A systematic response to systemic disadvantage: a response to Sander.

I. THE NOT-SO-GOLDEN AGE
II. MAKING ELITE BLACK LAWYERS
III. JUST MAKING IT
IV. SOME STRAIGHT TALK ABOUT STRAIGHT TALK

Consider the following story:

On November 22, 2004, Anthony R. Chase, CEO of ChaseCom, a call-center provider for telecommunications companies, returned to give the Traphagen Distinguished Alumni Lecture at his alma mater, Harvard Law School. Chase's lunchtime talk departed from the standard fare for such gatherings. Instead of reminiscing about how much he loved law school in the late 1970s and demurely cataloguing his professional success since graduation, Chase candidly told his young audience that he often didn't have a clue what his legendary professors were talking about--that is, when he bothered to stop in to hear what they were saying--and that his path from affirmative action admittee to CEO of his own company with gross revenues in the tens of millions had been anything but straight and trouble-free. He had failed many times, he told them, but had always been able to pick himself up. It was at these moments that his Harvard degrees (he has three of them) had proved most useful, providing both contacts and credibility that allowed him to pursue his dreams anew.

At the conclusion of the talk, the diverse group of students in attendance applauded wildly. As several confided in me afterwards, no one had ever talked to them like that before. Here was someone whose real struggles they could relate to--and whose honesty gave them the courage to admit that they had struggles of their own. A half dozen or so asked me for Chase's email so that they could tell him how much the talk had meant to them. One of those who did was a former student of mine from Taiwan. In addition to conveying her appreciation for Chase's talk and her admiration for his outlook on life, the student also wanted to make a connection. Her brother and his college roommate from MIT had recently started a telecommunications company in Taiwan. She wondered whether his company and ChaseCom might be able to do business. Four emails later, the black CEO from Texas and the entrepreneur from Taiwan were in discussions about a possible business deal.

Now ask yourself the following question:

Would Anthony Chase have been better off if he had gone to Boston College Law School instead of to Harvard?

If you buy the logic of Rick Sander's apparently well-researched, well-argued, and undoubtedly well-intentioned paper, the answer to this question would have to be yes. Notwithstanding a few qualifications along the way, even elite school graduates like Anthony Chase, and all of the other beneficiaries of affirmative action over the last forty years, would--on average--have been better off going to lower-ranked law schools where their entering credentials would have fit better with the entering credentials of their white peers.

To be sure, it is easy to lose sight of the fact that this is Sander's central claim. Many of the article's hundred-plus pages, and the vast majority of his statistical analyses, are devoted to other arguments--for example, that contrary to the Supreme Court's finding in Grutter, (1) law schools actually operate a dual admissions system in which blacks are evaluated separately from whites, or that law schools form a connected "system" in which admissions decisions taken by schools higher up the prestige chain have predictable consequences for the admissions decisions of lower-ranked schools. As interesting as these claims may be, however, they are, by Sander's own terms, peripheral to the "principal question of interest" his article is designed to address: to wit, "whether affirmative action in law schools generates benefits to blacks that substantially exceed the costs to blacks." (2) To put the point bluntly, even if law schools are operating a dual admissions system in violation of Bakke (3) and Grutter, so long as that system generates "benefits to blacks that substantially exceed the costs to blacks," (4) it would pass the minimal but crucial test that Sander poses for its legitimacy.

Similarly, it is easy to lose sight of the reason why Sander asserts that the costs to blacks of affirmative action outweigh the benefits. A good deal of Sander's analysis is devoted to arguing that many black students receive low grades in law school. Once again, this is an interesting and important issue--one that undoubtedly deserves the attention of those of us who spend our time teaching in law schools. But for Sander's overall argument, low grades are only important, as he concedes, if they produce "bad outcomes" such as "higher attrition rates, lower pass rates on the bar, [or] problems in the job market." (5) After all, in the days before affirmative action, all of the bottom places on the law school grade ladder were occupied by whites. (6) Once again, to put the matter bluntly, even if affirmative action systematically lowers the grades blacks receive in law school, if the beneficiaries of these policies go on to become successful and satisfied practitioners, then the costs to them of bad grades would be outweighed by the benefits that flow from a successful legal career.

Thus, once we clear away all of the underbrush, Sander's argument that affirmative action hurts blacks comes down to an assertion about the relative value of grades (and all that they represent) versus law school prestige (and all that it represents) in achieving the ultimate goal shared by all law students of becoming a lawyer and building a successful career. (7) By Sander's calculation, the primary effect of affirmative action is to allow black law students to attend schools that are twenty to fifty places above the schools that they would have been admitted to in the absence of these policies. (8) Without more, this would seem to be a clear benefit to those who receive this boost, as Sander concedes virtually every law student, black and white, firmly believes. Consequently, in order to prove his point that blacks are nevertheless harmed by affirmative action, Sander must demonstrate that "the net trade-off of higher prestige but weaker academic performance substantially harms black performance on bar exams and harms most new black lawyers in the job market." (9)

Indeed, although Sander devotes significantly more attention to bar passage than employment success, it is this latter claim--that attending a more highly ranked school actually hurts "most new black lawyers on the job market" (10)--that must in the end be the linchpin upon which his argument either succeeds or fails. Notwithstanding the grim statistics he amasses regarding black attrition and bar failure rates, the fact remains that most black students who begin law school--fifty-seven percent--do in fact become lawyers. (11) Moreover, a significant percentage of those who currently fail to graduate from law school and pass the bar would not become lawyers under Sander's proposal either. (12) Although one can argue that black students who fall into this latter category would have been better off if they had never been admitted to law school in the first place (an assertion that I will return to below), if those blacks who do become lawyers benefit significantly from affirmative action, then it is hard to see why blacks as a group are worse off simply because a higher percentage of blacks fail the bar than whites. (13) To make a persuasive claim that affirmative action harms "most" blacks, therefore, Sander must prove that grades are more important than law school prestige for those black law students who actually become lawyers.

