Indian Colleges Find Success with Affirmative Action

When the divisive affirmative-action case of Fisher v. University of Texas returned to the U.S. Supreme Court in December for the second time in two years, it brought into acute focus a years-long data analysis conducted by Carnegie Mellon University professors Dennis Epple, Thomas Lord University Professor of Economics, and Lowell Taylor, H. John Heinz III Professor of Economics, along with Surendrakumar Bagde, HNZ ’11 and a member of the Indian Administrative Services.

“Affirmative action is bound to be controversial,” Epple said, “because it increases opportunities for disadvantaged students at the expense of advantaged students. But many have gone further, arguing that affirmative action makes everyone worse off by crowding out advantaged students and getting disadvantaged students into colleges where they can’t keep up. My coauthors and I set out to investigate whether this ‘mismatch hypothesis’ is correct.” Their findings are reported in “Does Affirmative Action Work? Caste, Gender, College Quality, and Academic Success in India,” which is published in the American Economic Review, the flagship journal of the American Economic Association.

They find that affirmative action works largely as intended in college education in India.

“In India, affirmative action gets more lower-caste students into college,” Epple said, referring to their study of 210 engineering colleges in India.

He and coauthors found that “affirmative action improves the achievement of the lower castes,” Epple said. “And the lower castes get more competitive majors, the more sought after ones such as computer science and electrical engineering. Affirmative action doesn’t impact the rate of on-time graduation. So, even though affirmative action gets disadvantaged students into more competitive colleges and more competitive majors, it doesn’t slow them down.”

Epple explained, “within each caste, one third of seats are reserved for women, and they too benefit via increased college attendance and achievement without impacting their rate of on-time graduation.”

India’s objective admissions system is starkly different from the layered, subjective approach by U.S. universities, each with its own criteria, essays, interviews, tests and so on. In the Indian colleges the authors study, students take the same standardized entrance examination, graded by the same organization. Each caste obtains a designated share of seats in each college. Applicants are then ranked by score within their caste and gender, and this determines their priority for college choice.

“The Indian system is appealing to study because it is possible to determine precisely the extent to which a student is a beneficiary of affirmative action, and standardized examinations taken by students in all of the colleges make it possible to measure impacts on achievement,” Epple said. “In the U.S., each college has its own admission policy, each determines the weight it chooses to place on affirmative action, and there are no standardized examinations to permit comparison of achievement across colleges.”

“Affirmative action doesn’t give rise to mismatch in the colleges we study,” Epple said, “but it is natural to ask whether our findings translate to the U.S. context. We don’t know, but we believe the parallels are sufficiently strong as to cast serious doubt on the belief that affirmative action causes mismatch.”

The Indian constitution mandated affirmative action for the groups formerly known as “untouchables.” Successful lobbying by other groups has greatly expanded the set of castes that are beneficiaries of affirmative action. Half of the seats in colleges in the study are set aside for affirmative-action quotas. Those eligible for affirmative action — male and female — receive a benefit, according to the data that the coauthors studied. But it doesn’t benefit everyone.

“It crowds out upper-caste men,” Epple said. “So that’s the cost.”


Dennis Epple, Thomas Lord University Professor of Economics