When a computer system is able to reduce tax expenditures by millions of dollars and generate as much as $27 in savings for every dollar spent on expenses, you would think state officials would be scrambling to use it. But that's not the case with the Public Assistance Reporting Information System, better known as PARIS, a four-year-old system designed to reduce improper payments in public assistance programs.

So far, only 16 states are using PARIS to identify individuals or families who may be receiving benefit payments from TANF (Temporary Assistance for Needy Families), Medicaid or Food Stamps in more than one state, according to a report by the General Accounting Office (GAO).

In February 2001, PARIS identified almost 33,000 instances in which improper payments were potentially made to individuals who appeared to reside in more than one state. Just under half of the potential improper payments involved Medicaid benefits; the rest involved some combination of TANF, Medicaid and Food Stamps. So far, four states and the District of Columbia have collected data on the benefits of the interstate matching system and have documented $16 million in savings.

According to the GAO, their analysis suggests PARIS could help other states save program funds by identifying and preventing future improper payments. Among the 34 states not participating when the GAO released its report in September were California, Texas, Michigan and Ohio, all of which account for a significant portion of welfare expenditures.

Lack of Information

Each year, the United States spends approximately $230 billion on public assistance, Medicaid and Food Stamps. Millions are lost annually when individuals and families receive duplicate benefit payments from more than one state. Part of the problem has been the lack of information sharing between federal agencies that run the welfare programs and states that administer them.

In 1997, the Department of Heath and Human Services started PARIS so states could share eligibility information and identify improper payment benefits. PARIS works by comparing states' benefit recipient lists with one another using individual social security numbers, as well as name and address information. Computers at the Defense Manpower Data Center search for matches and any hits are forwarded to the appropriate state, where staff can take steps to verify the information and decide whether to cut off benefits.

Few states have taken the time to compare the program's costs to the benefits, but studies of its benefits clearly indicate that computer matching saves tax dollars. For example, Pennsylvania estimated that PARIS uncovered more than $2.8 million in savings in its TANF, Medicaid and Food Stamp programs. Maryland said that it saved $7.8 million in the Medicaid program during the first year PARIS was in operation. Kansas estimated that PARIS produced a savings-to-cost ratio of about 27 to 1.

According to the GAO, if states used data from all three public assistance programs in their matching activities (not all do), the net savings could outweigh the costs of PARIS. On average, the savings-to-cost ratio would be 5 to 1. Based on data provided by the three states, approximately 20 percent of match hits end up valid. In addition to the savings generated by participation in PARIS, states also gain from the program's internal controls that help ensure public assistance payments are only made to or on behalf of people who are eligible for them.

Limited Use

Even with the success so far, PARIS has been limited in its effectiveness. Most notably, only one-third of the states participate, leaving a large portion of the public assistance population not covered by the matching system. Second, PARIS has been hampered by coordination and communication problems among its participants. Third, some participating states give PARIS low priority, resulting in many duplicate payments left unresolved. Finally, the system suffers from the fact that it can't prevent duplicate payments from occurring, but can only identify and stop those