Finally it is important to recognize that to meet this crucial burden, Sander offers a single piece of evidence--and an equivocal piece at that. Sander argues that according to the first wave of responses to the After the JD Study (hereinafter "AJD Study"), black lawyers with high grades from low-status schools are as--if not more--likely to obtain high-paying jobs than their counterparts from higher-status schools with lower grades. (14) On the basis of this single finding, Sander declares that contrary to popular belief, "[r]acial preferences ... have not been an indispensable part of credentialing blacks for the job market; overall, they clearly end up shutting more doors than they open." (15)

In this brief Response, I will argue that this single piece of evidence does not come anywhere close to proving that most black lawyers would be better off in a world in which the vast majority of them would attend law schools twenty to fifty places below the ones that they currently attend, and where hundreds of blacks who currently attend law school today would be denied the opportunity to do so altogether. Affirmative action has played a crucial role in helping black lawyers to overcome the systematic and persistent obstacles that continue to make it more difficult for these new entrants to succeed notwithstanding the progress the country has made in reducing the overt discrimination that for the first two centuries of our history kept blacks out of virtually every desirable aspect of American society. Indeed, it is precisely because these policies have been so successful that for the first time blacks with high grades from lower-status schools have a plausible chance of gaining entry into high-paying positions in the legal profession. These tentative gains, however, are unlikely to continue if the number of black graduates from highly ranked schools were to decline dramatically. Although the picture is more complex with respect to those black students who are unlikely to end up in high-status jobs, they too have benefited more from affirmative action than Sander's analysis suggests. Equally important, the problems those in this group undoubtedly confront have as much to do with the way that bar exams are conceived and administered as they do with these students' potential to become competent practitioners.

The rest of this Response proceeds in four parts. Part I briefly reminds us of the long and sorry history of exclusion that gave rise to the need for affirmative action in the first place and examines how this legacy continues to disadvantage black lawyers. Part II examines Sander's contention that grades are more important than law school status in the context of black graduates from highly ranked schools. Contrary to Sander's assertion, black lawyers in this group gain benefits from their prestigious degrees that extend far beyond the starting salaries that they receive upon graduation. In turn, these fledgling members of the profession's elite provide important benefits to all black lawyers--and to society as a whole. Part III examines those black lawyers who appear to have benefited the least from affirmative action: those who attend lower-tier local and regional law schools. Although Sander makes a persuasive claim that these students face daunting risks in pursuing a career in law, a significant number plausibly benefit from their legal education, even if they do not go on to become lawyers. Moreover, given that bar passage is the most important obstacle facing blacks in this group, we can gain as many new black lawyers by reforming the way that such exams are administered and taken as Sander claims we will produce by eliminating the black students who currently have the most trouble passing. Part IV briefly concludes by arguing that, rather than improving conditions for black lawyers, Sander's proposal runs the risk of making many of the problems he identifies worse.

Before proceeding I want to make one thing clear. Rejecting Sander's bottom line does not mean that we also should reject his invitation to take a hard look at the costs and benefits of our current practices. Sander amasses many troubling statistics about how some black students are performing in law school and on the bar. It is imperative that we face up to these difficult realities. But when we examine these issues, we must also recognize that the "system" of affirmative action in place today is a response to a set of policies and practices that have--and continue to--systematically disadvantage black lawyers. It is only by placing affirmative action in the broader context of how careers are actually forged in today's legal marketplace that we can reach credible judgments about whether such policies hurt some of their intended beneficiaries, and, more importantly, what we might do to rectify this situation.

I. THE NOT-SO-GOLDEN AGE

In Part I of his article, Sander gives a quick nod to the historical conditions that prompted, to borrow his phrase, "the conscience of the legal academy to quiver," precipitating programs that for the first time affirmatively recruited black students to law school. This history, however, is more relevant to Sander's argument than his brief presentation suggests.

As Sander concedes, prior to the mid-1960s there were only a handful of black lawyers in the entire country. (16) What he fails to acknowledge is that this underrepresentation was the result of a deliberate and concerted effort by the profession's elite in both the North and the South to restrict entry to white, Anglo-Saxon, Protestant men of means. (17) Blacks--regardless of their qualifications--were overtly discriminated against by educators and employers alike. (18) The few blacks who managed to become lawyers during this period worked in solo practice or small minority firms, barely eking out a living serving an almost exclusively minority clientele. (19) As late as 1964, Erwin Smigel reported in his study of Wall Street lawyers that "in the year and a half that was spent interviewing, I heard of only three Negroes who had been hired by large law firms. Two of these were women who did not meet the client." (20)

Nor were blacks the only ones excluded. During the 1920s, leading lawyers sought to prevent a "pestiferous horde" of recent immigrants from entering the profession by closing night and part-time law schools and escalating admissions requirements--including requiring every applicant to pass a written examination and a character and fitness review before becoming a member of the bar. (21) Until the 1950s, many law schools maintained quotas limiting the number of Jews that they would admit. Even after the academy dropped their quotas, many law firms would not hire Jews--or anyone but the "right kind of Jews"--until well into the 1960s. (22) Overt and unabashed discrimination against women lasted even longer, with many firms refusing to hire or promote more than a token number of women well into the 1970s. (23) Given this history, it is no wonder that in 1961, President Kennedy's secretary of labor called the American legal profession "the worst segregated group in the whole economy." (24)

