Recent Research: The Merit of Merit Pay Programs

Published April 1, 2007

University of Arkansas education researchers Jay Greene and Marcus Winters touched a nerve when their study on teacher pay hit newsstands in February. Critics took aim, questioning the accuracy of Greene’s findings that teachers make an average of $34 an hour.

Yet most people failed to take stock of the larger point the authors were making: The real debate shouldn’t be over how much teachers are paid, they argued–it’s the way they’re paid that needs reform.

Enter merit pay.

Simple Idea

Merit pay programs are slowly inching toward the mainstream, a fact even the National Education Association (NEA) teachers union now concedes. Still, teachers unions continue to say merit pay is, in the words of former NEA president Bob Chase, “smart but it would be a mistake if a lot of schools tried to replicate it,” according to a 1999 Education Intelligence Agency memo.

Trying to halt this momentum suggests merit pay is some radical new approach, when in fact differentiated pay was common as early as the 1920s. Yet as even a cursory glance at their history shows, merit pay programs systematically have been halted before their benefits or detriments could be demonstrated.

Thus, we still don’t know that merit pay works–but as its proponents rightly note, sticking with the status quo continues to produce flat-lined student test scores.

Early Returns

Over the past few years, merit pay plans have been adopted in Charlotte, Dallas, Denver, and Minneapolis, to name just a few cities. The programs differ in terms of the bonuses offered and the criteria for gaining them. As a result, measuring the effectiveness of merit pay programs is like shooting at a moving target.

For instance, critics of Houston’s new merit pay plan say its bonuses are simply too small to encourage teachers to improve their teaching practices. Moreover, figuring out how music teachers, art teachers, and even physical education teachers can reap the benefits of a merit pay system is crucial if such programs are ever to reach a critical mass of support.

But does all this mean merit pay can’t work?

Differences in Details

As with many education reforms, the devil is in the details. While academic research on the topic has been surprisingly sparse, some well-designed attempts, such as David Figlio and Lawrence Kenny’s large-scale 2006 study, have found individual teacher incentives are positively associated with student gains.

Earlier this year, a pilot program in Arkansas’s Little Rock School District provided researchers an opportunity to discover the effects of merit pay.

In the Little Rock program, teachers could earn a bonus worth up to $11,000, and participating schools received more than $200,000 in total performance bonuses for the 2005-06 school year. According to a research team led by Gary Ritter and Joshua Barnett at the University of Arkansas’s Department of Education Reform, those bonuses led to significantly greater learning gains than had been achieved by the same students prior to the program, as well as by students at comparable schools.

Students in schools where the program operated in 2005-06 showed an improvement of 3.5 Normal Curve Equivalent points–a gain of nearly 7 percentile points for the average student.

Research Prospects

While the results from this first-year study suggest the future of merit pay is promising, the second-year study will add five more schools to the program–a move that will greatly assist in expanding and explaining the first-year findings.

This has some merit pay proponents saying it could be the missing piece for which education reformers have been searching.

Not so fast. While the early signs are encouraging, we won’t know if merit pay really works until various pilot programs are allowed to run unimpeded for several years. Detractors have managed through politically sensitive school boards to dismantle several promising plans.

Nevertheless, since almost no one is satisfied with the way teachers are paid, critics unhappy with current merit pay programs would do well to contribute to the debate by proposing viable alternatives.

Merit pay may not be the cure-all to education’s ills, but trying it and measuring its impact can’t be any worse than simply staying the course.

Brian Kisida ([email protected]) is a research associate for the School Choice Demonstration Project, and Brent E. Riffel ([email protected]) is the deputy director of the Office for Education Policy, both at the University of Arkansas Department of Education Reform.

For more information …

Education Intelligence Agency, September 20, 1999,

“Individual Teacher Incentives and Student Performance,” by David Figlio and Lawrence Kenney, National Bureau of Economic Research, Working Paper No. 12627, October 2006,

“Evaluation of Year One of the Achievement Challenge Pilot Project in the Little Rock Public School District” by Joshua H. Barnett, Gary W. Ritter, Marcus A. Winters, and Jay P. Greene, published in January 2007 by the University of Arkansas, is available through PolicyBot™, The Heartland Institute’s free online research database. Point your Web browser to and search for document #20772.