Rethinking Affirmative Action

Affirmative action has been defined as "voluntary and mandatory efforts undertaken by federal, state, and local governments, private employers and schools to combat discrimination, foster fair hiring and advancement of qualified individuals regardless of their race, ethnicity and gender; and to promote equal opportunity in education and employment for all." It is also a highly controversial policy, with few fans and many detractors.

Some of this is due to the history of expedient implementation, where affirmative action came to mean a ham-handed system of quotas. But much of the unease is due to disagreement with the policy’s intent.

Many conservatives argue that fairness requires that we do away with preferences and treat everyone exactly the same way. Meanwhile, some liberals criticize nondiscrimination statutes for their focus on race, religion, and gender to the exclusion of socioeconomic factors that can be more limiting. How, they argue, could you consider the son of an African-American neurosurgeon to be more disadvantaged than the son of an illiterate white sharecropper?  It’s a very good question.

The bottom line is that “classic” affirmative action is not exactly the flavor of the month.* Historically, there have been two main rationales for these policies. The first is compensatory; that is, the need to “right past wrongs”, such as slavery and its legal, political, and economic aftermath. The second is the notion that diversity is valuable and worth pursuing. To modern eyes, these may seem to be weak justifications. “Diversity” is a fairly abstract goal that must contend with a host of other, equally compelling goals, such as the maintenance of high standards. As for “compensation," it can seem manifestly unfair when one realizes that the individuals who benefit weren’t the actual victims of the past bad behavior, nor are those being “penalized” its perpetrators.

Having said that, I’d like to propose a third rationale, one that most of the nation claims to support – fairness. That is, Martin Luther King’s vision of a society that rewards each individual based purely on “the content of their character," without respect to race, religion, gender, class, or any other extraneous factor.

This, of course, begs the question: By that definition, wouldn’t a “color-bind” or “meritocratic” system be the most fair? According to the overwhelming preponderance of social science research, the answer is “No” (or at least “Not yet”). Indeed, the available evidence (see here and below), suggests that our current systems, organizations, and institutions, still favor – not always but often –  affluent white males. Not the poor, not racial minorities, not women.

In this light, affirmative action is needed, not so much as compensation for historic injustices, but as a sort of ballast, counterbalancing unconscious – but unquestionably still prevalent – current biases and prejudice.**

So, instead of eliminating affirmative action policies or continuing to limit them to a few protected categories, why not broaden them to encompass all social distinctions that are known to impede equal opportunity – including the social class of (and economic deprivation experienced by) the white sharecropper’s son?

If affirmative action was understood to be both relevant to the present and consistent with the nation’s meritocratic aspirations, perhaps it would enjoy greater support and also better serve the needs of the nation. Below, drawing on social psychological research, I hope to explain why.

Merit has become synonymous with fairness: the idea that, no matter who you are or where you come from, your talent and level of effort will determine how you fare in life. Surveys reveal that most Americans believe meritocracy is not only the way the system should work but also the way the system does work.

Despite these beliefs, research has established that, more often than not, true meritocracy is an illusion. This is because an implicit system of preferences organizes society’s perception of competence in a way that tends to benefit affluent white males over other groups. For true fairness to surface, this “default” system needs to be dismantled or, at a minimum, counter-balanced. In effect, “no preferences” continues to mean “preference for high status social groups." The idea that eliminating overt forms of discrimination, such as Jim Crow laws, also curtailed more subtle forms of prejudice is a myth unsupported by social science research.

Most of us do not like to think of ourselves as prejudiced or biased. We may be unaware that prejudice can persist in unconscious, covert forms and that stereotypes can shape our attitudes and behavior, even in the absence of conscious antipathies toward groups. Indeed, stereotypes often operate implicitly (also here and here), regardless of our own race/gender, and even when we disagree with its content.

But what do stereotypes have to do with merit? Social life is much about anticipation; as individuals we are always trying to gauge the relative value and utility of our actions and those of others. We are constantly forming expectations for how well (or poorly) we and others will perform in certain situations. These anticipations are partially formed based on attributes such as race, class, gender, or education – see here. Why these characteristics and not others (e.g., hair color)? Because collectively we associate valued abilities to certain - but not all - categories (e.g., men, white) of some social distinctions but not all distinctions. While we do not generally think red-haired people are better (or worse) at math than dark-haired people, on some level we are aware of the cultural belief that men posses greater math ability (a highly valued skill) than women.

So, what are the consequences of all this? In Reflections of an Affirmative Action Baby, Stephen Carter describes obstacles that African Americans face when they attempt to establish their competence: "Our parents' advice was true: We really do have to work twice as hard [as whites] to be considered half as good."

