Does Affirmative Action Work?

Marianne Bertrand, Rema Hanna, Sendhil Mullainathan present new evidence on the impact of affirmative action in university admissions – this time not from racial AA in the US, but caste AA in India. They find a neat equity-efficiency tradeoff, illustrating the political complexity of such programs.

Affirmative Action in Education: Evidence From Engineering College Admissions in India
by Marianne Bertrand, Rema Hanna, Sendhil Mullainathan
Many countries mandate affirmative action in university admissions for traditionally disadvantaged groups. Little is known about either the efficacy or costs of these programs. This paper examines affirmative action in engineering colleges in India for “lower-caste” groups. We find that it successfully targets the financially disadvantaged: the marginal upper-caste applicant comes from a more advantaged background than the marginal lower-caste applicant who displaces him. Despite much lower entrance exam scores, the marginal lower-caste entrant does benefit: we find a strong, positive economic return to admission. These findings contradict common arguments against affirmative action: that it is only relevant for richer lower-caste members, or that those who are admitted are too unprepared to benefit from the education. However, these benefits come at a cost. Our point estimates suggest that the marginal upper-caste entrant enjoys nearly twice the earnings level gain as the marginal lower-caste entrant. This finding illustrates the program’s redistributive nature: it benefits the poor, but costs resources in absolute terms. One reason for this lower level gain is that a smaller fraction of lower-caste admits end up employed in engineering or advanced technical jobs. Finally, we find no evidence that the marginal upper-caste applicant who is rejected due to the policy ends up with more negative attitudes towards lower castes or towards affirmative action programs. On the other hand, there is some weak evidence that the marginal lower-caste admits become stronger supporters of affirmative action programs.

One of my backburner projects is to look at the impact of affirmative action for Indigenous students at Australian universities. Watch this space.

About these ads
This entry was posted in Economics of Education. Bookmark the permalink.

7 Responses to Does Affirmative Action Work?

  1. christine says:

    “One reason for this lower level gain is that a smaller fraction of lower-caste admits end up employed in engineering or advanced technical jobs.”

    I should of course read the article before making this comment, but … if this is caused by discrimination in the labour market, it’s not a really a strong point against affirmative action.

    Also one of the bigger arguments for AA is the effects on subsequent generations, isn’t it? That might be a rather tough thing to identify.

  2. Kevin Cox says:

    Let me see if I have the logic of this study.

    We take two groups.
    Group A from a lower caste background had poor scores
    Group A displaced Group B from an upper caste background, with poor scores, but not as poor as the scores of Group A
    We measure the effect of the program by measuring the earnings of people in their first job.
    Group B had twice the earnings of Group A in their first job and hence it is said that it cost the economy in absolute terms.

    If this is the argument then it can easily be dismissed by the supporters of Group A. Christine has given one argument – the effect of discrimination carried on to the appointment of people in their first jobs so that the Group A were disadvantaged in obtaining jobs hence the lower scores. She also argues that the benefits to society are not just to the person themselves but to their wider family.

    Another argument against the measure of job is to take the economic gain as measured by taking the relative gain of Group A earnings against their income if they had not gone to the school, versus the relative gain of Group B. If the relative earnings are greater for Group A then it could be argued that there is a net gain to the community.

    I am sure others can think of many other arguments and measures that will be either for or against affirmative action.

    Certainly the monetary measure of first earnings is problematic. Monetary measures are at best a pale approximation of value. Perhaps the whole idea of measuring outcomes with monetary measures is largely a waste of time and effort. Maybe a different measure is needed or perhaps we need to revisit the money measure or perhaps we need to revisit the way we implement affirmative action or perhaps we can change the system so that affirmative action is not required.

    That is, instead of trying to show whether affirmative action is a good thing or not go about designing and building a system where people can compete for places irrespective of their financial background and we let the market place take care of the distribution of places and so remove the need for affirmative action.

  3. Patrick says:

    Kevin, although you are typically distracted (blinded…?) by money, that is a good idea. I also think it would be much simpler to have ‘blind’ admissions with no names, photos or interviews.

    Of course, that does mean removing a lot of information from assessors, but at University level I think it a reasonable sacrifice to make.

  4. gaddeswarup says:

    There are two recent papers on different aspects of reservations in India:
    http://www.uh.edu/~nprakash/
    One is about reservations in political representations and the other about reservations in jobs. Parts of the papers are too technical for me but they seem to complement the above paper.

  5. Kevin Cox says:

    Thanks Patrick,

    I was tempted to suggest how to implement the policy using special purpose money with extra information on it but I knew that would distract people:)

  6. christine says:

    “designing and building a system where people can compete for places irrespective of their financial background and we let the market place take care of the distribution of places and so remove the need for affirmative action.”

    Er … well, I think we’d all agree that removing the need (or perceived need) for affirmative action would be first best. I suspect the idea is that it might not be easily achievable?

    Also, aren’t virtually all Australian university admissions blind? (Except for medicine these days, I guess, though once upon a time back in the dim dark ages in Qld, even for medicine all you needed was a couple of science subjects at high school and a 990 TE score … )

  7. Patrick says:

    Very wise Kevin :)

    I am not sure Christine – I thought they had names. Maybe not.

Comments are closed.