Negative action: backfiring racial preferences.

COULD THE elimination of affirmative action policies in American law schools result in more black lawyers being admitted to the bar--perhaps as many as 8.8 percent more? That's the counterintuitive conclusion of a new study by UCLA law professor Richard H. Sander.

The central argument is one that critics of affirmative action have long made: Students admitted to a school under looser standards are, almost by definition, less likely to be prepared to succeed once admitted, undermining whatever benefit the preference was meant to confer. Sander set out to test that notion using data on 27,000 students from the Law School Admission Council. He found that more than half of black students were in the bottom tenth of their classes by the end of their first years. Black students entering law school were 135 percent more likely than their white counterparts not to complete a degree, and among those who did graduate, failure rates on the bar exam after repeated attempts were six times higher.

Sander projects that, while the elimination of preferences would reduce the number of black students matriculating at-and, to a lesser extent, graduating from--law schools, it would increase the total number passing the bar and going on to practice law. He suggests that most of the net loss in the larger pool would occur at the bottom end, because students denied spots at the most elite schools in the absence of preferences would fill spots in second-tier schools, pushing some students there into their next-choice schools, and so on. That shift to less-elite schools might, paradoxically, improve bar passage rates: Sander argues that students typically learn less (and do far less well) in classes that prove too difficult than they would otherwise be capable of.

Stanford Law Review, in which Sander's study appeared, will devote a future issue to critiques of his thesis.