Hearing on the Use of Data Matching to Improve Customer Service, Program Integrity, and Taxpayer Savings

Date: 
Friday, March 11, 2011
Management Issue: 

Good morning, Mr. Chairman, Mr. Doggett, and members of the Subcommittee. I would like to welcome the new members of the 112thCongress. It is a pleasure to appear before you, and I thank you for the invitation to testify today. I have appeared before Congress many times to discuss issues critical to the Social Security Administration (SSA) and the services the Agency provides to American citizens. Today, we are discussing how SSA uses data matching to improve customer service, ensure program integrity, and increase taxpayer savings. 

Data matching has become a critical issue for SSA and other Federal agencies, as they seek ways to improve payment accuracy and reduce or eliminate improper payments.  In November 2009, President Obama signed Executive Order 13520 on Reducing Improper Payments, and in July 2010 signed into law the Improper Payments Elimination and Recovery Act (IPERA).  This legislation sets a goal of reducing wasteful spending by $50 billion by 2012. In response, Federal agencies have increased efforts to pursue data-matching agreements among Federal, State, and local agencies, to help protect Government funds by ensuring that the right person receives the right payment at the right time. 

Identifying improper payments and offering recommendations for solutions to SSA has been an OIG priority for many years. To comply with IPERA, my office and other agency offices of inspector general (OIG), are working closely with their agencies, the Office of Management and Budget (OMB), and the Treasury Department. My office is currently serving as liaison for the Council of Inspectors General on Integrity and Efficiency, working with OMB on improper payment initiatives such as the implementation of Executive Order 13520. 

One of our initial reports on SSA’s computer-matching efforts has led to hundreds of millions of dollars in projected savings for the Agency. The report, Effectiveness in Obtaining Records to Identify Prisoners, released in May 1996, examined whether SSA adequately obtained complete and timely information to determine if prisoners in Federal, State, or county and local corrections facilities collected retirement and/or disability benefits while incarcerated—which the Social Security Act prohibits. SSA entered into computer-matching agreements with corrections agencies, matching prisoner records against Agency benefit records, in accordance with the Computer Matching and Privacy Protection Act of 1988 (CMPPA). 

Despite these efforts, we determined SSA achieved “only limited success” in obtaining prisoner information. SSA had agreements to obtain prisoner data from 47 of the 50 States plus the District of Columbia and the Federal Bureau of Prisons, but the Agency had agreements with just 156 of 3,316 (4.7 percent) county and local corrections agencies, according to our findings. The absence of these agreements at the county and local levels led to significant estimated overpayments to prisoners. We made several recommendations to SSA to improve procedures for obtaining prisoner information, including instituting agreements with corrections agencies to obtain information on all prisoners; and seeking a CMPPA exemption for prisoner-related data matches. 

Because of our work, SSA undertook a major initiative to obtain prisoner data from all State, county, and local corrections departments, and pursued legislation to improve the cost-effectiveness of prisoner data matching. The Ticket to Work and Work Incentives Improvement Act of 1999 eliminated the need for SSA to enter into CMPPA agreements for prisoner matches; the law also included provisions for SSA to provide incentive payments from Old-Age, Survivors, and Disability Insurance (OASDI) program funds to State and local corrections institutions that report prisoner data to SSA. The Agency’s efforts proved successful, as a follow-up OIG report in July 2003 found that SSA had active agreements to obtain prisoner data from all 50 States, the District of Columbia, the Federal Bureau of Prisons, and more than 3,000 county and local facilities. 

The change to CMPPA requirements for SSA and prisoner records matches, as well as the Agency’s expanded efforts to increase its matching agreements, resulted in significant savings for SSA programs. Today, SSA receives prisoner data from corrections facilities monthly, and matches that data against the Agency’s OASDI and Supplemental Security Income (SSI) records, halting benefit payments to prisoners.  In 2006, the most recent year available, SSA’s Office of the Actuary estimated savings from OASDI prisoner suspension provisions were over $580 million per year. 

