Penalty Structures and Deterrence in a Two-Stage Model: Experimental Evidence

Increasing penalty structures for repeat offenses are ubiquitous in penal codes, despite little empirical or theoretical support. Multi-period models of criminal enforcement based on the standard economic approach of Becker (1968) generally find that the optimal penalty structure is either flat or declining. We experimentally test a two-stage theoretical model that predicts decreasing penalty structures will yield greater deterrence than increasing penalty structures. We find that decreasing fine structures are more effective at reducing risky behavior. Additionally, our econometric analyses reveal a number of behavioral findings. Subjects are deterred by past convictions, even though the probability of detection is independent across decisions. Further, subjects appear to take the two-stage nature of the decision making task into account, suggesting that subjects consider both current and future penalties. Even controlling for the fine a subject faces for any given decision, being in a decreasing fine structure has a significant effect on deterrence.