Improper payments from government agencies have been a long-standing and significant problem. While only a small percentage of overall government payments, they account for hundreds of billions of dollars in unnecessary expenses for federal, state and local government agencies. Despite the efforts of the 2002 Improper Payments Information Act and the Eliminating Improper Payments initiative, the federal government estimates that, in 2006 alone, more than $40.5 billion in improper payments were made by federal agencies.1 When combined with the improper payments being made by state and local agencies, the magnitude of the problem seems insurmountable.

 

Improper payments are defined as “when federal funds go to the wrong recipient, the recipient receives the incorrect amount of funds, documentation is not available to support a payment, or the recipient uses the funds in an improper manner.”2 These payments largely result from incomplete or inaccurate data, human errors and fraud.

 

 

Federal, state and local agency officials are faced with the challenge of identifying both the extent and root cause of improper payments and with reducing risk and correcting problems. While much progress has been made in recent years, agency officials are looking at new solutions that will provide more timely and accurate identification of potential payment errors or fraud.

 

Reducing Improper Payments with Improved Data Quality

 

Most agencies today are already using sophisticated, statistics-based technology to assist them in identifying transactions that may represent improper payments. These detection applications look for anomalies in transactions or related sets of transactions (e.g., bills for incompatible treatments for the same individual). If a transaction is identified as suspect, it is either denied or investigators are alerted to manually inspect it to determine whether a payment is improper.

 

These systems determine related sets of transactions using fixed identifiers such as Social Security number, drug enforcement administration number, taxpayer identification number, etc. Errors in these identifiers, either deliberate or unintentional, limit the ability of the detection applications to identify improper payments. While these detection programs are asking the right questions, they don’t always have a complete and accurate set of data which they can be applied. Master data management (MDM) can help agencies solve this problem.

 

Instead of relying on one identifier, MDM solutions integrate disparate data simultaneously, providing a single, comprehensive view of all the different data associated with a particular person or other entity. The most accurate systems apply probabilistic matching techniques to the data, rather than less accurate rules-based or other traditional approaches.

 

Using Probabilistic Matching Within MDM

 

Probabilistic matching systems reduce error rates by using likelihood ratios, statistical theory and data analysis to accurately identify relationships between disparate, fragmented data, even data with complex typographical errors and error patterns. Because probabilistic systems pinpoint variation and nuances to a much finer degree, they are able to identify associations that elude more traditional methods. Agencies that apply existing statistical-based detection applications will be able to more accurately identify improper payments, reduce the need for manual review of suspect transactions and benefit from more accurate results.

 

In addition to ensuring that detection and analysis systems have the most complete and accurate information possible, MDM also provides an integrated view of large amounts of data in real time without requiring that data be moved into a centralized repository or hub. MDM solutions that use a registry-based model enable agencies to easily and rapidly deploy solutions that match and link data from disparate systems without causing significant disruptions to their operations.

 

Registry-style solutions house data in each existing contributing source system or database. Data owners remain autonomous and retain control and responsibility for maintaining their data. More importantly, a registry-style solution can enable data from hundreds of databases to be easily shared, even when data cannot be centralized because of regulatory, privacy or other business reasons.

 

Agencies seeking to solve their improper payments problem can achieve significant benefits from deploying advanced MDM solutions that are highly accurate, do not require that data be centralized and are capable of handling massive volumes of data in real time. When the MDM solution feeds this significantly higher quality data into the detection systems already in place, agencies should be able to significantly reduce improper payment rates, which can translate into billions of dollars in cost savings annually for federal agencies alone.

 

Benefits Beyond a Single Agency

 

Once registry-based MDM solutions are in place at multiple agencies, accurate data can be shared between agencies, which have the potential to provide additional benefits to both agencies and citizens.

 

Cross-agency data sharing can help detect situations in which people are receiving benefits from multiple agencies when they should only be receiving benefits from one. An example would be when a person is receiving temporary assistance for needy families or other state-based cash benefits on behalf of the same dependent child from two states at once. This situation is sometimes encountered in states where there are two cities near state lines, which makes it easy for people to register and receive benefits in both states. These types of errors or improper payments would not be detected with data from a single agency, but could be identified when data is shared between agencies and accurately identified as belonging to the same person.

 

In addition, cross-agency data sharing could help identify situations where people should be receiving benefits but aren’t because they did not realize they were entitled to them. If data is shared across agencies, when someone registers for Medicaid, for example, the agency would be able to view whether that person is receiving food stamps, aid to a family with dependent children and other benefits provided to low-income individuals. The agency could then inform the applicant that she/he is likely to qualify for these additional benefits or even to enroll them into those services as well.

 

Optimized Data Saves Tax Dollars

 

The pressure to consolidate programs, reduce duplication and streamline processes is only going to get stronger for government officials, as will the public demand for direct and easy access to government programs and information.

 

The good news is that data quality can be improved to help agencies better use and manage citizen data to meet a host of missions. MDM solutions can help increase the efficiency and effectiveness of government programs and still properly safeguard the public’s personal information.

 

By using MDM to provide accurate, real-time identification and validation of citizen data at multiple touchpoints, government organizations will gain more confidence in the accuracy of its citizen data. This will result in a bevy of benefits for a variety of federal, state and local programs in areas such as health and benefits administration, law enforcement, homeland security, and defense and intelligence.

 

References:

  1. United States Government Accountability Office. " Improper Payments: Federal Executive Branch Agencies’ Fiscal Year 2007 Improper Payment Estimate Reporting." www.gao.gov, January 23, 2008.

  2. United States Government Accountability Office.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access