What is the point of merit pay for teachers? Isn’t that the first question we should be asking? And then, given that answer, do we think that it will work? Let’s take a look!

(Let me be clear here: for the sake of this discussion, I am talking about some sort of compensation structure by which individual teachers get bonuses and/or raises based on their own students’ performance on some sort of test or measure.)

One possible reason to adopt merit pay is that it seems more just, in a way. That is, it would be only be fair if better teachers should be paid more money. I mean, who wouldn’t think that better teachers — even if by “better teachers” we mean “teachers who get better results” — deserve more money? We could argue about which outcomes and measures are meaningful, but the basic idea has a lot of appeal.

There is another big potential reason to adopt merit pay — a far more compelling possibility. If a merit pay program would somehow lead to improved outcomes, who would oppose such a thing? But I wonder if that result is even plausible, given the realities and details that so many people want to ignore. So, let’s think about what the theory of action (i.e. how/why such programs lead to the desired outcomes) might be. How could a merit pay program lead to improved outcomes? I have three big questions.

1) My first question about any merit pay program would have to be, “Where is the money going to come from?” Obviously, that is related to “How big a program are we talking about?”

There are lots of reason to building large program, covering many teachers and with large bonuses (see below). However, this runs up against a problem that is not seen in most other industries. You see, better teaching – even better outcomes – does not produce the revenue to pay for these bonuses. A better salesperson will generate greater revenue, revenue than can be used to pay for his/her bonus. A more efficient repairperson can cover more repairs, saving the employer on headcount, thus economically justifying a bonus. But what about a better teacher?

The public schools already have enormous market share. More importantly, additional students drawn from private schools do not generate additional revenue. Additional student drawn from charter schools generate less than average revenue. So, where would the money come from?

Washington, D.C., appears to be on the brink of starting a large new merit pay program. However, it is being paid for with private funds. What happens when those grants run out? New York City’s public school system is 25 times the size of D.C.’s. Are there private grants that would pay for an equivalent program in NYC? This is simply not a scalable solution, neither as a long term solution nor as a broad solution for many districts.

2) My next question is, “How much money will it take to alter the behavior of existing teachers?”

That is one of the basic principles of merit pay programs, right? That is, by providing monetary incentives, teachers will work differently than they would otherwise. Well, would what it take?

We already know that teachers are not particularly sensitive to monetary incentive. After all, they chose to go into a famously low-paying profession, one without enormous financial upside. (Yes, given decades of experience and very advanced degrees, they can make over $100,000 in some districts, but that amount of experience and education yields far greater compensation in other fields.) There are decades of research to back this up — research that shows that teachers are not particularly motivated by financial rewards. We should, therefore, expect them to need even larger bonuses to alter behavior in education than in other fields.

But we are still missing a theory of action. There is a still a black box or underwear gnomes scenario (i.e. a situation in which a key step that is never explored or explained). If teachers do not know what to do to be better, how is paying them bonuses going to help solve that problem? If, on the other hand, teachers know but are too lazy to do what it takes, how is offering these already lazy and monetarily insensitive folks bonuses going to change that? How is the possibility of bonuses going to alter behavior?

I think that this line of thinking contains the most common mistake people who support merit pay make. They confuse how they might respond to such a structure with how others might respond. People in finance are accustomed to a large bonus component of their compensation, and they believe that this system works. However, the people who have gone into that industry already knew that going in. This is a system that works for them. But teachers did not go into such a system, and clearly were not looking for even the possibility of massive compensation. Why expect them to respond like people in finance might?

I am not suggesting that there is not some level of bonus that would alter behavior. Rather, I am asking what it is, or what the rational basis for determining it might be. Is there a research basis for establishing the size of these bonuses? Is there even a logic behind the setting of merit pay bonuses for teachers beyond “This is how much money we have raised”?

3) My last big question is, “How much money would it take to significantly alter the composition of the teacher pool?”

There are many who believe that we have the wrong people in teaching — that we need harder workers or smarter people, or whatever. Many of them are merit pay proponents, and they see such programs as attracting more of the right people to the classroom.

An argument that I have heard quite a few times suggests that some people choose not go into teaching because they resent the idea of not being rewarded for their excellence. This argument suggests that what is important to them is being recognized as being better than their colleagues. I suppose that I have some sympathy for this argument. After all, who does not want their accomplishments recognized? But that is not the same thing as insisting that one’s recognition come in the form of a check! Is anyone seriously suggesting that making $100 bonuses available to the best teachers will significantly alter the pool of teachers? That is simply a laughable suggestion, right? If recognition of excellence is important, that can be addressed in ceremonies, announcements and/or public bulletins, and done far less expensively than merit pay programs while offering far greater inventive than small checks.

I understand that there are many people who never consider going into teaching, or who rule it out, because of the level of compensation. But small or even moderate bonuses are not going to change that. I understand that people want to be rewarded for their hard work and accomplishments, and some people expect more money for that than others. So, again, how large would the bonuses have to be, and how many would have to be available, to alter the career decisions of a significant number of people?

For example, if a teacher could potentially double his/her first year salary, but only one one out of a thousand teachers would get the bonus, would that possibility attract a lot of new applicants?

4) My last question is really a composite of the first three. “If the money available for such a program is neither enough to alter behaviors nor attract new people to teaching, isn’t merit pay just a way to reward existing high performers? And how does that actually improve our schools?”