It is important to understand that these policies and practices expressed more than simple discriminatory animus. Instead, they reflected deep and widespread beliefs about the very essence of lawyer professionalism. Competence and identity were inextricably intertwined in the minds of lawyers in what some like to refer to as the profession's "golden age." (25) Nowhere was this ideology more prevalent than in the profession's elite. As Smigel observed, "[Wall Street firms] want lawyers who are Nordic, have pleasing personalities and 'clean-cut' appearances, are graduates of the 'right' schools, have the 'right' social background and experience in the affairs of the world, and are endowed with tremendous stamina. " (26) Even after a lawyer was hired, firms considered the associate's ability to fit into the firm's tightly knit social structure to be almost as important as the number of hours he worked or his potential for bringing in business when deciding whether he should be elevated to partnership. (27)

This is the background against which law schools finally instituted policies designed to ensure that more than a paltry number of blacks and other traditional outsiders had access to legal education. (28) As Sander suggests, these programs were typically justified on the ground that black students needed "preferential treatment" in order to gain admission to law school and therefore to increase the size of the black bar. (29) The above history, however, suggests a deeper and more powerful justification. Law schools and legal employers were justified in taking affirmative steps to assist black students and lawyers in order to counteract the systematic and pervasive preferences that had been accorded to white, Anglo-Saxon, Protestant men of means for more than one hundred years. (30)

Judged against this historical background, the last forty years have produced remarkable progress. There are now approximately 40,000 black lawyers in the United States--an increase of almost twentyfold since 1960. (31) Black lawyers can now be found in virtually every sector of the profession and in many of the high-end jobs in corporations, government, and the judiciary for which elite jobs in the legal profession typically serve as a gateway. (32) Moreover, every study examining the careers of this new generation of black professionals has concluded that these beneficiaries of affirmative action have gone on to lead successful and productive careers--careers that from all accounts are equal to, and in some respects even exceed, those of their initially better-credentialed white peers. (33) Given that this progress coincides exactly with the period in which law schools have been pursuing affirmative action programs expressly designed to bring about this result, anyone claiming that such policies have been counterproductive (the strong form of Sander's claim) or are no longer necessary (the weak form) appropriately bears a very heavy burden of persuasion. (34)

At the same time, there is ample evidence that the profession's long history of exclusion continues to affect black lawyers adversely even if we concede, as the evidence also indicates, that overt discrimination against blacks has decreased markedly over the last three decades. I have written extensively about the racialized disadvantages that blacks continue to face in the legal marketplace, especially with respect to elite jobs. (35) For present purposes, it is sufficient to simply say this: Notwithstanding the important progress documented above, a mounting array of evidence confirms that most whites continue to hold a broad range of negative stereotypes about blacks even as they consciously profess to believe in racial equality. (36) At the same time, an equally long line of research confirms what any observer of human nature takes for granted: that people instinctively prefer to work with others who are like themselves. (37) The combination of these common and widespread prejudices can be deadly for blacks seeking to make it in the corporate law firm world. Law is fundamentally a relationship business. At every stage, decisionmakers are asked to make important choices on the basis of limited and imperfect information. Law firms continue to give substantial weight to issues of "comfort" and "fit" when making hiring decisions. Partners choose which associates to train and to select for plum assignments on the basis of first impressions, secondhand information, and generalized reputation. Clients employ similarly impressionistic information when selecting lawyers and law firms for important assignments.

None of this should be taken to mean, as Sander rightly insists, that factors such as "cognitive skill and subject mastery" play no role in career success. (38) They clearly do. This point sometimes gets lost in all of the talk about the "old boy network" or "white privilege." Where Sander errs is in assuming that law school grades accurately map these important qualities. Grades certainly measure something about certain kinds of analytic ability and whether a student has understood and can replicate what the professor believes that he or she was intending to teach. And Sander is also correct in assuming that grades measure these qualities more accurately at the two tails of the distribution (i.e., for those at the top and the bottom of the class) than they do for the majority of students who fall in the middle of the curve.

Nevertheless, grades still remain a very imperfect proxy. It is not simply that grades do not purport to measure, as Sander concedes, how "conscientious, well-spoken, diligent, likeable, or ethical" someone is, although these factors are clearly an important part of what makes someone a good lawyer. (39) They also do not do a particularly good job of measuring creativity, the ability to work well in and manage teams, the ability to listen to and to empathize with others, and, most important of all, that critical but elusive quality that lawyers refer to as "good judgment." Indeed, there are good reasons to believe that the typical way that grades are administered and evaluated in law school may select against traits like teamwork and creativity. And teamwork and creativity--let alone good judgment--are at least as big a part of being a "good problem-solver," to borrow Sander's phrase for what clients value in a lawyer, as cognitive ability and substantive skill in a profession like law where success frequently depends upon the ability to "play well with others" and to persuade them to do what you and your client want. It is not surprising, therefore, that studies that attempt to understand the causes of long-term career success have not documented a robust correlation between grades and other entry-level credentials and long-term career success. (40)

It is precisely because qualities such as creativity, skill, and judgment are both difficult to observe directly and complex to measure that decisionmakers frequently rely on proxies--such as whether someone "has the right background," "exudes confidence," or "makes me feel at ease"--when deciding whom to trust among the broad range of arguably qualified applicants. In a world in which most whites continue to hold negative stereotypes about black intellectual inferiority, these common heuristics systematically disadvantage the career prospects of black lawyers. (41)

To be sure, a growing number of black lawyers have found ways to circumvent these obstacles. And, as the AJD Study data Sander relies on suggests (although, as I will argue below, falls far short of proving), employers of all kinds may be more receptive to hiring black lawyers than they have ever been in the past. These new realities, however, owe much more to the past success and current practice of affirmative action than Sander suggests. To see why, it is necessary to take a closer look at the crucial role that black graduates from elite schools have played in opening the doors of opportunity for all black lawyers.