These words describe a common observation by members of lower status groups: Due to status beliefs that disadvantage them, they must actually perform at higher levels than members of higher status groups to be judged as equally competent. More generally, ascribed characteristics such as race influence the level of performance required for inferring ability. In fact, the standards and criteria used in evaluations can (and do) change as a function of attributes (i.e., class, race etc.) of the person being assessed.

So, despite reductions in blatant forms of discrimination, some groups continue to experience subtle but systematic barriers to advancement. Individuals who possess characteristics that produce higher expectations (and/or lower standards) are generally held in higher esteem and tend to have more influence over others. In a self-fulfilling fashion, because we anticipate that these individuals will make more competent contributions, we give them more opportunities to make them, pay more attention to their feedback, and perceive such feedback as more meritorious.

For example, in an audit study of employer hiring behavior, researchers Bertrand and Mullainathan (2003) sent out identical resumes to real employers, but varied the perceived race of the applicants by using names typically associated with African Americans or whites. The study found that the “white” applicants were called back approximately 50 percent more often than the identically qualified “black” applicants. Similarly, Greenhaus and Parasuraman (1993) found that the achievements of black managers were less likely to be attributed to ability or effort, and more likely to be attributed to help from others, than were those of white managers.

The same patterns has also been documented for gender. For example, in a well-known natural experiment, Goldin and Rouse (2000) found that, after symphony orchestras switched to blind auditions (with musician performing behind a screen), the rate of hire for female musicians increased by 25 percent. Trix and Psenka (2003) found that letters of recommendation for female faculty applicants tend to put them at a disadvantage relative to similarly qualified male applicants – that is, the letters for women were shorter, contained fewer descriptions of their research accomplishments, and included more doubt-raising language.

These are but a few of the numerous studies that demonstrate how attributes such as race, gender, and class*** – today – systematically organize the ways in which we perceive competence, influence, and exercise deference to the general advantage of affluent white males. As a consequence, valid and objective assessments of merit for women, people of color, or the poor can often be harder to achieve than for affluent white men.

So, while the principle of merit should govern our society, in most situations, it is unlikely that it does. In fact, blind belief in the fairness of a “meritocratic” system can, in itself, be a concern. In a recent paper Castilla and Benard (2010) showed that, when an organizational culture promotes meritocracy, managers actually show greater bias in favor of men over equally performing women. The authors provide a potential explanation, which is that individuals may be more prone to express prejudiced attitudes when they feel they have established their moral credentials as non-prejudiced. Castilla and Benard conclude that their findings are “a cautionary lesson about the potential unintended negative consequences of organizational efforts to reward merit. If not implemented carefully, such efforts may prove unhelpful or even harmful."

In short, the members of socially devalued groups are frequently caught in an invisible web of contemporary biases. As a result, selective opportunities – such as admittance to an elite university or landing a high-paying job – may be unfairly limited for these groups. If we are serious about fairness, equality, and diversity, we need to rethink and broaden – not eliminate – affirmative action. Particularly important is the need to expand affirmative action measures to encompass social class – poor people not only face prejudice in the form described above, but also the tangible impediments associated with impoverishment.

As we move further and further away from the era of overt, legal discrimination, it is to be hoped that perception and reality will come into ever closer alignment, and affirmative action will become less and less necessary as a means to ensure a fair and just society. Until that time, these policies can be viewed as a useful mechanism by which to bring merit and reward into proper calibration.

- Esther Quintero

*****

* Actual research on who supports affirmative action suggests that it is necessary to hold two views simultaneously. First, the conviction that meritocracy is the way the system should work. Second, the belief that the system is not meritocratic. In contrast, people who oppose affirmative action also value meritocracy, but think that the system is already meritocratic, so no change is needed.

** When biases affect the way we perceive and assess other people, we refer to this as discrimination. When biases affect the way we perceive ourselves, we talk about “stereotype threat” – i.e., fear that we will be judged based on stereotypes about our social group. Although my discussion focuses on the former set of issues, both phenomena have to do with the fact that people are not a blank slate or a tabula rasa but instead, they come to develop shared preconceptions about themselves, others, and the world around them. At times these preconceptions can be useful shortcuts, but at other times they can impede objectivity and accurate perceptions of one’s own talent and that of others.

*** Although I am not aware of comparable studies that look at social class, we know that perceptions of the poor (and particularly welfare recipients) reflect attitudes and stereotypes that ignore these individuals’ strengths and competencies and attribute their poverty to personal failings (also here), rather than socio-economic structures and systems. Also, within the “Expectation States” and “Status Characteristics Theory” research programs (cited throughout this essay), there are studies on “social rewards” and “performance expectations”, which find that when two people are rewarded differently for doing the same job, the one who is paid more is also subsequently treated as higher status.