'Smell Something, Say Something!' Teachers' Unions Do Not Hurt Student Outcomes.
Welcome to the first edition of a new On The Economy feature, dedicated to the parting admonition of the great Jon Stewart: when it comes to BS, "smell something, say something!"
To be clear, I'm not trying to emulate the fact checkers out there. Nor am I going to peruse the papers, like Dean does so effectively, to find errant economics reporting. Instead, I'm just going to occasionally pounce on a specific brand of assertion: a stylized, accepted fact that isn't a fact at all.
For example, conservative partisans (as well as many centrist Democrats) consistently assert that teachers' unions are bad for student outcomes, and if we want to improve such outcomes, we must diminish the impact of teachers' unions. Most recently, this negative role of unions was a featured assertion in a Republican primary debate.
That claim smelled bad to me, as in I know of no body of evidence to support it. I know it's a constant refrain, but I figured I'd have seen something from the deep academic community that runs analyses of such issues over the years to support it, and I haven't.
Maybe I missed it. So I asked some experts in this field and they confirmed my intuition.
-- Berkeley econ prof Jesse Rothstein, who's done important work on "value-added-measurement" in teacher evaluations, confirmed my priors that such evidence is wanting.
-- He and education policy expert Kevin Carey made the same interesting point: there's a significant measurement challenge in that school districts that don't have unions, and would thus serve as a useful control, "tend to have teachers associations and/or contracts that aren't too different from what unionized districts have" (Rothstein).
-- The unions themselves will correctly tell you that states with fewer unions, including "right-to-work" states, have worse student outcomes. And there are countries, like Finland, that have very high unionization rates and consistently rank highly in international comparisons of student outcomes. But, as Carey stressed, right-to-work states are also poorer, and Finland ain't the US, and there's the quasi-union arrangements noted above, even in non-union states. So it's very hard to make an all-else-equal run at this question.
-- Larry Mishel shares this paper by himself and Emma Garcia. It tests -- rigorously, I thought -- for correlations -- again, we're not talking causality -- between the strength of teachers unions and whether unions shift more experienced and higher credentialed teachers away from poorer schools. Their results fail "to show an association between the strength of unions in the states and the allocation of teacher credentials across schools. We find no negative or no association between the allocations of credentials in average schools or in high poverty schools and the unions' strength...we find no association between the unions' strength and the misallocation of credentials among high poverty schools relative to the average."
In other words, there is nothing like a well-established consensus that teachers' unions have any impact one way or the other on student outcomes. That doesn't mean teachers' unions are great for kids either. It means that when you hear a politician bashing teachers' unions on behalf of students, they're BS'ing... so: smell something and say something.
***
This has been reposted from Jared Bernstein's On the Economy blog.
