Using Michigan Sentencing Guidelines to Estimate Effects of Incarceration and Probation on Reoffending and Employment

在过去的三十年里，美国被监禁的人数急剧增加，这引发了关于监禁对犯罪行为和经济机会的影响的关键问题。关于监禁与缓刑等替代制裁的影响的实证研究往往缺乏明确的因果解释，因为被判入狱/监禁与缓刑的罪犯之间的结果差异，即使是同一罪行，可能是由于这些群体之间存在的和未测量的差异，这些差异影响了量刑决定和所研究的结果。这项拟议的研究旨在通过基于密歇根州量刑指南的两项自然实验，对监禁与监禁或缓刑对累犯和就业的影响提供合理的因果估计。一种设计依赖于随机分配的法官来比较那些仅仅因为被分配到的法官而被判处不同刑罚的罪犯。另一种设计利用了在两种量刑指导变量上对罪犯进行正式评分的方式中出现的不连续性。犯罪记录和犯罪严重程度。该分析比较了在量刑变量中得分非常相似的罪犯的结果，尽管如此，由于他们略低于或高于每个变量的阈值，他们面临着显著不同的监禁概率。这两种设计都限制了对有资格判处相同类型刑罚的个人的比较，这些人因此处于监禁和替代制裁之间的政策边缘。该项目将为刑事司法系统的决策者和行为者提供新的证据，为今后关于监禁与非监禁制裁是否适当的决定提供信息。

The dramatic increase in the number of people incarcerated in the U.S. over the last three decades raises critical questions regarding the effects of incarceration on criminal behavior and economic opportunity. Empirical research on the impact of prison vs. alternative sanctions like probation often lacks a clear causal interpretation because differences in outcomes between offenders sentenced to prison/jail vs. probation, even for the same offense, may be due to pre-existing and unmeasured differences between these groups that affect both the sentencing decision and the outcome under study. The proposed research aims to provide plausibly causal estimates of the effects of incarceration in prison compared to jail or probation on recidivism and employment using a pair of natural experiments based on sentencing guidelines in Michigan. One design relies on the random assignment of judges to compare offenders who are given different sentences due only to the judge to which they were assigned. The other design exploits discontinuities that emerge from the ways that offenders are officially scored on the two sentencing guidelines variables ? criminal history and offense severity. This analysis compares the outcomes of offenders with very similar scores on either/both sentencing variable(s) who are nonetheless exposed to dramatically different probabilities of incarceration because they are slightly under or over threshold values of each variable. Both designs restrict comparisons to individuals who are eligible for the same types of sentences and who are therefore on the policy margin between prison and alternative sanctions. This project will provide new evidence for policy-makers and actors in the criminal justice system that could inform future decisions on the appropriateness of custodial vs. non-custodial sanctions.