Merit Pay Cannot Work — Here Are Three Reasons Why

So Vivek is talking about bringing merit pay to Ohio, and this idea seems appealing to a lot of people, which reveals to me just how little people know about education and what we've already tried in this profession.

This has been studied for decades.

This has been tried for decades.

People have been trying to come up with merit pay schemes that work, for my entire life and they are never going to succeed.

We know now for a number of different reasons that merit pay will never work.

And the first reason it will never work is the measurement issue.

We simply cannot measure teacher performance the way people think we can.

We can get test scores, we can compare test scores over time, but those do not allow us to measure teacher performance.

And there was a huge research project funded by the Gates Foundation about 15 years ago called the Measures of Effective Teaching Project.

And just an enormous amount of money was spent to try to measure teacher effectiveness and turn it into, you know, basically a science.

And it just utterly failed.

There was absolutely no way to measure teacher performance.

They'd get one score one year.

would say that you know this is the best teacher ever and then the next year they would you know be in need of improvement like there was just no stability it was just all over the place and obviously the signals they were getting were were just noise it was it was just random fluctuation and i think we can understand that measurement failure if we understand that in the first place teachers have a very small influence on student test scores who your students are has a huge influence on your test scores.

And there are school level factors, there are community level factors, and of course, family level factors are by far the biggest.

But teachers don't actually make up that much of the outcome in terms of test scores.

And if the signal is, 10% and the noise is 90%, that is a very difficult signal to pick out.

And it has so far turned out to be impossible and not for lack of trying.

Do we think we haven't spent enough money on testing in this country?

Do we think we have not tried hard enough?

The last reason I want to talk about that merit pay will never work is that it creates perverse incentives, both within the school level where You know, you're incentivized to teach the kids who are going to get you the best pay raise, and you're not incentivized to teach the kids who need your help the most.

That's not great, but that also operates at the labor market level.

Teaching is a labor market.

People aren't forced to work where they work.

They have a choice.

And if you incentivize people with bonuses, one of the things you're going to do is incentivize them not to work in schools where they're less likely to get their bonus for whatever reason.

So I think we've really got to think about the practicalities, the details, but also the second order consequences of something like merit pay.

And if you're not in the education profession, here's some advice.

Don't just assume that everybody in the profession is dumb and hasn't thought of the idea that you had five minutes ago.

It's probably a pretty safe assumption that your idea that you think is so smart has actually been thought of and tested and tried a long time ago.

And if it's not currently being practiced, it's probably because it doesn't work.

Let me know what you think.