Computer matching crosses state lines in the Southeast - food stamp program

Every day food stamp managers are discovering more ways to use computers in program operations. At first, managers used computers simply to speed up ongoing functions or for storage and retrieval of information, but now they are using computers more creatively--for example, to detect and measure potential problems,

Food stamp directors in the eight Southeastern States (Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, and Tennessee) have taken advantage of current computer technology to determine how many food stamp recipients are collecting benefits in more than one state in the region.

This effort, called the Southeast Interstate Match Project, is the Largest Cooperative undertaking by state agencies to measure and analyze duplicate food stamp participation. State Exchange funds helped

Representatives from the eight states agencies met in Atlanta last fall to discuss the feasibility of a multi-state match of food stamp files. Their planning session was funded as a state exchange project by the Food nad Nutrition Service's Southeast regional office.

Billy Davis, director of Florida's Division of Public Assistance Fraud, led the discussionand offered to coordinate the initial phase of the project since he and his staff have had extensive experience with computer matching.

Despite idfferences in the way each state's data was collected and maintained as well as other obstacles, all Southeastern state agencies agreed to participate in the project.

State directors realized that not only would they be able to detect those persons fraudulently collecting benefits in more than one state, they would also find answers to some questions that could affect the integrity of the Food Stamp Program on an ongoing basis.

They hoped to find answers to such questions as: Where is duplicate participation occurring--is it just in border counties of neighboring states or in widely separated locations? What are the common characteristics of persons committing this type of violation? What is the most cost-effective way to curb or eliminate the problem?

Although there had been a number of matches to detect duplicate food stamp participation between neighboring states or in selected adjacent counties in neighboring states, no widespread check had been conducted. A multi-state effort was needed to identify trends.

After obtaining food stamp eligibility lists for November 1983 from the various state agencies, Davis' staff created a computer program which matched the Social Security numbers and other available identifiers, such as names and dates of birth, of the 3.3 million food stamp recipients on the eight lists.

According to Davis, the computer matching process was like comparing the contents of one file cabinet with the contents of seven others and then continuing in the same manner until every file cabinet was compared with all others. States found some surprises

Preliminary data from the computer match revealed 6,557 persons who were potentially participating in more than one state. After screening this list to eliminate cases where the Social Security numbers were incorrect or where households, although eligible, were not participating, state coordinators eliminated 26 percent of the matches, leaving 4,521 to be investigated.

Even during the preliminary stages, some surprises appeared. "What surprised me most was that we had cases in Florida that matched with cases in Kentucky," says Davis. "In the past, in considering interjurisdictional matches, people thought the problem was a matter of recipients crossing state lines into bordering states.

"We still found that the majority of each state's matches were with bordering states, but they were not limited to bordering states. Matches were scattered throughout the eight states."

Using forms and procedures field-tested by FNS' Southeast regional office, participating states spent 3 months investigating and documenting cases of duplicate participation. They funneled their results to the regional office for compliation and analysis. A report of the findings will be available shortly. Duplication is not widespread

In evaluating the Interstate Match Project, Cherie Morris, FNS' food stamp director for the Southeast region, says, "We went into this project not knowing what we would find but hoping we could dispel the myth that duplicate participation is widespread.

"Our data indicates that while duplicate participation is occurring, it is not a serious problem. As we analyze our data further, we should be able to tell more about where instances of duplicate participation occur most often and what type of recipient is likely to commit such a violation.

"I'm very produ of the directors and staff of the Southeastern state aencies who overcame technical problems with a cooperative spirit to achieve this match," she adds. "I hope the project will not only help eliminate those problems which have been uncovered but will also serve as a model for other efforts of this type."

For more information about the Interstate Match Project, contact: Cherie Morris Regional Director Family Nutrition Program Food and Nutrition Service, USDA 1100 Spring Street Atlanta, Georgia 30368 Telephone: (404) 881-4981

COPYRIGHT 1984 U.S. Government Printing Office
COPYRIGHT 2004 Gale Group