T C A Srinivasa-Raghavan: The way reservation works out

OKONOMOS

At long last, an empirical study of the costs and benefits of reservation in engineering colleges.

Affirmative action everywhere is controversial. But how controversial it is depends on how politicians design the policies.

India is no exception to this general rule. For example, as long as reservations (22.5 per cent of government jobs) were meant only for the dalits and the tribals, no one objected because it was evident that they needed assistance to become members, as Andre Beteille has pointed out, of the middle class.

But as many commentators and experts have shown "" most recently Dipankar Gupta of JNU in an article in this newspaper on April 13 "" problems arise when politicians begin to use reservations not only to advance their own personal or extended intra-caste group causes, but also do it by adopting policies that have a zero-sum game flavour.

This seems to be the key message, an unintended one though, in a recent paper* by Marianne Bertrand, Rema Hanna and Sendhil Mullainathan who have done an empirical analysis of the effect of reservations in engineering colleges. They find that these policies work quite well and also as intended.

What I found of special relevance was their statement that "the marginal upper-caste applicant comes from a more advantaged background than the marginal lower-caste applicant who displaces him." The term marginal is critical here, I think.

If an upper caste applicant with 90 per cent is displaced by a lower caste applicant with 80 per cent, it does seem fine, provided of course that his father or mother is not already a member of the middle class "" lower, middle or upper. It never made any sense for the offspring of a dalit or tribal foreign service officer to enjoy the benefits of reservation.

The authors also say it is a myth that very low-scoring lower caste students don't benefit from the education they receive as a result of reservation. Thus, "despite much lower entrance exam scores...we find a strong, positive economic return to admission. These findings contradict... (the view) that those who are admitted are too unprepared to benefit from the education."

The authors have also found that "the marginal upper-caste entrant enjoys nearly twice the earnings level gain as the marginal lower-caste entrant." This happens because very few of the lower caste persons get absorbed in higher-paying jobs. "In this particular context," say the authors, "we find that for Re 1 lost to the upper caste, the lower caste members gain Re 0.4 to 0.6." This is a devastating finding and can be checked for accuracy simply by looking at your own peer group in the office.

Another is that "the individuals who are displaced by the program come from stronger socio-economic backgrounds than the displacers. Hence, by targeting disadvantaged caste groups, the policy achieves some income-targeting without generating any of the behavioural distortions typically associated with income-targeting."

They go on to say that "our findings do not support the view that the academic resources that are devoted to the lower-caste students are totally wasted on them." Well, hold on, whoever said that? When? Where?

The study has also found no evidence that "the marginal upper-caste applicant who is rejected ends up with more negative attitudes towards lower castes or towards affirmative action programs." Well, maybe, but perhaps the authors should have asked the parents what they thought or think.

Few know it "" and here is something for future study "" it is the parents who get hurt more, not just because of their prejudices but mainly because as far as sacrifices go, the ones made by the upper-caste parents are not any the less than anyone else.

In the end it boils down to the old problem: social justice to groups vs the abridgement of equity for individuals. But now that the chief justice has said that, "if any Constitutional Amendment is made which moderately abridges or alters the equality principle under Article 19(i)(g) it cannot be said that it violates the basic structure of Constitution," the issue has been settled for the moment, at least for practical purposes. But the philosophical implications of zero-summing remain.

*Affirmative Action in Education: Evidence From Engineering College Admissions in India, NBER Working Paper No. 13926, April 2008