Michigan school districts are required by law to make performance a "significant factor in determining compensation" for teachers.

This is uncharted territory for nearly all schools, since historically they've paid teachers based only on years on the job and academic credentials. Some public schools in India, on the other hand, have paid teachers based on performance for years, and a new study finds large and positive impacts for Indian elementary school students.

Specifically, students taught by teachers eligible for performance-based pay for five consecutive years scored 20 percent higher in math and 14 percent better in language arts. The researchers estimate that this policy is 15 to 20 times more effective than a 25 percent reduction in class sizes.

Not all merit-pay programs are the same, and design matters. Here are some specific details about India's program that Michigan school districts might consider when designing their plans:

•  Teachers were paid bonuses for each percentage point increase of their students' test scores, and there was no limit to how much teachers could make.
• The average amount teachers received was about 3 percent of an average teacher salary. In Michigan, that would equate to about $1,800 or more.
• Rewards that had the most impact were based on individual performance, not group performance.
• Teachers were paid bonuses only for individual growth in math and language arts.
• Students who did not progress could negatively impact a teacher's bonus, but to a limited extent.

While other studies of some merit-pay programs in the United States have shown mixed results, this study from India is relevant and instructive for three reasons.

First, it uses randomized control trials — the gold standard in educational research. Second, the study analyzes student performance over a five-year period — the longest for any merit-pay program ever studied, according the study's authors. Lastly, it tested and demonstrated that the results were not due to "gaming" or "teaching to the test" — students taught by teachers paid based on performance did better than their peers in all subjects and on all types of test questions (multiple choice and short answer, for example).

Obviously, there might be many differences between Michigan's public schools and those operated in India. But what happens inside schools — teaching and learning — is probably more similar than it is different.

India's merit-pay program could provide important lessons for Michigan's schools.

Permission to reprint this blog post in whole or in part is hereby granted, provided that the author (or authors) and the Mackinac Center for Public Policy are properly cited.