School Meals Programs: USDA Has Reported Taking Some Steps to Reduce Improper Payments but Should Comprehensively Assess Fraud Risks, May 21, 2019

This post first appeared on GAO Reports. Read the original article.

What GAO Found

The Department of Agriculture (USDA) has reported various actions aimed at lowering estimated improper payment error rates in the National School Lunch Program and School Breakfast Program (school meals programs). Examples include a new application prototype intended to reduce applicant errors and training for food service workers to reduce administrative errors. USDA uses a model based on a periodic study to estimate improper payments, and reported error rates will generally not reflect the effect of most actions until USDA’s next study is released, likely in 2020. However, in fiscal year 2018, USDA redefined what it considers an improper payment. Specifically, meal claiming errors—for example, meals that are missing a required nutritional component but that are counted as reimbursable—are no longer considered improper payments, resulting in error rates for fiscal year 2018 that are not comparable to prior years.

USDA has not assessed fraud risks in the school meals programs, which hinders its ability to ensure that its key oversight practices—extensive processes designed for broad monitoring purposes—address areas at risk for fraud. The assess component of A Framework for Managing Fraud Risks in Federal Programs (Fraud Risk Framework) calls for managers to plan regular fraud risk assessments and to assess risks to determine a fraud risk profile. USDA officials stated that the agency considers fraud risks through efforts to assess overall program integrity risk in the programs, which include research projects and consideration of specific risks when allocating monitoring resources. However, GAO found that USDA’s efforts to assess risk do not comprehensively consider fraud risks. As a result, these efforts are not aligned with the overarching concepts of planning and conducting fraud risk assessments in the Fraud Risk Framework. Establishing a process to plan and conduct regular fraud risk assessments that align with the leading practices in the Fraud Risk Framework—including those in the figure below—will help USDA design and implement an antifraud strategy, as well as evaluate and adapt its strategy to improve fraud risk management in the school meals programs.

Key Elements of the Fraud Risk Assessment Process

Why GAO Did This Study

In 2018, almost 30 million children participated in the National School Lunch Program and over 14 million participated in the School Breakfast Program, with cash payments totaling almost $17 billion. Historically, the school meals programs have reported high estimated improper payment error rates, which suggest that these programs may also be vulnerable to fraud.

GAO was asked to review improper payment error rates and potential fraud in the school meals programs. This report (1) describes steps USDA has reported taking since 2015 to lower improper payment error rates and (2) examines the extent to which USDA has assessed areas of risk for fraud in the school meals programs.

GAO reviewed the results of the most recent study USDA uses to estimate improper payments in the school meals programs, as well as the error rates and actions to reduce them reported in USDA’s agency financial reports from fiscal years 2015 through 2018. Further, GAO analyzed guidance for key oversight practices and documentation regarding USDA’s risk assessment processes. GAO examined these processes against the leading practices in the Fraud Risk Framework for assessing fraud risks. GAO also interviewed agency officials.

What GAO Recommends

GAO recommends that USDA establish a process to plan and conduct regular fraud risk assessments for the school meals programs that align with the leading practices in the Fraud Risk Framework. USDA generally agreed with the recommendation.

For more information, contact Seto J. Bagdoyan at (202) 512-6722 or bagdoyans@gao.gov.

Leave a Reply

Your email address will not be published. Required fields are marked *