We examine the effect of Michigan's 2011 reforms to teacher evaluation and tenure policies on teacher retention. Our data are drawn from administrative records containing the population of public school employees from 2005-2006 through 2014-2015. To identify the causal effects of these reforms on teacher attrition, we utilize a difference-in-differences (DD) strategy that compares the exit rates of teachers with the exit rates of other professional staff in the same school districts who were not affected by the policy changes. We find that, on average, Michigan's teacher reforms had little impact on teacher attrition overall. However, further analyses provide strong evidence that early-career teachers assigned to hard-to-staff districts were more likely to exit post-reform.