What GAO Found
Over the last 10 years, the U.S. Department of Agriculture (USDA) has reported that improper payment rates for the Supplemental Nutrition Assistance Program (SNAP) have ranged from an estimated 5.8 percent to 3.2 percent of all payments, likely reflecting, in part, certain policy changes and calculation methods. Many factors affect low-income households' eligibility for SNAP and the amount of benefits they receive, creating multiple opportunities for errors in the eligibility determination process conducted by states. However, GAO found that certain state or federal program changes can affect the likelihood of these errors. For example, when states adopted available policy flexibilities that simplified or lessened participant reporting requirements, these changes reduced the opportunity for error and led to a decline in the improper payment rate, according to a USDA study. Conversely, other changes may have led to an increase in the improper payment rate. USDA cited the change from only counting errors over $50 in the rate to counting all errors over $37 as a key factor in an increase in the rate in fiscal year 2014.
SNAP's improper payment rate calculation methodology is generally similar to that used by other large federal programs for low-income individuals, including Medicaid, the Earned Income Tax Credit, and Supplemental Security Income (SSI), though some differences may affect the resulting program improper payment rates. To generate improper payment rates, all four programs draw representative samples of their recipients and report their improper payment rates at high levels of precision. Yet, some methodological differences among the programs likely affect the resulting rates. For example, when there is insufficient information to review eligibility and benefit determination for a selected case under review, Medicaid counts the full benefit amount as an error, SNAP makes an adjustment in the improper payment rate calculation but does not include the full benefit amount as an error, and SSI removes such cases entirely from the sample.
Fraud is also a key indicator of program integrity, and in 2014, GAO found that selected states employed a range of tools to detect potential SNAP recipient fraud, though they faced some challenges. Recipient fraud can occur when applicants make false or misleading statements to obtain benefits or when recipients misuse benefits by exchanging them for cash or non-food goods or services. All 11 selected states that GAO reviewed matched information provided by SNAP applicants and recipients against various data sources to check for accuracy, but efforts varied widely among these states. Some states suggested changing the financial incentive structure to promote fraud investigations. Some states also reported limitations of USDA's required approach to monitoring benefit card replacements, and GAO developed a more targeted approach by combining data sources to identify households potentially engaged in trafficking. In addition, although USDA had increased its oversight of state anti-fraud activities since fiscal year 2011, GAO found that USDA did not have consistent and reliable data on states' activities because its reporting guidance lacked specificity. This limited USDA's ability to monitor states and find more effective ways to combat recipient fraud.
Why GAO Did This Study
In fiscal year 2015, SNAP, the nation's largest nutrition assistance program, provided about 46 million low-income people with $70 billion in benefits. USDA and the states partner to operate the program and address issues that affect program integrity, including improper payments and fraud.
This testimony summarizes GAO's recently completed work on SNAP improper payment rates and GAO's 2014 report on recipient fraud. It addresses: (1) the effects of SNAP policies on the rates; (2) how the SNAP improper payment rate calculation methodology compares to those of other federal programs for low-income individuals; and (3) GAO's 2014 findings on efforts to combat SNAP recipient fraud. GAO reviewed relevant federal laws, regulations, guidance, documents, and program data; interviewed relevant federal officials; and gathered information from states. For the 2014 report, GAO also interviewed officials from 11 states that served about a third of all SNAP recipient households, though GAO's results are not generalizable to all states. This testimony also includes USDA's actions to date on GAO's 2014 recommendations.
Reissued on July 8, 2016
In 2014, GAO recommended that USDA take several steps to improve state financial incentives, fraud detection tools, and reporting methods. USDA agreed with these recommendations and has taken some steps to address them. GAO is not making new recommendations at this time.