It’s been 13 years since I wrote the article below (for The Weekly Standard) on Richard Sander’s intensive study of affirmative action in American law schools. Sander, a lawyer and economist, found “a system of racial preferences that, in one realm after another, produces more harms than benefits for its putative beneficiaries.” Sander became one of the most compelling critics of affirmative action, the co-author with Stuart Taylor of Mismatch. Read my article as an introduction to the mismatch school of thought. Pick up the book for the full treatment of the topic.
Affirmative action emerged in the 1960s as a policy intended to help black Americans. How, then, would institutions committed to affirmative action respond if it could be shown that the policy does blacks more harm than good? Richard Sander, a law professor at UCLA, is about to find out.
This week the Stanford Law Review will publish his article, “A Systematic Analysis of Affirmative Action in American Law Schools.” By “affirmative action” in the law schools, Sander means the racially preferential variety used in admissions, and his focus is exclusively on preferences extended to blacks, the original beneficiary group, the other such groups having been added later (and for less compelling reasons).
The title of the 117-page study is as dull as Sander’s conclusion is sharp. “What I find and describe,” he writes, “is a system of racial preferences that, in one realm after another, produces more harms than benefits for its putative beneficiaries.” Sander makes the further, riveting point that “the annual production of new black lawyers would probably increase if racial preference were abolished tomorrow.”
Who is Richard Sander anyway? Perhaps not the man you would imagine from the analysis above. A lifelong Democrat, a liberal on most issues, he has a long record of involvement in civil rights issues, including housing segregation. His son is biracial. “So the question,” he notes in the article, “of how nonwhites are treated and how they fare in higher education gives rise in me to all the doubts and worries of a parent.” Because he favors race-conscious strategies in principle, his article is a classic instance of following an argument wherever it leads. At volokh.com–the blog where he summarized his findings–he wrote that he was “surprised and dismayed” by his “generally negative conclusions,” which “put me at odds with many close friends.”
In an interview, he told me there could be “a significant professional cost” for having written the article, which is already under attack. But he remains confident of his findings, and he’s prepared to defend them. He’ll reply to critics in a future issue of the law review. And he’s writing a book looking at the impact of affirmative action on all favored groups, not just blacks.
There’s something else to report about Sander, perhaps the most important thing. Besides being a lawyer, he’s also an economist. And “A Systemic Analysis” is plainly the work of an economist. Sander doesn’t address the familiar legal issues involved with affirmative action (no parsing of strict scrutiny here) but instead asks whether preferences “meet their simplest goals of producing more and better black lawyers.”
To answer it, Sander needed relevant data. Through FOIA requests he collected 2002 and 2003 admissions data from seven public law schools, some of them among the nation’s very best. He also worked from the data gathered by the Law School Admission Council on one national cohort of law students–27,000 students who entered law school in 1991. LSAC, which tracked the students (representing 95 percent of all accredited law schools) through 1997, collected information on the students’ admissions credentials (LSAT score and undergraduate GPA), race, academic performance, and bar exam outcomes.
Sander was working with his data sets 18 months ago when the Supreme Court upheld the admissions preferences used by the Michigan law school. Justice O’Connor’s opinion for the Court in Grutter v. Bollinger accepted the notion advanced by Michigan and its large roster of amici that preferences benefit those they target, which in this case included blacks, Latinos, and Native Americans. Sander had come to the opposite conclusion, and what he noticed about the Michigan law school case (and its companion case from the University of Michigan) was, he told me, how “saturated” they were with social science. “So much of it was really bad. That was one of the reasons I wrote [the article].”
