Association for Public Policy Analysis & Management

Since 2009, states across the country have implemented substantial changes to teacher labor markets. Most widespread have been new systems of teacher evaluation based on student outcomes, but many reforms also include new restrictions on or even the elimination of tenure, reductions on collective bargaining protections, and new limits on union membership. Although policymakers and advocates of these reforms have argued that such changes will promote a more effective teacher workforce, critics have worried that they may drive effective teachers from the classroom and dis-incentivize potential new, high-quality teachers from taking their place. Thus far, however, rigorous evidence for the effects of labor market change on either current or potential teachers remains scarce.

In this paper we consider the effects of reform in Michigan, focusing in particular on changes to the pipeline of potential teachers. Beginning in 2011, the state implemented new teacher evaluation requirements, lengthened the time to teacher tenure and linked tenure to on-the-job performance, and reduced the scope of collective bargaining to prohibit evaluation, dismissal and other provisions from teacher contracts. Later, the state also passed a new Right-to-Work law that prohibited districts from requiring public school teachers to pay union dues. Because Michigan has also realized increases in teacher attrition over the same time period, teacher advocates have claimed that the state’s reforms have harmed the profession and, by extension, children themselves.

 Using rich administrative data on all Michigan high schoolers in the National Student Clearinghouse between 2007 and 2015, we investigate patterns of entry into the teaching profession by first providing descriptive estimates of the post-reform change in new education majors in Michigan-based teacher preparation institutions. Because more than 90 percent of public school teachers in Michigan are trained in an in-state institution, we are able to provide a comprehensive overview of the pool of future teachers. Next, to obtain plausibly causal estimates of the reform impact on potential teacher quality, we compare pre- and post-reform high school test scores of students who choose an education major or minor to those of students entering a comparison group of programs including social work, nursing and accounting that should not have been impacted by Michigan’s teacher-related reforms. We estimate overall impacts as well as important sub-group changes associated with student demographics and local economic conditions. As such, our results provide a new and important systematic consideration of policy question that has thus far been informed largely by conjecture and debate.