II. MAKING ELITE BLACK LAWYERS

Sander is willing to concede that black lawyers like Anthony Chase who graduate from Harvard and other schools at the very top of the law school hierarchy may in the end benefit from affirmative action. (42) Even with respect to this most favored group, Sander's concession is grudging at best. Sander does not dispute that black graduates from top schools, like their white counterparts, almost never fail to complete law school or eventually pass the bar. (43) Sander also concedes that elite school graduates earn substantial wage premiums which can reach as high as forty percent. (44) Nevertheless, Sander asserts that even for black graduates from the best schools, this wage premium is at best "a wash" or a "small plus" when weighed against the increased risk that blacks who attend these institutions will receive low grades. (45) For all other black students, the trade-off between school status and grades is at best a substantial penalty for those blacks in highly ranked schools outside the top ten, and at worst leads to an "overwhelmingly negative" reduction in income for those graduating from middle- to lower-ranked schools. (46)

Grudging though it may be, Sander's concession that blacks from the very top schools are likely to benefit from their prestigious degrees is more important than he acknowledges. After all, even under Sander's modified proposal in which the top schools would arbitrarily cap their black enrollment at fifty percent of current levels, half of all of the blacks who now attend top ten schools would be bumped down to schools where they would, by his own admission, not enjoy the benefits that they currently receive. By his estimation this might be as many as 150 people. (47) If affirmative action were eliminated altogether, the number of blacks who would be harmed would be nearly twice as large, since, as Sander concedes, in the absence of such programs there would be few if any black students in the top law schools. In a proposal that claims to value blacks "as individuals," (48) this cost to blacks who currently benefit from the system should not be overlooked.

Moreover, given that the population of black law students is disproportionately concentrated in schools at the top of the law school hierarchy, the actual number of blacks in danger of being hurt by the elimination of affirmative action under Sander's own calculations is actually much larger. Thus, in the entering class of 1991 used by the Bar Passage Study, there were 419 black students enrolled in the top thirty law schools. (49) By contrast, if one excludes blacks enrolled in the seven historically black institutions, which Sander concedes are properly viewed as a special case, (50) there were 1023 blacks enrolled in the remaining 123 schools that participated in the study. Put somewhat differently, the top twenty percent of all law schools account for over one-third of black enrollment in nonhistorically black institutions. If we use Sander's top three categories of schools, which cover the forty most highly ranked institutions, the percentage of all black students who attend schools ranked in the top quarter of all law schools is likely to be even greater. (51)

By Sander's own calculations, eliminating affirmative action runs the risk of damaging the job market prospects of all of these black students. In Table 7.4, Sander estimates that law school status is highly significant in terms of its effect on starting salaries for schools in his top three tiers. (52) Although he predicts that the black students who currently enjoy this benefit would gain as much or more if they attended lower-ranked schools but received grades that were two standard deviations better than they are currently receiving, this is only a guess that depends upon many unknown factors. For example, if a black student currently admitted to a Tier 2 school ends up attending a school in Tier 5 (as opposed to the Tier 4 school Sander predicts), the projected benefit associated with receiving better grades would likely be negated by the status penalty associated with attending a significantly less prestigious school. (53) Similarly, even if most blacks fall only the twenty to fifty places Sander predicts, if those who do only improve their grades by one standard deviation instead of the two Sander posits, they would, once again by Sander's own calculation, be worse off under the new regime. (54) For the moment, however, I simply want to emphasize that Sander's assumptions about the gains blacks will receive by getting better grades are just that--assumptions. Weighed against these assumptions is the very real wage premium that hundreds of black law students currently enjoy as a result of being able to attend schools in the top quarter of the law school hierarchy.

Thus, even on the very narrow terms Sander proposes for valuing the benefits of law school prestige--salaries earned two years after graduation--it is far from clear that a proposal that jeopardizes more than one-third of all black law students who do not attend historically black institutions can be justified as being in the best interest of blacks as a group. At a minimum, this cost must be weighed against the benefits that Sander asserts blacks who attend schools outside of the top forty would receive under his proposal. I return to this question in Part III. To get an accurate sense of the trade-off Sander is proposing, however, it is crucial to recognize that the benefits of attending a highly ranked law school extend far beyond the initial boost in starting salary discussed above.

Sander's analysis suggests that top schools confer two important benefits on their students: an educational benefit, which he assumes blacks are not receiving because they tend to get low grades, (55) and a placement benefit, which he concedes at least some do receive in the form of higher starting salaries. Educational and placement benefits are undoubtedly a large part of why students of all races, creeds, and colors fight so hard to get into top schools. As important as these benefits are, however, they fail to capture anything approaching the full value of attending an elite law school. In addition to acquiring substantive knowledge and gaining preferential initial access to the employment market, students attending elite schools are also socialized into the habits and possibilities of eliteness and granted a lifetime membership in the elite networks to which the graduates of such institutions automatically belong. Just as important, elite school graduates also obtain a visible and durable credential that they can use to signal to employers that they have received all of the valuable goods that elite schools provide--and that employers can use to signal their own competence and connections to clients and potential recruits. Over time, the socialization, networking, and credentialing benefits of a degree from an elite law school dominate the educational and placement advantages discussed by Sander. This is particularly true for black lawyers.