SSA’s Access to Financial Institutions (AFI) Project is another example of a data-matching initiative that has helped the Agency prevent payment errors that were common in the past. During the initial claims process and later reviews of eligibility, SSI applicants and recipients are required to report their resources to ensure they are eligible to receive payments; SSA studies have found that money held by SSI recipients above the resource limit is a leading cause of payment errors. 

To reduce those overpayments, the Agency implemented AFI as an alternative to the traditional SSI asset-verification process of recipient self-reporting and direct contacts with financial institutions. 

The AFI system checks an applicant’s or recipient’s known bank accounts, and searches for unknown accounts. Because it allows SSA offices to request and receive financial account information electronically, AFI should help the Agency reduce SSI payment errors. AFI has been implemented in 25 States, covering about 80 percent of the SSI population, and SSA plans to implement AFI in the remaining States this year. SSA expects AFI to yield $20 in savings for every $1 spent on the program—for FY 2011, the Agency expects to save $100 million, and by 2013, SSA projects approximately $900 million in lifetime program savings for each year the Agency uses AFI. 

In recent years, my office has released two reports related to electronic bank data: SSI Recipients with ATM Withdrawals Indicating They Are Outside the United States, in April 2008; and SSI Recipients with Excess Income and/or Resources, in July 2008. 

The first report relates to SSI recipients who might not have been eligible for payments because they were outside the United States for more than 30 days. The Agency relies considerably on individuals’ self-reporting their absences from the United States, but because reporting such events might result in ineligibility for SSI payments, individuals have little incentive to communicate with SSA. 

We issued subpoenas to obtain the financial information of SSI recipients, and analyzed the resulting data. Based on a sample, we estimated that SSA failed to detect about $225 million in overpayments because 40,560 recipients did not inform SSA of their absence from the United States. We recommended that SSA explore alternatives that might help detect unreported residency violations, including assessing the feasibility of obtaining electronic bank statements with transaction-level data, so that foreign transactions could be identified and possibly investigated. 

In our second report, we further analyzed the financial information we obtained for the ATM withdrawal audit, and concluded that SSI recipients in our sample failed to inform SSA of changes in income or resources, causing overpayments.We again recommended that SSA obtain electronic bank statement information, in the most cost-effective manner, to include bank account summary and transaction-level data, so that the Agency could identify and investigate additional income and resources. 

SSA’s and the OIG’s efforts to expand the use of prisoner and SSI recipient banking data have detected program vulnerabilities and achieved significant Agency savings. These successful initiatives lend support to a suggestion from Chairman Davis for all government agencies to explore the possibility of developing common data elements and a central point for agencies to share information, with the goal of reducing improper payments and improving customer service.  

My office supports the Chairman’s suggestion for further examination of this issue across the government, as evidenced by the extensive work we have done on the subject in relation to SSA. We have made, to SSA, the following computer-matching recommendations:

  • Use of State Bureau of Vital Statistics (BVS) Records to Detect Unreported Marriages and Divorces, released in June 2003, recommended that SSA establish guidelines to monitor the cost-effectiveness of computer matching, working with State BVS agencies to obtain matching agreements and purchase marriage records to identify beneficiaries who did not report their marriages.
  • Title II Disability Insurance Benefits with a Workers’ Compensation Offset, released in November 2006, recommended that SSA work with States to standardize the format used to report workers’ compensation to SSA; and explore electronic exchanges with the States that maintain automated workers’ compensation databases. 
  • SSA’s Controls and Procedures over SSI Death Alerts, released in May 2007, recommended that SSA continue to encourage State BVS agencies to develop and implement an electronic death registration. 
  • SSI Recipients with Unreported Vehicles, released in July 2009, recommended that SSA assess the costs and benefits of obtaining vehicle information from States or from LexisNexis for SSI recipients, so the Agency can verify individuals’ resources during initial applications and redeterminations. 
  • Disabled Beneficiaries Hiding Wages, released in July 2009, compared beneficiaries’ payment information against their employment information from LexisNexis; identified and referred 300 potential fraud cases to SSA; and recommended the Agency perform work continuing disability reviews (CDRs) on the cases.

