February 6, 2008 — Green for Grades

Virginia and Maryland schools experiment with financial incentives for academic achievement
By Ernie Siciliano

This spring, schools in two Fifth District states will use greenbacks instead of gold stars as a motivational tool. Baltimore students who improve their scores on Maryland's standardized tests will be able to pocket up to $110. In Virginia, as many as 15 high schools will pay students if they receive a three or higher (out of five) on their Advanced Placement (AP) exams.

Paying students for academic achievement is gaining traction across the country. The Virginia program was based on a pilot project in Texas that has expanded to seven other states. The most high-profile of the "pay-for-performance" programs is Opportunity NYC, established by Harvard economist Roland Fryer last September with an enrollment of 9,000 fourth- and seventh-grade students.

Empirical studies of these programs are rare, but generally show signs of promise. That appears to be enough for educators constantly trying to better the nation's schools.

"A lot of traditional formulas for improving education haven't really worked," says Edward Miguel, an education economist at University of California-Berkeley. "There's been a lot of research of classroom size and teacher size but less on the side of motivating students."

Paul Nichols heads a Virginia-based nonprofit organization that seeks to increase the number of high school students in AP classes. He says many students aren't motivated to do the extra, college-level work required to take an AP class and do well on the exam. They are more likely to focus on after-school jobs or activities.

Paying students to take AP classes may direct more of their attention to the classroom by offering immediate compensation. While academic achievement is strongly correlated with financial success, the gains often come much later.

"Children are quite myopic and you might be able to use incentives to make them work harder in the short run," explains Eric Bettinger, an education economist at Case Western Reserve University who conducted a pay-for-performance experiment in Ohio. "If children knew all the long-term benefits of college, we wouldn't need a program like this."

For the most part, pay-for-performance programs are being funded by the private sector. Virginia's program is supported by grants from Exxon Mobil and money raised from local businesses that could total more than $13 million. The Baltimore program is an exception as it is funded from the city's regular school budget. "In our capitalist society, pay for performance certainly does strike at the heart of what goes on privately," Nichols says.

Does that mean what works in the private sector also works in public schools? According to Nichols, research on the Texas program found that students were more likely to take the AP exam and perform well when they were paid. Enrollment in after-school AP preparation programs also rose.

Stanford's Edward Miguel and two other economists conducted a study in Kenya, where female students earned academic scholarships and cash grants if they scored in the top 15 percent in a test. Miguel found that financial incentives boosted test scores, as well as contributed to a 25 percent drop in teacher absenteeism. "Even though the incentive was targeted to the students, it affected the whole classroom," he notes.

Case Western's Bettinger conducted his own experiment in Coshocton, Ohio. Students received "Coshocton Bucks," gift certificates redeemable at area businesses, based on their performance on state standardized exams. Bettinger found that their math scores significantly improved when they were paid, although improvement was less noticeable in other subjects. (His research results are in the process of being published.)

Despite these early successes, paying students for performance may have some drawbacks. For example, students could focus too much on the extrinsic reward of getting money and less on the intrinsic rewards of learning and acquiring new skills. When extrinsic motivation overtakes intrinsic motivation, total motivation decreases. Sociologists call this phenomenon "crowding out."

Both Miguel and Bettinger surveyed students to measure their attitudes toward learning. Neither economist found much evidence of crowding out.

Still, the research on pay-for-performance programs is relatively new and there are questions to resolve. A 2007 study in Israel found that performance rose for boys but not for girls, a difference that Bettinger did not notice. Another study in Canada found paying students was effective only when coupled with an increase in academic services.

Even pay-for-performance advocates are reluctant to label it a panacea. "The incentives help some, but I don't think that's the entire puzzle," Nichols says.

"I'm a little guarded as to give it a full thumbs-up," Bettinger agrees. Still, the Coshocton school district requested that he run his experiment another year. He is optimistic that he will continue to have success.

Ernie Siciliano is a Publications intern in the Richmond Fed's Research Department.