The Volokh Conspiracy

Entry Four: What Would the Black Bar Look Like With a Reduction or Elimination of Law School Racial Preferences?

Perhaps one of the reasons that few legal academics have delved very far into the issue of law school racial preferences is the widespread assumption that, whatever the other costs and benefits of the system, admissions preferences are the only way to racially integrate the bar. The claim that has been repeated many times - and which figured prominently in the Grutter briefs -- is that black enrollment in law school would drop 50-90% if preferences were abolished. Since a decline of that magnitude seems unimaginable, so does a serious questioning of admissions preferences.

In my research, I found that the usual method academics use to project admissions in a race-neutral world is seriously flawed. Most such projections assume that minority students will continue to apply to the same schools - and only those schools - if preferences disappear. But if preferences are large, and minorities (blacks in particular) take preferential policies into account in deciding where to apply, then it is necessarily the case that minority applications will overwhelmingly be rejected in a race-neutral world.

If one instead asks what proportion of the black applicant pool would be admitted to some law school under a race-blind system, one gets a much rosier result. A 2003 study by Dr. Linda Wightman (who headed LSAC's research operations for many years) found that 86% of blacks admitted to law school in 2001 would have qualified for some law school under a race-blind system. That number has risen sharply over the past decade, because of a steady rise in the number of black applicants and a gradual but steady narrowing of the black-white credentials gap.

Now, recall that in my research I found racial preferences tend to systematically lower black performance in law school, black graduation rates, and black rates of success on the bar exam. Only 45% of entering blacks under the current system graduate and pass the bar exam on their first attempt; another 12% pass the bar on some later attempt. I estimate that the 45% figure would rise to 74% under a race-blind regime. If the pool of entering black law students shrinks a little, but their survival rate rises sharply, it's not hard to see why we might well end up producing more, not fewer, black lawyers in a race-blind system. My best guess is that the total number of blacks passing the bar on their first attempt would rise about 22% in a race-blind system; the number passing after multiple attempts would rise 9%.

The emphasis here should be on the word "guess". Obviously, no one knows what would happen to black interest in attending law school in a world without preferences. There might be a surge of interest in a law school world where blacks perform much better, have much higher chances of success on the bar, and dispense with any stigma from affirmative action. There could be an erosion of interest if blacks are faced with attending less elite schools or if many blacks view the end of preferences as a signal that they are unwelcome. I would hope that my findings, and those of other researchers, would reassure black applicants that going to less elite schools is an excellent career move for them, but of course no one knows. My simulations make the neutral assumption that the total volume of black applications won't increase or decrease.

The stories I've read about my article almost invariably cite the "9% increase in black lawyers" projection and infuse it with an air of artificial precision. That's unfortunate, though perhaps inevitable given the exciting aura such a claim carries with it. I'm really trying to make two other points. First, the end of preferences clearly no longer implies a massive hit to the production of black lawyers. The claims of a 50-90% decline were misleading nonsense. (My most vocal critics are now suggesting a 25-35% decline.) I will defend my 9% increase as a much better guess, but the important point is that the range of debate has shifted.

Second, there can be no question that in a race-blind system, the black bar would be healthier in a number of ways. The proportion of practicing black attorneys who have failed the bar at least once would fall from 22% to less than 10%, and black scores on bar exams (even for those who pass the first time) would be dramatically higher. The median earnings of black attorneys (at least for the early career years I have measured) would be significantly higher, and blacks would be distributed across job sectors in a way much more similar to white patterns. And of course, many, many fewer black law students would spend years earning a degree that failed to gain them admission to the profession. With all this said, I am not at all convinced that a total elimination of preferences is the best way to go. It's certainly the cleanest solution, and perhaps the only one that policy-makers could legislate. But I think there are a variety of alternatives worth exploring. Consider, for example, what I call the "4% solution." Suppose that all law schools agreed that, if they use racial preferences for blacks, they would not apply those preferences to more than 4% of the class. Schools would of course continue to admit all blacks who qualified without preferences, so the 4% would be a floor, not a ceiling.