The first of these benefits is nicely captured by Robert Granfield's excellent book about Harvard Law School, entitled Making Elite Lawyers. (56) Granfield argues that more than learning the rule in Shelley's Case, what Harvard students are really acquiring for their $40,000 in tuition is the knowledge of how to assume their rightful place in the nation's elite. This socialization into what Granfield calls "collective eminence" occurs primarily outside the classroom and has as much to do with shaping placement choices as it reflects these opportunities. (57) Anthony Chase's lunchtime talk typifies this socialization process. Schools like Harvard routinely bring in speakers who represent the cutting edge of professional practice. (58) By observing and interacting with these leaders, students get to see firsthand how successful lawyers present themselves and pick up invaluable lessons about what it takes to succeed in the real world. As the example also underscores, the mechanisms by which students are socialized in elite schools frequently provide important opportunities for relationship building.

The network effects of elite schools are so obvious that one would think that they would have to be a large part of any examination of the benefits of attending such institutions. Yet Sander ignores them entirely. The graduates of top schools go on to occupy prominent positions in the professions, industry, government, and the arts. Having the opportunity to form relationships with these future leaders is clearly helpful to one's ability to build a successful career in virtually any field. Moreover, as Anthony Chase's story highlights, school ties extend well beyond friends or even classmates. Being a Harvard, Yale, or Stanford graduate means being inserted into an exclusive club that extends back for more than one hundred years and forward with every new graduating class. The simple fact that one is a "Harvard man" or a "Stanford woman" opens up opportunities for relationship building across generations and domains of expertise and interest. Whether it is being invited back as a distinguished alumnus, exchanging pleasantries at the Yale Club, or simply breaking the ice in conversation in a job interview with a fellow NYU grad, being a member of this exclusive club pays handsome dividends throughout the course of one's career.

There is ample evidence that these elite networks are especially valuable for lawyers. When Smigel studied the partners of Wall Street law firms in the 1960s, for example, he found that an astounding 70% were graduates of either Harvard, Yale, or Columbia Law School. (59) In the intervening years, the tremendous growth in both the number and size of large law firms has caused most to expand the range of schools from which they are willing to hire new lawyers. (60) Nevertheless, elite law school graduates continue to dominate law firm hiring. (61) Indeed, although the AJD Study found graduates from every school tier working in large law firms, only 9% of those in Tier 3 and 3% of those in Tier 4 were employed in firms with more than one hundred lawyers. (62) The corresponding percentages for respondents in top ten and next-ten schools were 40% and 33% respectively. (63)

The concentration of elite school graduates among the partners of these institutions is even greater. Thus in 1995, Mitu Gulati and I found that 70% of all of the partners in five large firms in top legal markets had graduated from one of the thirteen most prestigious law schools in the country. (64) Ninety-three percent of Cleary Gottlieb's partners graduated from these institutions. (65) Studies of the backgrounds of leading lawyers in other high-profile positions show a similar prevalence of graduates of elite law schools. (66) While there may be many factors that explain these patterns, a point to which I will return below, it should go without saying that being a part of a network that includes these powerful decisionmakers carries real value.

The socialization and networking benefits of graduating from an elite school undoubtedly help to explain why a degree from such an institution is universally considered to be a valuable credential in the job market. Even if we credit Sander's analysis of the information on starting salaries in the AJD Study, it is evident this credential provides a more visible and durable market signal than all but the highest grade point average. Over time, employers simply stop inquiring about grades. (67) Similarly, applicants are likely to have an increasingly difficult time working their good grades into the conversation. (68) Where one went to law school, on the other hand, is always apparent. At the same time, firms have traditionally used the number of elite law school graduates they employ as a means of signaling their quality to clients and potential recruits. (69) Although superstar grades and other academically related honors such as law review membership and Supreme Court clerkships can also be used for this purpose, (70) cumulative grade point averages cannot.

As important as these socialization, networking, and credentialing benefits of an elite law degree are for all aspiring lawyers, there are good reasons to believe that they are especially valuable--both singly and, most importantly, in combination--for black lawyers. Black law students are far less likely than their white peers to have absorbed the ways of eliteness at home or in their earlier educational experiences. (71) As a result, my interviews are filled with examples of black lawyers whose eyes were opened to the mores of the power elite while in law school, and with cautionary tales about what happens to those who failed to learn these lessons. (72) Similarly, in a world that continues to be dominated with negative images and stereotypes about blacks, most of the black lawyers in my sample, like Anthony Chase, have found that having relationships and connections with classmates and fellow alumni who have been able to steer them business or otherwise vouch for their competence has played a critical role in their success. (73) And virtually all believe--whether or not they attended such an institution themselves--that having an elite law school degree is an invaluable tool in overcoming the initial barriers that make whites skeptical of their competence and hesitate before giving them important opportunities--especially, as Chase discovered, when times get tough. (74) The sum total of all of these benefits provides blacks who attend elite schools with a significant advantage in the job market. (75)

The numbers bear this out. In 1995, I surveyed the nation's largest 250 law firms to determine how many associates these firms hired in the last year, how many of those in this group were black, and where all of these new recruits went to law school. (76) One-third of the firms responded to the survey. In these firms, the percentage of blacks graduating from one of thirteen specified elite schools was somewhat higher (57.3%) than the number of white graduates (51.7%) from these same institutions. The differences grew larger, however, at the top end of the elite spectrum. For example, in the two cities with the highest response rates, New York (51%) and Washington, D.C. (50%), more than half of all black associates hired graduated from either Harvard or the top schools in the local market: Columbia and NYU in New York or Georgetown in Washington, D.C. The corresponding numbers for whites were 40.4% in New York and 23.2% in Washington, D.C. (77)