In the picture Sander draws, the admissions preferences for blacks are very large. This is the case with respect to almost all law schools. Which is to say, contrary to conventional wisdom, preferences aren’t confined to the elite schools. Indeed, “affirmative action has a cascading effect through American legal education.” The top-tier law schools enroll not only the small number of blacks who don’t need preferences to get in but also less-qualified black applicants whose credentials would have sufficed to gain them admission under a race-blind standard to a second-tier school. Second-tier schools are then forced to choose between having few if any black students (under a race-blind standard) or using preferences to reach their racial goals. The second-tier schools make the latter choice, and so the effects cascade to the third-tier schools and on down the law-school ladder. There is thus a “system” in place whose net effect is “to move nearly all blacks up a tier (or two) in the law school hierarchy.” Only at the bottom–in the lowest-tier schools–do you find black students who are probably unqualified for any law school.
The cascading effect leaves most black students “mismatched” with peers whose academic credentials (in terms of LSAT scores and UGPA) are superior. Which means, as Sander puts it, that “nearly all blacks [are placed] at an enormous academic disadvantage in the schools they attend.” And so there are “mismatch effects.” In their first year, about 50 percent of black law students end up in the bottom tenth of their class, and roughly two-thirds in the bottom fifth, with only 8 percent placing in the top half. The grades of black law school students go down a bit from the first to the third year. Black students have a much higher attrition rate than white students (19 percent compared with 8 percent). Sander finds that fact unsurprising, since students (of whatever race) with the very worst grades are those who are expelled or drop out. Finally, black law school graduates fail the bar exam at four times the rate of white graduates. Sander concludes that more than 40 percent of black students starting out to become lawyers never reach that goal.
Sander faults the preferences used to admit blacks for the systematic mismatch and its effects. He contends that black students perform the same academically as whites with similar test scores, and that if blacks were not mismatched, they would have better grades and learn more. And he adduces evidence showing that grades are powerful predictors of bar exam performance. Not incidentally, employers, he also found, prefer to hire graduates with good grades.
Sander envisions a law school universe without preferences, which is not, he says, “an unthinkable Armageddon.” There would be many fewer blacks enrolled in the top 20 schools, he says, but overall the number of black students would shrink only slightly–by about 14 percent. And without mismatching and its effects, black students would perform better, fewer would fail to graduate, and 8 percent more would pass the bar.
Because elite schools would have few black students–perhaps 1 or 2 percent of their student bodies–and because those schools fancy themselves as critical in shaping the views of future national leaders, Sander suggests “an intermediate step” by which the elite schools would still use preferences, but on a limited basis, such that a class is 4 percent (and not the current 7 to 8 percent) black. Sander says this would dampen the cascading and bring an end to mismatching “at some point fairly high in the law school spectrum.” It would appear that Sander offers the 4 percent solution as a concession to elite opinion, a way to advance a dialogue on the costs and benefits of preferences, not because he believes in it himself. After all, his basic conclusion is this: “Blacks are the victims of law school programs of affirmative action, not the beneficiaries.”
To judge by the academic response so far, Sander is going to be busy discussing his analysis. A group including the main architect of the Michigan law school admissions policy has written an article disagreeing with Sander’s conclusions, including his assessment of what would happen under race-blind admissions. They claim the black law school population would shrink by 35 to 45 percent, and that the number of black graduates who pass the bar would decline by 25 to 30 percent. But other academics working from Sander’s data sets are reaching the same conclusion as he. If Sander wins the social-science battle, that could have repercussions in litigation over admissions preferences, especially in the Supreme Court, where one of the justices who voted to sustain the Michigan policy, Justice Stephen Breyer, practices a jurisprudence (applauded by liberals) that looks to the consequences of a policy in assessing its legal merit.
Sander’s article promises to reopen an issue that seemed settled in Grutter, notwithstanding that it was a 5-to-4 decision. Sander even imagines the next major legal challenge to admissions preferences: a lawsuit brought by black plaintiffs “who were admitted, spent years and thousands of dollars on their educations, and then never passed the bar and never became lawyers–all because of the misleading double standards used by law schools to admit them, and the schools’ failure to disclose to them the uniquely long odds against their becoming lawyers.”
That would certainly disturb the sleep of all the law school admissions officers who breathed a sigh of satisfaction after Grutter.
