collateral or completing UCC filings.
Quality Assurance Reviews of Loss Verifications
This week we issued a final report on the integrity of SBA’s Quality Assurance Reviews (QAR) of loss verifications. Loss verifications provide the initial damage estimates for repair or replacement of real, personal and/or business property and are used to establish loan amounts under SBA’s Disaster Assistance Program. The report addressed an employee
complaint that QARs of individual loss verifications were inappropriately altered in order to meet performance metrics required under the process outlined in OMB Circular A-76.
Under the A-76 process, government employees may submit a bid to compete with outside contractors to perform the work based on a “Most Efficient Organization (MEO)” study. If the employees win the competition, they must operate as the MEO and meet performance metrics.
In February 2005, a group of SBA employees assigned to SBA’s Office of Disaster Assistance was determined to be the MEO of an A-76 competition and awarded a 5-year contract to conduct the initial loss verifications. The MEO was required to perform within a 2 percent exception. SBA monitored the MEO’s performance through a QAR process, and in July 2006 SBA conducted its first review of the MEO’s performance using a team of ODA loss verifiers that were not assigned to the MEO. The team concluded that
the MEO performed within the 2 percent exception rate. Our audit determined that the QAR process was compromised, allowing a senior official to alter results in order to avoid penalties and to retain the work under the A-76 contract. Specifically, we found that 72 of 246 QARs reviewed were materially altered, lowering the MEO’s true 4.8 percent exception rate for the 246 QARs to 0.6 percent. The 0.6 percent rate was well below the 2 percent maximum exception rate stipulated in SBA’s Letter of Obligation.
We recommended that SBA take steps to ensure that the initial QAR results are documented and any changes justified in writing and coordinated with the reviewers.
The Expedited Loan Program
The Expedited Loan Program was established as a pilot in November 2005 to reduce the processing time for underwriting disaster loans. The pilot was revised in December 2005 with an expiration date of May 15, 2006. It was then extended four times. The Expedited Loan Program used credit scores and a series of critical questions as the means to underwrite loans and to make general loan approval. Unlike standard processing, loans processed under the Expedited Loan Program are approved without an analysis of an
applicant’s repayment ability or verification of his/her stated income and debt. Additionally, loan making decisions by loan officers were not reviewed by a supervisor. The Expedited Loan Program was suspended on April 9, 2007.
As of April, 2007, SBA had approved close to 70,000 expedited loans, totaling $3.7 billion, specifically related to the 2005 Gulf Coast Hurricanes. Loans approved under expedited procedures accounted for nearly 45 percent of all disaster loans approved during this time period.
Our audit disclosed that, based on a statistically relevant sample, about 32 percent of the Gulf Coast Hurricane loans approved under the Expedited Loan Program were made to applicants who lacked repayment ability. Based on the sample results, we estimate that over 21,000 loans, totaling $1.5 billion, were awarded to high-risk applicants who may not be able to repay their loans. Loans for these individuals most likely would not have been approved if they had been processed under standard loan processing procedures. This occurred because the loans were approved based solely on the applicants’ credit scores, without regard to their income and expenses.
Consequently, had loans for these applicants been declined, SBA would have referred the applicants to FEMA for possible grant assistance. While recognizing the great needs of the borrowers, we are concerned that there is a high probability that a significant number of loans processed under expedited procedures will default.
After we briefed the Office of Disaster Assistance on our audit findings, SBA terminated the Expedited Loan Program. In its place, SBA introduced the RAPID Pilot Program. This program was initiated as part of the Administrator’s Action Commitment Excellence (ACE) Campaign. We commend the Agency for expeditiously addressing the shortcomings
identified in our audit of the Expedited Loan Program.
Our audit staff has accomplished a great deal of work on the disaster loan process in a very short time frame. I would be remiss, however, if I did not mention the outstanding efforts that our criminal investigators have made in this same area. With no presence at all in the region one year ago, as of today we have obtained 9 indictments and 1 conviction, and prevented losses to the government of millions of dollars. I am very proud of the work our criminal investigators do each day to help ensure that the money allocated
for victims is not lost to theft or to fraud.
We hope these comments have been both helpful and constructive. These audits I have mentioned are only a portion of the work done in this area. It
is our most sincere desire to use these audits to work further with the Agency to help fix the problems I have outlined above.