When we turn our attention to partners, the percentage of blacks who have broken into this exclusive club who are also graduates of elite law schools is even more startling. For example, in 1993, 77% of the identifiably black partners profiled in the ABA's directory of minority partners in majority-corporate law firms were elite school graduates as I have defined that term. (78) As with the previous comparison, this percentage is somewhat but not dramatically higher than the percentage of white partners who graduated from these institutions in a sample of five top firms from around the country. (79) Once again, however, when we look more closely, it becomes clear that black partners tend to be concentrated at the top end of the range of elite schools. Thus, nearly half (47%) of the black partners graduated from either Harvard or Yale. Although two firms in the sample (Boston's Ropes & Gray and New York's Cleary Gottlieb) have percentages of white Harvard and Yale partners that rival this total, the average for all five firms was 33%. An analysis of the black partners listed in the 1996 directory of the Chicago Committee on Minorities in Large Law Firms reveals a similar pattern. (80) Seventy-three percent of those listed are elite school graduates, with 53% being the graduates of only three institutions: Harvard, Michigan, and Northwestern. As a rough comparison, in that same year 67% of the partners at Chicago's Sidley & Austin, one of the largest and most prestigious firms in the city, were elite law school graduates, with less than one-third coming from the three schools that contributed over half of the entire population of black partners.

None of these comparisons definitively establishes that black lawyers are more in need of elite school credentials than comparable whites. Nevertheless, they are consistent with the countless anecdotal reports that I have collected from respondents concerning the importance of elite school credentials for blacks in law and other high-status jobs. (81) As one black partner in Chicago put it in a recent news account, "If you're not from Harvard, not from Yale, not from Chicago, you're not adequate. You're not taken seriously." (82) Whether or not this is literally true, it does appear that black lawyers are taken more seriously if they have elite educational credentials. (83)

Sander seems to argue that affirmative action in the job market will make up for any problems that black lawyers encounter as a result of eliminating affirmative action in law school. According to Sander, "the strong positive coefficient associated with black lawyers in our regression shows that the legal market as a whole is more willing, not less willing, to hire blacks into good jobs" and that it is "obvious" that they would continue to do so in the absence of affirmative action. (84) Needless to say, justifying the elimination of affirmative action in one realm by assuming its continuation in another is ironic, to say the least. I return to this irony in Part IV. But once we inspect more closely the evidence Sander cites from the AJD Study to prove his point that blacks receive "substantial preference" in the job market, it is clear that to the extent that such a preference exists, it is largely the product of the very affirmative action programs in law schools that Sander seeks to abolish.

Sander treats the success of the black graduates from elite law schools as if this were solely a benefit to the individuals involved. This seriously underestimates the importance of developing a viable black elite. Blacks who have achieved positions of power and influence in the legal profession, most of whom, as we have seen, are graduates of elite law schools, have played a critical role in opening the doors of opportunity for other black lawyers. (85) Black partners routinely serve on hiring committees, form organizations dedicated to improving the hiring and retention of minority lawyers, and devote countless hours to mentoring black associates and law students. As many blacks have left law firms for in-house counsel positions, these newly minted purchasing agents have sought to use their new positions of authority to distribute work to minority lawyers and to press firms to report their progress on diversity and improve upon it.

These efforts have improved the job prospects of every black law student. Predictably, recent black graduates from elite schools have benefited the most from these efforts by their alumni predecessors. This, of course, is simply a reflection of the many benefits of joining the elite networks to which such graduates automatically belong. But black lawyers from lower-ranked schools are also better off as a result of the doors that have been opened by elite black lawyers. Black partners have been at the forefront of pushing law firms and other employers to look beyond the graduates of top law schools when recruiting black students. Employing a variation of Willie Sutton's famous quip about the reason why he robbed banks--"it's where the money is!"--black partners and general counsel have vigorously pressed firms to recruit at historically black law schools and to attend minority job fairs in order to increase the number of black applicants that they see. At the same time, black general counsel have worked to ensure that minority law firms gain access to at least some corporate business and that blacks in small- and medium-sized firms are included as well. These efforts have been critical in producing the improvement in the earnings prospects of black graduates from lower-ranked schools identified in the AJD Study.

The examples Sander cites clearly demonstrate this effect. Sander argues that "there is no observable difference in outcomes" for black graduates with GPAs of 3.5 or higher, regardless of whether that student graduated from "NYU or Northwestern, at the elite end, or schools such as Howard, Texas Southern, or Santa Clara University, on the low-prestige end." (86) If this is true, it is almost certainly the product of the tireless efforts of the past beneficiaries of affirmative action. In 1993, for example, Texas Lawyer ran an article entitled Perceived Notions: Recruiters Pigeonhole TSU Graduates, Leaving Top Students in the Cold. (87) The article bemoans the fact that "firms that talk grandly about increasing their minority hiring don't look to do it at [Texas Southern University's] law school, where half of the 587 students are African American...." (88) As proof of this neglect, the article cites a TSU student who did not receive a single offer from any large law firm in Texas even though she graduated "magna cum laude, second in her class, [was a] law review editor, and ... clerk[ed] at the Texas Supreme Court." (89) Revealingly, when questioned about why the young woman failed to attract his firm's interest, one hiring partner explained "we usually can fill our needs at the top schools." (90) Clearly, there was no "preference" operating in favor of top black graduates from lower-ranked schools in 1993. (91) To the extent that this reality has changed--and remember, we have only two years of salary data to suggest that it has--it is almost certainly the result of the actions of black graduates from "the top schools" who have insisted that firms look beyond their ranks to fill their hiring needs.

