Affirmative action in the US: does it work?

Daniel Sabbagh, senior researcher at Sciences Po's Centre for International Studies and Research (CERI), has been working on the issue of race in the United States for over twenty years. Today he sheds light on the thorny question of affirmative action in North American universities. What criteria do these policies employ? Are they fair and effective? The answers provide a nuanced picture of this type of programme.

Sciences Po: The affirmative action policies encountered in American universities are more varied and complex than we often imagine. Could you describe them briefly?

Daniel Sabbagh: All affirmative action policies share a common goal: to rectify the under-representation of certain groups at university relative to their proportion in the general population. Affirmative action is based on the premise that these groups – blacks, Hispanics, Native Americans and sometimes Asians – are under-represented, at least in part, due to past or present discrimination.
Besides this common goal, such measures can be distinguished from each other by their methods and classified into three categories: outreach schemes, which aim to widen the pool of candidates from the groups concerned; what we might call “direct” affirmative action, in that it openly confers an advantage on its targets; and finally “indirect” affirmative action, which is more recent. The latter is based on supposedly non-racial or colour-blind criteria, such as place of residence or level of income, but which in fact are highly correlated with belonging to a given racial group.

SP: Has “direct” affirmative action had significant results?

D.S: In large part, yes. Overall, the affirmative action implemented since the 1960s has proven effective in terms of its immediate objective. It has been highly instrumental in raising the proportion of blacks in the student population from two percent to nine percent between 1960 and 1994 (blacks represented 13 percent of the US population in 2010). But of course other factors also came into play, particularly the adoption in 1964 of the Civil Rights Act against discrimination, which effectively increased the returns on blacks’ investment in education. However, we do know that if direct affirmative action was done away with altogether, blacks would only make up two percent of students in the most prestigious universities.*

SP: Despite their proven impact, direct forms of affirmative action are coming under increasing criticism. What are their drawbacks?

D.S: In many cases, these drawbacks are posited without being proven. For example, there is no evidence that minority students make less effort because they expect to be given a leg up. There is also no indication that this form of affirmative action undermines the image that the beneficiary groups have of themselves. But the debate is still open. Currently, the main controversy relates to something called mismatch.

SP: What is mismatch?

D.S: Mismatch is what supposedly results from the fact that, due to affirmative action, black and Hispanic applicants are more likely than white and Asian applicants with equal SAT** scores to gain admission to a more prestigious institution. However, the high standards in such universities may not be suited to their initial level, causing their academic performance to suffer. When they are admitted to a top-class university through affirmative action, black and Hispanic students are ranked lower there than they would be in a second-class university. But this is not enough to deter them from accepting offers of admission, which is understandable; students admitted to Harvard have a better chance of having their tuition fees covered and of getting a high salary after graduation. Moreover, this adverse effect is reproduced at lower levels of the system. By ripple effect, the less prestigious universities in turn lower their standards for black and Hispanic applicants, without which they would not be able to admit enough students from these groups.

SP: Has highlighting this adverse effect led to affirmative action being discontinued?

D.S: No, so far there have been practically no instances of direct forms of affirmative action being simply abolished. Instead, the dominant trend is to gradually replace them with more or less equivalent indirect measures.

SP: Such as?

D.S: The main ones are percentage programmes, where the top-ranking students from each of the high schools in a state** are given automatic access to a public university in that state, regardless of their SAT scores.

SP: Do these programmes have the desired outcome?

D.S: Only partially. First, in some high schools – that is, “mixed” ones – blacks and Hispanics are often relegated to the less advanced classes. As such they are less likely to come out on top. For example, at the University of Texas at Austin, in a state where this scheme has been implemented, the percentage of blacks admitted to first year only increased from three percent to just over four percent between 1997 and 2012. However, the argument of those opposed to this type of indirect affirmative action, which maintains that white parents would be tempted to enrol their children in “bad schools” to increase their chances of being ranked at the top, has proven inaccurate. The study conditions are so deplorable in these schools that it would make for a dangerous bet indeed.

SP: Have any other alternatives proven more effective?

D.S: The first point to stress is that no alternative can be effective on its own. Only a combination of alternative criteria would be able to produce the desired outcome. For instance, it has been shown that simply replacing the racial criterion by that of parental income is useless. The majority of blacks who have the initial level necessary for admission to a prestigious university are not poor, most poor people are not black, and among the poor, the academic performance of blacks and Hispanics remains lower than that of whites and Asians.

SP: Beyond the direct effects of affirmative action on access to education, do we have an idea of its impact on society?

D.S: We know that in some cases, “direct” affirmative action has produced positive results over and above its immediate objective. For example, it has been found that black and Hispanic doctors tend to settle in areas populated by blacks and Hispanics. We know that this trend is more pronounced the higher their level of qualification. This results in an improvement in medical coverage for black and Hispanic populations and in the care they receive.

SP: So we haven't heard the last of affirmative action in the United States?

D.S: Certainly not. Its reason for existing has not gone away, and it has been the subject of many studies. In the 1970s, research into affirmative action was mainly of a legal or philosophical bent. These days, we can base our reflection on empirical knowledge about the effects of these programmes. The effects are often positive, even if they do not eliminate the gap in performance between members of different racial groups and may even exacerbate it in some cases.

SP: Are there lessons to be learned from this experience to improve equality in education in France?

D.S: The comparison is made difficult by the fact that we know much less about the effects of policies implemented in France. Our culture of evaluation is still weak. Evaluation is also hampered by the absence of what are incorrectly called “ethnic statistics”. This makes it impossible to comprehensively assess the consequences of indirect affirmative action policies, which nonetheless exist. The Americans do not have this problem, and are therefore able to compare the effectiveness of different forms of affirmative action. In France, the lack of legitimate alternatives means that indirect affirmative action is implemented by default, and relatively little is known about its results.

*Contrary to popular belief, about 70 percent of American universities are open to everyone. Less than a third of them are selective.
** In the US, educational policy – especially secondary education – is a state responsibility. The federal government is supposed to play a residual role.

More articles by Daniel Sabbagh
- “Affirmative Action: The US Experience in Comparative Perspective”, Daedalus. Journal of the American Academy of Arts and Sciences, 2011
- “The Rise of Indirect Affirmative Action: Converging Strategies for Promoting “Diversity” in Selective Institutions of Higher Education in the United States and France”, World Politics, 2011

Learn more about social responsibility at Sciences Po