Computer-Aided Techniques Against Public Assistance Fraud - A Case Study of the Aid to Families With Dependent Children (AFDC) Program

• NCJ Number: 86514
• Author(s): Michael Fischel, Mitre Corporation; Lawrence Siegel, Mitre Corporation
• Date Published: January 1980
• Publication Type: Publication

Annotation

An assessment of computer-aided techniques to address fraud in the Aid to Families with Dependent Children (AFDC) program concludes that such methods play only a supporting role in preventing and detecting fraud and that their effectiveness has been limited by poor data quality and inadequate administrative and managerial support.

Abstract

Report information was collected from a literature survey, discussions with Federal officials, telephone interviews with AFDC staff in 19 States, and site visits to 6 of those States. The study identifies four issues that are pertinent to understanding the use and difficulties in applying computer-aided techniques to AFDC fraud: the complexity of the program's rules and regulations, the intergovernmental nature of its administration, pressures to control increasing administrative costs, and the nature of the electronic data processing support. HEW estimates that 3 to 13 percent of all AFDC cases are involved in some form of fraudulent claims, obtaining as much as $600 million in public funds illegally. Misrepresentation of facts concerning income apparently is the most prevalent fraud. The report describes and compares the following antifraud techniques: computer-aided matching in wage, jurisdictional, and benefits areas; selective case action; and selective case screening. Hard evidence on their effectiveness is lacking, but detection of fraud based on computer-generated leads is highly dependent on the availability of local staff to conduct case reviews and collect evidence. Cost-effectiveness data is also scarce, but preliminary information raises questions about the utility and cost of computer techniques as currently employed. The effective use of computer-aided techniques is significantly influenced by the sufficiency of data bases used, the adequacy of administrative and managerial support, reliable cost and evaluation data, and restrictions emanating from privacy considerations. Suggestions regarding the LEAA's future role in supporting welfare agency efforts to curtail AFDC fraud conclude the report. Tables, diagrams, and approximately 120 references are included. For related material, see NCJ 67366, summary.