theirview Reservations: gains at a cost
Affirmative action in higher education remains a divi sive issue in India. One cannot forget the wide spread debates and pro tests that accompanied the recommendations of the Mandal Commission. More recently, affirmative action has risen again to the forefront of national politics, as the Supreme Court ruled last week to uphold the legal quota of 27% for backward class students in state-run institutions of higher education, which is over and above the 22.5% reserved for scheduled castes (SC) and scheduled tribes (ST). In addition, the court’s decision clarified that the “creamy layer,” or the elite, would be exempted from the policy.
The debate surrounding this controversial topic has been hampered by a lack of rigorous evidence on the effectiveness of affirmative action programmes. Do these programmes successfully reach the poor, or is it the rich only taking advantage of the opportunities affirmative action provides? Does gaining access to higher universities as a result of these programmes improve one’s ability to find a higher paying job after graduation?
To provide further information on these questions, we collected data on individuals who had applied for entrance to engineering colleges considered among the most prestigious, between 2004 and 2006-approximately 8-10 years after their entrance exam held in 1996-in one Indian state.
We, then, conducted a survey with both the applicants and their parents to gauge life outcomes, including income and occupation, job satisfaction, social networks and caste identity-all of which help address three critical questions central to the debate:
1. Does caste-based targeting result in aid for backward classes and SC and ST or is it simply benefiting the “creamy layer” (richer members) of the groups?
2. Do those who gain admissions to university as a result of affirmative action actually benefit?
3. Does the policy harm those who did not gain admissions due to affirmative action?
First, we compare the background characteristics of those lower-caste individuals (members of the other backward caste, or OBC, and SC categories) who gained admission to universities due to the reservations, with those of the general caste who would have otherwise gained admission if the reservation policy was not in place. We find that the parental income of these upper-caste applicants is nearly double that of the lowercaste applicants. Thus, it appears that the affirmative action programmes are helping lower-income students gain admission to engineering universities. These results include members of the OBC group-a group that many think is too wealthy to benefit from reservations. While OBC applicants come from, on average, higher-income families than SC, both come from lower-income groups than the upper-caste students.
Do these students actually reap benefits from attending the engineering schools? One side of the debate argues that the programme places minorities in an academic environment for which they are unprepared and, thus, they are more likely to drop out of school and less likely to find a high-paying job after graduation. Others argue that those who are admitted due to reservations “catch up” once at university and, in fact, end up with jobs that are as good-if not better-than those from the general caste.
Using our data, we empirically estimated the returns to an engineering degree for members of OBC and SC groups. We find that attending engineering school increases the monthly income of these lower-caste individuals between Rs3,700 and Rs6,200, which corresponds to an increase of 40% to 70%. Getting an engineering degree, therefore, helps individuals from traditionally disadvantaged groups get a better job! But, there is one caveat to this result. Specifically, we find that, in the lower-caste group, it is those from higher socioeconomic backgrounds that appear to get the largest economic returns, suggesting that a policy expanding affirmative action to the very poor may not yield the same results.
The final question then becomes: What happens to those in the general caste who lost a seat due to affirmative action? Is there a social cost, i.e., are the losses incurred by the general caste larger than the gains of the traditionally disadvantaged groups? We find that there is indeed a social loss: Attending engineering colleges increases the monthly income of the upper-caste members by around Rs5,000 more than it increases the monthly income of the lower-caste members.
Thus, under affirmative action, it is plausible that upper-caste appli cants who were rejected due to the programme are worse off down the road, given that they may lack access to similar opportunities afforded to the less advantaged in order to maximize their potential. In addition, we find that the affirmative action policy appears to hurt female applicants, given that a higher percentage of those let in through affirmative action are males. This implies that while affirmative action helps some traditionally disadvantaged groups, it may do so at a cost to other disadvantaged groups.
Overall, affirmative action is a redistributive policy: It benefits those who enter university due to the reservations, but the economic losses incurred by the general caste are slightly larger than the gains for the disadvantaged groups. Assuming that people value having a more equal society, the relevant question is whether affirmative action can achieve such redistribution at a lower cost than other programmes or policies.
This paper can be accessed at www.nber.org/papers/w13926.pdf Sendhil Mullainathan is a professor of economics at Harvard University. Rema Hanna is a professor of public policy and economics at New York University. Marianne Bertrand is a professor of economics at the University of Chicago. Sudha Krishnan is a research analyst at the Policy Design Initiative, Harvard University.