What We Know About Teacher Tenure—Not Much

Teacher tenure is one of those contentious issues that everyone seems to have an opinion about. Recently, Stephen Blum outlined many of the “positives, negatives, problems and some suggestions for tenure.” But one thing that complicates any debate over tenure is that there is virtually no quantitative evidence of how tenure impacts teacher labor markets or student achievement. Sure, many people have stories about a teacher who should have been fired but was protected by tenure, just as others have stories about a teacher who angered her principal and would have feared for her job if she weren’t tenured. And whenever anyone proposes changing tenure in some way, such as lengthening the probationary period, we hear more stories about how on the one hand, bad teachers get tenure because the probationary period is too short for principals to really assess their effectiveness; or, on the other hand, good teachers don’t get tenure because the probationary period is too short for principals to really assess their effectiveness.

But what do we really know about the effects of tenure? From a research (as opposed to anecdotal) perspective, not much. This is partly because it’s next to impossible to identify the “effect” of a policy that has no control group; that is, all public schools have some sort of tenure so we can’t compare schools that have it to schools that don’t. However, there is some variation across states in the details of tenure policy, particularly the length of the probationary period. In a recent study, "Probation Length and Teacher Salaries: Does Waiting Pay Off?," Eric Brunner and I tried to use that variation to ask how the length of the probationary period affects teacher labor markets. Specifically, we focused on beginning teacher wages and relied on variation across districts that are located near state borders, where schools may be competing for teachers with schools that have different probation lengths. What we found is that districts in states with longer probationary periods pay higher salaries, all else equal, and that wage premium is much larger in collective bargaining districts.

It isn’t too hard to imagine why we see these results: if we think of job security as a benefit of being a teacher, then teachers will demand higher wages in exchange for lower benefits/increased uncertainty. But this pressure on wages will only occur when teachers have alternatives, which is why we see the effect in districts close to states with different policies.

What does this mean for California? California is one of eight states with a two-year probationary period; the majority of other states have at least three-year probations. If we were to increase the length of probation, at least some districts, particularly those near the border with Nevada (which also has a two-year probationary period), may need to raise salaries in order to continue to attract as many new teachers. On the other hand, there may be no wage impact in most districts. Our study also says nothing about whether a change in probation length would affect who is attracted into teaching, or how districts might react to such a change in terms of evaluation or training policies. My larger point is that our study is the only one I am aware of that even tries to quantify the impact of tenure policy on teacher labor markets, so no one can really predict well what would happen if we were change state policy. More research would certainly help inform the debate about this important issue.