You can disprove any position if you force your imaginary opponents to take the maximal side. So if you say of teacher's unions "smashing them will not magically raise test scores", all I can say is, "Well, d'uh". And while I understand that teachers also lobby for things that are good for kids, like better supplies, this does not make powerful teacher's unions a good idea. Teacher's lobby for kids when it happens to coincide with their interest. Unfortunately, in urban areas, it often doesn't.

I should probably clarify that I'm talking about twenty, maybe thirty failing urban school districts/agglomerations in the United States. I could care less whether Scarsdale has a powerful teacher's union that negotiates triannual ten month paid leave in Hawaii. And the problem in rural areas is not the teacher's unions, it's the geographic fact of no possible competition, and often the net outmigration of educated people who might make good teachers.

But in those urban areas, the teacher's unions are a big honking problem. This is not some crazy right wing opinion about unions in general; it is a specific problem with public employee unions. The cops and firefighters have their own issues, about which I will happily wax lyrical some other day, but in the end most of them boil down to getting paid ridiculous amounts of money to do no work. If the laziest ten percent of New York's teachers spent all day drinking coffee and doing "literature review", this would be a fiscal problem, but not a desperate one. The problem is, we stick the teacher's union's problems in our classrooms.

But getting rid of the teacher's unions would not lead to some happy paradise where all the students were Doogie Howser. The teacher's unions are one cog in an enormous dysfunctional system. The school boards, the education bureaucracy, the principals, the other "political stakeholders"--precious few of them poor parents--also factor in. If you got rid of the teacher's unions and left the rest of the institutions in place, I would be shocked if the schools noticeably improved.

But while taking away much of the teacher's union's power is definitely not sufficient, it does seem to be necessary. They resist changes to their work practices that the best evidence (see Ayers, Supercrunchers) seems to show works with disadvantaged kids: rote memorization, and phonics. These replace the tools that upper middle class give their kids earlier--even if you went to a whole language school, if you're reading this blog it's a safe bet you had phonics, too, when your parents taught you to "sound it out".

Instead, they agitate for things like smaller class sizes. It is true that schools with smaller class sizes tend to do better--but this is not surprising, since they tend to be more affluent. Pilot programs with disadvantaged kids also seem to show a benefit, but these suffer from the same problem that I discussed in a previous post about the Perry Pre-School: who's staffing your smaller class sizes? If smaller class sizes means employing more marginal teachers, it's far from obvious that this is a net boon. To the kids, I mean. It's an obvious win for the union.

This is why almost all educational ideas fail: they don't scale when you take the highly motivated grad students and gifted teachers out of the equation. That's why I'm tepidly gung ho about Direct Instruction: it has been proven to work with ordinary teachers using ordinary resources.

I don't care if the teachers have unions to negotiate over salary and benefits. But I think the power to block terminations and set work rules should be entirely stripped from them.

But this will not do anything unless you also take on the principal's union, prune the rapidly multiplying deadwood in the educational bureaucracy, get someone who knows their way around a regression analysis to pick your curriculum, and get serious about accountability for the schools. I don't want to defang the teacher's unions for the fun of it; unless you're planning to do these other things, you might as well leave it alone, too.