| Abstract: | This paper examines the impact of the Child Support Performance and Incentive Act (CSPIA) of 1998 on child support performance measures that are rewarded financially as well as outcomes that are not rewarded. Three of the five performance measures explicitly rewarded by CSPIA are reconstructed in this analysis, as are two child support outcomes that were considered for financial rewards but were ultimately rejected. Using a panel interrupted time series model with state fixed effects and state‐specific trends, this analysis finds that CSPIA had a statistically positive impact on just one rewarded performance goal, cost‐effectiveness, and negatively impacted an unrewarded child support outcome—collections sent to other states. Effect sizes suggest that CSPIA had little impact on child support performance, on balance. These results provide more evidence to the ongoing debate about the ability of performance incentives to improve public sector performance. It also suggests that reforming performance systems in response to perceived problems may create new gaming responses. |