In addition, we have planned the following reviews: 

  • Follow-up: Individuals Receiving Benefits Under More than One Social Security Number at Different Addresses 
  • SSI Recipients with Unreported Real Property 
  • OASDI Benefits Affected by State or Local Government Pension 
  • SSI Recipients Who Alleged Being Separated or Divorced 
  • Follow-up: Survivors’ Benefits Paid in Instances When SSA Removed the Death Entry from a Primary Wage Earner’s Record 

My office, while encouraging SSA to pursue computer matches to improve the integrity of its operations, has also sought to use computer matches and data analysis effectively in our own work.  However, the CMPPA has been an obstacle to many OIG projects. Enacted in 1988, the CMPPA amended the Privacy Act of 1974 by adding certain protections for subjects whose records are accessed in computer-matching programs. The CMPPA was passed in response to a growing concern that government agencies would match databases in ways that would invade individuals’ privacy. 

The CMPPA contains several useful and practical exemptions, specifically exempting matches performed for law enforcement purposes, statistical reviews, and congressional investigations, among others. However, computer matches that primarily affect benefit determinations require a formal computer-matching agreement pursuant to the CMPPA. The main objective of many of our audits and investigations is to ensure that only eligible individuals receive payments from SSA; thus, we are required to secure a computer-matching agreement to complete some of our work. 

To conduct a full-scale match, and take action based on the match’s results, our office must go through a lengthy administrative process within SSA before receiving final approval from the Agency’s Data Integrity Board. This review typically takes more than a year, and sometimes several years, to complete; the process can derail planned audits or investigations because the related work is time-sensitive. We can conduct computer matches for research and statistical purposes, but we are unable to use the resulting data to affect benefits, make arrests, or take other meaningful action in response to fraud uncovered through such a statistical match. 

In a June 2010 Government Accountability Office (GAO) report, Cases of Federal Employees and Transportation Drivers and Owners Who Fraudulently and/or Improperly Received SSA Disability Payments, GAO matched SSA’s disability beneficiary and recipient data against Federal payroll data to identify Federal employees who were working while collecting disability payments. GAO referred its findings to SSA and OIG, but we would not have been able to undertake this type of work on our own without a computer-matching agreement, under CMPPA limitations.

The IG community is pursuing an exemption to the CMPPA that would permit computer matches related to audits, inspections, or investigations designed to identify weaknesses that make programs vulnerable to fraud, waste, or abuse and to detect improper payments, but the legislation has stalled to date. In 2010, the Patient Protection and Affordable Care Act amended the CMPPA to exempt matches performed by the U.S. Department of Health and Human Services or its Inspector General related to potential fraud, waste, or abuse. We do not have a similar exclusion, though we have proposed similar legislation that would amend the Social Security Act. A CMPPA exemption could serve as a vital tool in facilitating our ongoing mission to combat fraud, waste, and abuse in SSA’s programs and operations.

Our recommendations related to computer-matching agreements support the organization’s primary focus on integrity. Further, we continue to pursue the establishment of a self-supporting program integrity fund for activities such as our Cooperative Disability Investigations program (CDI), CDRs, and redeterminations, to ensure that applicants and beneficiaries are eligible at the time they apply and as long as they remain in payment status. The proposal would provide for indefinite appropriations to make available to SSA 25 percent, and to OIG 2.5 percent, of actual overpayments collected based on detection of erroneous overpayments SSA collects. These funds would be available until spent for stewardship activities. 

In conclusion, my office is dedicated to working with SSA to identify data matches that can improve the efficiency and integrity of the Agency’s operations and the delivery of benefits to the American public. Data matching serves as one piece of a large integrity puzzle for SSA. Increased Agency efforts to pursue future matching agreements, and a CMPPA exemption to allow my office to pursue computer matches related to potential program fraud, waste, or abuse, would further our collaborative effort to protect SSA funds for the Americans who are eligible for them. As Chairman Davis has suggested, the Federal Government as a whole should explore the possibility of data matching across all agencies and programs to improve its service to the American public.  Just as SSA strives for payment accuracy, so too should all other government agencies. We will continue to provide information to Agency decision-makers and this Subcommittee, and we look forward to assisting in these and future efforts. 

I thank you again for the invitation to be here with you today. I’d be happy to answer any questions.