Indeed, the presence of black lawyers in the nation's legal, business, and government elites confers benefits to the black community as a whole--and to our nation and to the world--that extend far beyond the important but arguably parochial employment prospects of other black lawyers. Sander alludes to the collateral costs of affirmative action: "the national competition between Democrats and Republicans, ... interracial goodwill, the belief held by whites that they are 'already' making sufficient sacrifices for the cause of racial justice, and the credibility of institutions that are often trapped in deceptions by their own policies." (92) He fails, however, to mention any of the collateral benefits associated with creating a significant, although still fledgling, black presence among the nation's elite. Black graduates of elite institutions, virtually all of whom Sander would classify as beneficiaries of affirmative action, have played pivotal roles in increasing home ownership in minority communities, (93) reducing employment discrimination in large corporations, (94) making AIDS drugs more affordable in South Africa, (95) and, as we recently saw in the case of Barack Obama's stirring speech at the Democratic National Convention, reminding all Americans of the ties that bind them, regardless of race, religion, or political affiliation. (96) To fail to account for these and other similar benefits is to ignore the extent to which our profession and our country have been fundamentally transformed for the better in the last four decades.

Of course, one can acknowledge all of these benefits and still credit the weaker form of Sander's argument that affirmative action is no longer necessary for the continued production of these gains. Once again, it is important to emphasize that unlike the benefits chronicled above, any such claim rests on little more than speculation and hope. The speculation is that the graduates of lesser-ranked schools will be at least as successful as their predecessors from more highly ranked schools at building long-term careers even though they lack the multiple and reinforcing benefits that such a degree provides. (97) The hope is that those who have already benefited from affirmative action will continue to be successful in a world in which their own credentials are subject to even greater attack. In Part IV, I will offer some additional reasons to doubt both propositions. For now, however, I simply want to maintain that neither the income data in the AJD Study, nor any of the other studies that Sander cites, come anywhere close to proving his sweeping claim that the "conventional wisdom that elite schools are the only path to coveted jobs" is false. In the absence of such proof, it is hard to see why black students who are currently admitted to schools in the first three tiers of the law school hierarchy should be willing to trade in the proven benefits of their elite status for the speculative gains from improving their grades that Sander promises.

The same is true of the psychological benefits that Sander argues blacks would gain in a world without affirmative action. Although he initially discounts such "soft" issues, a large part of why Sander believes that blacks will be better off under the regime he proposes has to do with the benefits to black lawyers of living in a world in which their success will not be discounted by presumptions about affirmative action. The benefits of such a world should not be minimized. (98) Whether blacks are likely to receive these benefits in the world Sander anticipates, however, is a much more difficult question. Although affirmative action undoubtedly contributes to the feeling among many whites that blacks are less competent, both the lessons of history and the findings of contemporary research suggest that the roots of these perceptions run far deeper in the human psyche than any set of policies or practices of specific institutions. At the same time, the socialization, networking, and, to a lesser but still significant extent, credentialing benefits of attending an elite school exist independently of whether one is or is not considered an affirmative action hire. Given these two dynamics, it is speculative at best whether black graduates from top schools would be better off--the standard Sander sets for himself--in a world in which some whites might have a harder time consciously dismissing their competence (and in which some blacks might find it easier to have confidence in themselves), but in which black lawyers also have fewer protections against the unconscious stereotypes and biases that are likely to remain.

Although the calculation is admittedly more complex for blacks who attend lower-status schools, Sander has still not satisfied his burden of persuasion that the costs of affirmative action to this less-privileged group outweigh its benefits.

III. JUST MAKING IT

Sander has performed a valuable service by focusing attention on the fate of black graduates from local and regional law schools. This group is largely absent from the writings of academics studying diversity in the legal profession, including my own. As Sander makes clear, this is an important omission. A majority of black law students--although as we have seen, not as large a majority as Sander's analysis might lead one to believe--attend such schools. If affirmative action is harming most of these students, then this should properly count as a significant cost of how we currently conduct law school admissions.

Sander produces strong evidence to support his claim that this is indeed the case. Attrition and bar failure rates are alarmingly high among this group. (99) Yet before we conclude that even these students are harmed by affirmative action, it is important that we both take a closer look at some of the statistics Sander proffers, and at the same time step back from these numbers to observe the general fate of those likely to attend lower-ranked law schools. When we do, we see that it is far from clear that the majority of black law students in this most vulnerable group are harmed more than they are helped by affirmative action--which is, it bears repeating, Sander's sole claim.

Let's begin at the bottom. Sander concedes that, under his plan, 14% of all the black students currently admitted to law school would no longer be eligible for admission to any school. (100) This amounts to 524 blacks who currently have the chance of becoming lawyers but who would no longer have this possibility under Sander's proposal. Sander dismisses this cost on the ground that these students "have such weak academic credentials that they add only a comparative handful of attorneys to total national production." (101) Sander's own calculations, however, suggest that this dismissal is far too quick. Notwithstanding their comparatively weak entering credentials, 65% of matriculants in this category graduate from law school. Twenty-nine percent eventually pass the bar and become lawyers. (102) This translates into 152 practicing lawyers who will never have an opportunity to be lawyers under Sander's plan.

This is not a trivial loss. For starters, it is only seventeen short of the number of new lawyers that Sander claims his plan will produce. (103) Thus, even if one credits all of Sander's admittedly optimistic assumptions, his plan loses almost as many black lawyers out of the bottom of the distribution as it gains from increasing bar passage rates for those in the middle. (104) Moreover, even if one were to believe that it was a net gain to society, or even to the black community, to replace the 152 black lawyers excluded by Sander's plan with the 169 blacks gained by eliminating affirmative action--a question I address below--there can be no question that with respect to those blacks who will no longer have a chance to become lawyers, the question of their exclusion is not trivial to them. Once again, in a proposal that claims to value blacks "as individuals," this cost certainly cannot be overlooked.

