The study found that under TANF, as implemented in 81 Iowa counties in 2000, there was no impact on the incidence of the types of crimes examined; however, the variables of median family income, percent urban, racial composition, female-headed household, poverty rate, and unemployment did show a significant relationship to the level of both expressive and instrumental crimes. The strength of some of these variables differed across the two crime categories depending on the model observed; for example, more variation in the instrumental crime category was explained by including the percent-urban variable in the complete model, and the variation explained by adding the female-headed household variable was greater for expressive crimes compared to the model that omitted this variable. Possible explanations for the study findings are discussed. Crime data for the 81 counties were obtained from the National Incident-Based Reporting System (NIBRS). Based on NIBRS crime descriptions, they were divided into the instrumental and expressive categories based on the motivation behind the offense. The independent variable used to examine the welfare-crime relationship was the average amount of welfare dollars spent in 2000 per county per recipient. The control variables were categorized as economic structure, population structure, and family structure indicators. Data on all control variables were obtained through the U.S. Census Bureau for 2000. 6 tables, 6 notes, and 37 references