Jumping To Conclusions

Over at Reason's Hit & Run, Nick Gillespie blogs about a new study by the National Bureau of Economic Research which suggests that job applicants with identifiably "black" names have a tougher time than job applicants with more neutral names. The study uses the phrase "white sounding names," but gives as an example "Greg Baker," which doesn't sound stereotypically "white" to me. (I was initially confused when I read this, getting a strange sense of deja vu, but apparently this is the second such study in recent months.)

From the study's press release:

A job applicant with a name that sounds like it might belong to an African-American – say, Lakisha Washington or Jamal Jones – can find it harder to get a job. Despite laws against discrimination, affirmative action, a degree of employer enlightenment, and the desire by some businesses to enhance profits by hiring those most qualified regardless of race, African-Americans are twice as likely as whites to be unemployed and they earn nearly 25 percent less when they are employed.

Gillespie adds:

The study has weaknesses (e.g. it only measures callbacks rather than job offers; it relies only on newspaper ads; etc.) but it raises tough questions about fairness in American society. What's worse, any sort of near-term remedy is far from self-evident. As the authors note,

From a policy standpoint, this aspect of the findings suggests that training programs alone may not be enough to alleviate the barriers raised by discrimination, the authors write. "If African-Americans recognize how employers reward their skills, they may be rationally more reluctant than whites to even participate in these programs."

Similarly, even supporters of affirmative action will have to acknowledge that a) these results occurred in an employment system that already has affirmative action and b) if the problem is that prospective employers worry that "race signals lower productivity," those employers will work around policies designed to ameliorate racial disparities in hiring.

Surprisingly, neither Gillespie nor the study nor any of the participants in the long comment discussion which follows note that affirmative action, rather than being an ineffective solution, may actually be a cause of the problem. Given the existence of affirmative action, and given two applicants of similar credentials on paper, the minority applicant, on average will be less qualified, and there is no way for the decisionmaker to distinguish between the black college graduate with the 1200 SATs and the black college graduate who only had 900 SATs and needed the boost from affirmative action to get in.

I'm not naive enough to suggest that having a name like LaKeisha might not create a subconscious effect in the mind of the decisionmaker -- though as some commenters pointed out, other identifiably non-black ethnic names such as Billy Bob might have similar effects -- but to ignore the effects of affirmative action is dishonest. A rational decisionmaker, without any racist intent, would maximize his chances of getting a better applicant even given appparently equal credentials by choosing the one less likely to be black.