How Changes to Compensation and Union Power Affect the Composition of the Teaching Workforce: Evidence from Wisconsin's Act 10

Abstract:  How does the composition of the teaching workforce respond to policy changes that affect the compensation and attractiveness of being a teacher? I address this question in the context of Wisconsin's Act 10, a policy change which weakened teachers' unions, capped wage growth, and increased required pension contributions by teachers. I first ask what effect Act 10 had on the retention of teachers in Wisconsin. I find that older teachers are quite responsive to the policy change: in the year following the reform, turnover rates doubled from 17 to 35 percent for teachers of retirement-eligible age, and this increase was similar across observable teacher and school characteristics. By contrast, I find little change in the attrition rate of younger teachers. I then analyze the relationship between the exit of older teachers and changes in education quality using school by grade-level value-added estimates. Surprisingly, I find that value-added increased substantially in grade-levels with teachers who retired following the reform, and I obtain similar results when instrumenting for retirement using the pre-existing age distribution of teachers. I consider two channels that could have potentially driven these results: first, retiring teachers tended to have lower value-added than their replacements; second, certain schools (or school districts) enacted new policies following the reform that simultaneously raised test scores and induced teacher retirements. I find some evidence suggesting that the second channel played at least some role, although the evidence is consistent with both channels having been important. Lastly, I discuss ongoing work to explore particular changes at the school and district level that may have contributed to these results. 

Preliminary Draft