Signaling and Racism

Racial stereotyping illuminates the dark side of signaling. People judge books by their covers because the covers are visible and easier to grasp than the book’s contents. Similarly, individuals’ physical characteristics are easier to evaluate than their personalities. Unfortunately, therefore, people sometimes make decisions based upon race when ignorant of a person’s vastly more significant qualities.

Although perhaps not moral, using race as a signal can be rational. This rationality is not predicated upon genetic differences among the races. Using race as a signal is rational if race is merely correlated with less visible characteristics. For example, imagine that your company wants to hire a recent Malaysian college graduate. You want the smartest student you can find. Let’s assume that Malaysian colleges discriminate against ethnic Chinese in admissions. Consequently, it is harder for an ethnic Chinese than an ethnic Malay to be admitted to a Malaysian college. Given this discrimination, you would expect in Malaysia that, on average, Chinese college students are more capable than Malaysian college students. If the colleges discriminate against a group, then the school must have higher standards for that group, implying that students from this group will be on average better than the rest of the college’s student population.

So, you want to hire a Malaysian college graduate, and you believe that on average, Chinese graduates are more capable than the Malaysian graduates. How important is this racial difference? It’s of absolutely no importance if you can determine each job candidate’s quality. You care about competence, not race. If you can determine a candidate’s competence, then race becomes irrelevant. If, however, competence is difficult or even costly to evaluate, then it becomes rational for you to use race as a factor when hiring, because race is correlated with competence. Even if a book’s cover would have no effect per se on how much pleasure you would derive from reading the book, it’s still rational to base your purchase decision on the cover if it tells you something about the contents. Signaling theory shows that if colleges discriminate against some race, then employers might desire to discriminate in favor of this race. Alas, the reverse also holds true.

Affirmative action can harm racial groups to the extent that a college is a signal of quality. Assume that high school students can be academically scored from 0 to 100. Let’s say that some highly selective college admits students from race X with a score over 90 and, because of affirmative action admits students from race Y with a score over 85. Imagine that the primary benefit of attending this college is that it signals your high rank. Unfortunately, if this school’s affirmative action policies are known, the signaling benefit of attending this college will be lower for group Y than X. Tragically, even members of group Y who have scores of 100 will be hurt by affirmative action, because potential employers could more easily judge their race than their intelligence or score.