Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
An important source of errors in transfer payments programs is the fraudulent misreporting of earnings received by recipients. We propose a model of the recipient's decision to report income and the expected penalties if caught engaging in fraud. We discuss several features of the detection and penalty structures in the AFDC and Food Stamps programs and present estimates of a model of the duration of fraudulent earnings-misreporting episodes. We find that recipients respond to increases in the returns to fraud by lengthening their episodes of misreporting. However, the level of fraudulent activity at any given time is rather low.