The beauty of the 4% approach is that it breaks the cascade effect. The top ten law schools would presumably fully use the 4%, and would thereby preserve more racial diversity at the top than we would have (at least initially) in a race-blind system. But the next tier of schools would now have many black applicants who formerly went to top-ten schools; they would be significantly less reliant on preferences. In the third tier and below, preferences would be nearly irrelevant. Enough blacks would have shifted down-market so that schools would have very substantial black enrollments (often larger than current black enrollments) with minimal or no preferences.

This approach has several advantages. It effectively confines the aggressive use of preferences to the top tier of schools, where the academic mismatch is most benign in its effects. It blunts the fear of those who believe that the most talented blacks will shift to other fields if they are unlikely to attend top-ten schools. It mitigates the diversity impact on the most elite classrooms, and provides some reassurance that pipelines of talented blacks into prestigious clerkships and legal academia remain open.

My hope is that, by developing some rough consensus on how to model the systemic effects of affirmative action, we can have a much richer dialog and can identify and test possible compromises, like the 4% solution, that break the ideological logjam.

***Thanks to readers who have sent me comments and questions. On Monday, I'll respond to a number of these.***

Entry Three: Black law graduates in the job market

(My apologies for posting my planned Wednesday post today)

Tuesday, I wrote about some of the perverse effects racial preferences have upon blacks in law school and on the bar. There was already some awareness that some of these problems existed, though most of my colleagues were surprised by their scale and by the tendency of preferences to exacerbate existing gaps. But it has been almost universally believed that an unmitigated blessing of racial preferences, from the perspective of black beneficiaries, is the entrée preferences give to good jobs. Since employers like to hire new attorneys with famous name-brands, giving blacks boosts into more elite schools is thought to enhance greatly their marketability and, eventually, their power and position in the legal hierarchy.

A new database that I describe in my article provides the first nationally representative data on young lawyers and their jobs. Although the data is not perfect, it makes it possible to weigh the effect of over a dozen background factors in shaping the types of jobs new lawyers get. Analyses of the data show, quite strikingly, that employers care — and care a lot — about how job-seekers did in law school. Law school prestige is important, but for law graduates as a whole, good grades are a much more powerful predictor of getting a higher-paying job than the eliteness of one's school.

What this implies about racial preferences is not completely obvious. One needs to estimate both how much of an "eliteness" boost the typical black applicant gets in the admissions process, and how much the average black student's law school GPA would go up if admissions were race-blind and the student went to a lower-ranked school. Both calculations are difficult, and subject to some debate. That said, I think the general pattern is fairly clear. Anywhere outside the most elite schools, new black lawyers are hurt by preferences more than they are helped. For a typical black graduating from a middle-ranked law school, the grades/prestige tradeoff that goes with affirmative action lowers her earnings by about twenty percent.

I found that at the most elite (top ten) law schools, blacks gain enough from the enhanced prestige of their school to roughly offset the grade disadvantage. This seems intuitively plausible, too. Yale, Harvard and Stanford are universally known as blue-chip schools, while many employers won't know what distinctions to draw between Fordham, Iowa and Case Western. And it is probably true that some very elite employers largely limit recruiting to the top ten schools. But even at the top, I was unable to find a clear net plus for blacks from preferences — just a wash.

One might suppose that some employers use grade cutoffs, in part, as a device to discriminate against minority job candidates. This may be true for some, but the general pattern is just the opposite: legal employers pay a premium to recruit junior black lawyers. In other words, when one controls for lots of background characteristics, new black lawyers earn seven to nine percent more than other lawyers with comparable backgrounds (I found no similar premium for Hispanic or Asian attorneys). Blacks with high GPAs do extremely well in the market regardless of where they went to school.

But there's the ironic rub. Law school preferences create a situation where high-GPA blacks are a rare commodity. Blacks with very low GPAs - or even worse, blacks who have failed a bar exam once or more - are very common, and they are penalized substantially in the job market.

One by-product of this state of affairs is the large concentration of black lawyers in government and in small firms. Part of this seems to reflect personal preferences; there's a variety of evidence that black lawyers, on average, do more pro bono and community service than the average white lawyer and are more inclined by personal philosophy to work in government. But the concentration is also partly due to the preferences/low grades connection. Black law graduates with grades at the middle of their class or higher have a pattern of early careers that more closely resembles the white pattern than the pattern of other blacks. Jobs with local government agencies or small firms are sometimes a last resort for students with poor grades - and this is a choice blacks face far more often than they would in the absence of large admissions preferences.

