A Real Analysis of Merit Pay

by: Eric Luedtke

Tue Jul 08, 2008 at 09:08 PM EDT

One of Red Maryland's writers has managed to work himself into a lather over the Maryland State Teachers' Association's tentativeness concerning merit pay proposals. Not surprisingly, the limit of analysis consists of insisting that a few quick things (merit pay, ending teacher tenure, charter schools) will magically fix education, and calling teachers unions socialist. Because calling the other side names makes you right.

The touchstone for the posting is an article in the Sun that talks about increasing use of merit pay schemes in Maryland. And it's a good read, because, quite frankly, this is something that will see increasing play in education. Politicians like it, it sounds good to the public, and, despite the Red Maryland writer's gesticulations about MSTA, the majority of teachers, and most teachers unions, support some forms of merit pay. In a recent national survey, 80% of teachers said they would support merit pay for teachers who work in tough neighborhoods with low performing schools, and 64% for national board certified teachers (a form of merit pay that already exists in Maryland and has been supported by MSTA). And a close reading of the article, heck, any reading of the article, points out that the system of merit pay being implemented in Prince George's County was negotiated between the teachers union and the school system. In other words, teachers union leaders were at the table in creating the system.

You see, the question isn't whether merit pay is a good idea. Unions are all for getting their members more money, and merit pay is one way to do that. They just want such systems to be created with input from teachers, not just by legislative or administrative fiat, and they want to make sure any such systems are fair. So the real question is how to implement merit pay, and whether it is good public policy to implement it in certain ways. Here are just some of the problems that need to be dealt with:

1. As Dan Kaufman of MSTA points out, there is no research anywhere in the country that definitively proves merit pay significantly improves performance. One would hope that before the state or local school boards jumped whole hog into the idea, they would have some proof it would work. There needs to be better research on this, and the federal government should fund it.

2. What research is out there in terms of how teachers view merit pay suggests that the rewards have to be significant (approximately a minimum of $5,000) to influence teacher behavior. So we're talking about a huge budget item here. Meanwhile, conservatives who favor merit pay schemes complain about how much money is going to public schools, hoping to have their cake and eat it too.

3. Merit pay systems that have been implemented in jurisdictions in the past have subsequently been eliminated when the costs of the program expanded beyond the willingness of local officials to pay for it. The fact is that merit pay systems that reward teachers for meeting arbitrary targets are financially unpredictable. If Montgomery County implemented a merit pay reward of only $5,000, the difference between 20% of its teachers hitting the target and 80% hitting it would be about three million dollars. And there is no way of knowing when the budget is set how many would hit the target. Prince George's County is paying for their program with a federal grant, and I'd predict a one-in-four chance at best that they will pick up the tab when that grant runs out.

4. Measuring success of individual teachers with students is incredibly difficult. High stakes standardized tests measure a snapshot of a student and not improvement over time, which makes their use as a measure of professional success a little arbitrary. Some vital subjects and jobs are incredibly difficult to test - for example, the job of a guidance counselor who supports student success is almost impossible to measure. This last suggests that school-wide rewards would work better than individual ones, except that this would not create incentives for individual behavior and effort.

In other words, like most things involving education, creating a merit pay system that works will be a much more complex challenge than it seems at first glance. And, in any case, it's almost certainly impossible to close the achievement gap, raise student achievement, and meet the dozen other goals of public education with only this one tool. School reform needs to be a multi-faceted process, and the facets need to be carefully designed, rather than the 'throw a few good slogans out there and hope it makes a difference' strategy pursued by the Bush administrator over the last few years.