My findings about the job market tradeoff between school eliteness and grades have implications for all law students, not just blacks. The implication of my findings is that going to the best law school one gets into - a strategy almost everyone seems to follow - may not be a very good strategy at all. It is important for students to realistically assess how well they will do at the schools that will have them, and to pick a school where they are likely to be at least in the middle of their class. Middle- and low-tier law schools, under this view, deserve a lot more respect than the very hierarchical world of legal education tends to accord them.

Entry Two: The Effects of Preferences on Grades, Graduation, and the Bar

As I discussed yesterday, a very large majority of American law schools essentially race-norm black and white academic credentials when they admit their classes. Since the black/white credentials gap in the applicant pool is quite large, this means a typical law school has very little overlap between the highest credentials of its black students and the lowest credentials of its white students. This wouldn't matter very much if, as critics have long argued, the LSAT and UGPA were poor predictors of law school performance. And it's true that, individual by individual, these credentials are only rough indicators of performance. But applied to groups they are extremely accurate.

Consequently, blacks as a group have academic trouble in law school in very consistent and predictable patterns. At American law schools that use large racial preferences, half of all black students end up in the bottom tenth of their first-year class. Put a little differently, the median black student performs in the first-year at about the 7th percentile of the median white student. The gap is statistically no different in legal writing classes than in classes with timed exams. And, when we adjust for dropouts, the black-white gap gets slightly wider over the second and third year of law school.

It's important to note that this performance gap has nothing to do with race per se; whites who attend law schools where their credentials are far below most of their peers have pretty much the same types of troubles. The performance gap is a function of preferences. (Now, it's true that the preferences come about in the first place because of the black/white credentials gap - but that is another story, which I'm happy to address later if readers are interested.) There is no credible evidence I've seen that, if schools used race-blind admissions, blacks would underperform whites at all. (I will discuss this issue further, and respond to some reader commentary, this Friday.)

The most obvious consequence of the grade gap in law school is that blacks are expelled, or drop out, at much higher rates than whites (19% of blacks don't complete law school compared to 8% of whites). Almost all of the attrition is among students with very low grades. The more serious consequence is that students at the bottom of the class apparently learn less than the same student would learn at a lower-tier school where the student was closer to the middle of the class. This is what's known as the "mismatch effect".

A number of studies of college students have found various types of mismatch effects. Black law students who go to schools where their credentials are far below most of their classmates are less likely to graduate, more likely to switch out of science majors, and more likely to abandon aspirations for an academic career than blacks who attend college where their credentials place them closer to the middle. These studies have been hampered, however, by the absence of any general test that college students take after graduation; no one could demonstrate that blacks actually learned less at more elite schools.

For my research, I was able to capitalize on a massive database compiled by the Law School Admissions Council in the 1990s, which tracked law graduates through up to five attempts to pass the bar. Bar exams vary state by state, but they have much in common and there are ways to control for the variations. I found that law school grades predicted bar passage rates far more powerfully than school eliteness did -- it was more important to be near or above the middle of the class than at a higher-ranked school -- and, again, this was equally true for blacks and whites. However, since the preferences system pushes most blacks to attend more elite schools, the tradeoff has devastating effects on black bar passage. Blacks are 50% to 100% more likely to fail the bar on their first attempt than are whites who started law school with identical credentials. Combined with the admission, at the bottom of the law school hierarchy, of blacks with very weak academic backgrounds, and one finds that nationally, blacks fail the bar at four times the white rate.

Taking the graduation effect and the bar effect together, and one finds that only 45% of blacks who started law school in 1991 graduated and passed the bar on their first attempt (compared to 80% of whites). Again, this is not a "racial" effect, but a preferences effect. I find in my analysies that the graduate-and-pass rate for blacks would rise to 74% in a preference-free system -- still a little lower than the white rate, but only because the distribution of black credentials is lower than the white distribution.

Tomorrow: Black law graduates in the job market