Sander is likely to respond that he is valuing the humanity of the blacks in this part of the distribution precisely by not allowing them to embark on the long, difficult, and, as Sander rightly emphasizes, often expensive path of becoming a lawyer when they have only a three out of ten chance of actually achieving their goal. (105) Upon close inspection, however, even this well-intentioned concern rings hollow given the alternative choices blacks in this part of the distribution are likely to have. Although less than one-third of those excluded by Sander's plan currently become lawyers, almost two-thirds become law school graduates. This is a credential that we should not assume to be worthless.

One way to remember this is to notice that, according to Sander's estimate, approximately seven percent of law school graduates never even bother to take the bar. (106) For some of these students, this was undoubtedly their plan from the start. For others, the decision to forgo the potential benefits, but certain costs, of taking the bar exam is the result of a complex mix of events that transpire during law school, including changes in family status, evolving interests, and perceptions about the job market. (107) In either case, the education--and the credential---can still be quite valuable. As we in the academy continually assert, legal education is good preparation for a wide range of activities that go far beyond the practice of law. Even if one dismisses much of this rhetoric as self-serving, there are good reasons to suspect that having a law degree is valuable in a range of jobs--real estate, insurance, law enforcement, human resources, and, of course, the business of law itself (i.e., paralegals, court clerks, court reporters, etc.)--that do not require one to be a member of the bar. In addition to gaining substantive knowledge and a visible credential that employers are likely to value, law graduates working in these fields will also benefit from having formed relationships with classmates who do become lawyers, many of whom will be in a position to assist their law-related careers. Given these potential benefits, it is far from clear that the 188 blacks with index scores under 480 (Sander's cutoff point for being admitted to law school under his plan) who currently graduate from law school but who do not pass the bar would be better off in a world in which they were not allowed to enroll at all.

This is particularly true in light of the fact that the other avenues for social and economic advancement open to this group are likely to be equally problematic. It is possible, of course, that blacks with low index scores have simply missed their true calling in medicine, the arts, or some other promising route to a high-paying and high-prestige career. It seems more likely, however, that these are students who will face long odds no matter what they attempt. (108) The real question from the point of view of students in this position--once again, the sole criterion Sander sets out for himself--is whether the risks and benefits associated with going to law school are better or worse than the reward structure of the alternatives that this group is likely to have either elsewhere in higher education or directly in the job market.

Needless to say, I do not have the information necessary to answer this question here. Given what we know about the earnings of black college graduates, particularly those from working class backgrounds whose degrees are not from elite colleges or universities, it seems wrong to presume that even a fully informed, rational, utility-maximizing black student in this position would not consider law school a relatively good bet. According to the AJD Study, even a student graduating at the bottom of his or her class in a Tier 4 law school and entering private practice can expect to earn $50,000 a year within two years of graduation. (109) Although a far cry from the princely sums paid to first-year associates at top law firms, it still compares quite favorably to entry-level salaries for college graduates in many fields. Just to put the matter in some perspective, the median salary for all black college graduates one year after graduation in 1999-2000 was $30,506. (110) For those blacks with low entering credentials who go on to beat the odds and not only become lawyers but build successful careers in areas such as personal injury law or worker's compensation, the gap in earning potential between college and law school graduates over the course of a career can be substantially higher. Even those who fail to become lawyers may receive an important earnings boost over the course of their careers from the skills, credentialing, and relationships they received from going to law school. (111) Ex ante, these rewards may outweigh the risk that a black matriculant will end up with neither a bar card nor a law degree for the time, effort, out-of-pocket expenses, and forgone opportunities that they invested in trying to obtain these goods.

Indeed, it is hard to see how one can answer the question of whether law school is a good investment for black students enrolling in schools at the bottom of the law school hierarchy without knowing much more than we currently do about whether such schools are a good investment for any student. As both the Bar Passage Study and AJD Study make clear, students enrolled in such schools are less likely to graduate and pass the bar and earn substantially less than their peers from more highly ranked institutions. And while some go on to build successful and financially rewarding careers, by most accounts the life of the average solo practitioner or "bread and butter" tort lawyer is getting increasingly difficult with each passing year. (112) Whether law schools in this range remain a good deal in such an environment is something that no one really knows--least of all the schools whose survival depends on persuading a steady stream of new recruits to believe that it is. As the AJD Study follows the class of 2000, we hope to shed some light on this and other similarly vexing questions about how the market for lawyers works at various levels of this increasingly stratified profession. Until we do, it will be very hard to untangle the fate of black lawyers (or law school graduates) from local and regional schools from a general assessment of the career returns from attending such institutions for all students.

Finally, it is impossible to assess Sander's proposal without pausing to consider the validity of the bar exam itself. As the above analysis indicates, the biggest obstacle faced by blacks in the part of the distribution Sander seeks to eliminate is the bar. Approximately fifty percent of the blacks in this group who graduate from law school and who take the bar never pass it. (113) Indeed, bar passage rates are at the core of the entire story Sander tells. By his own analysis, the number of blacks graduating from law school and taking the bar would decrease by 6.8% under his proposal. It is only because he predicts that eliminating affirmative action will significantly increase the number of blacks who pass the bar that he is able to assert that the overall production of black lawyers would increase if affirmative action were eliminated.

Once we are clear on the mechanism by which Sander hopes to expand the size of the black bar, it is evident that there is another way that one might achieve the same objective. Rather than cutting off the opportunity that those with low index scores currently have to go to law school, we could instead change the bar exam so that more black law school graduates in this part of the distribution are able to pass it. Needless to say, for critics of affirmative action, even uttering such a proposal will serve as confirmation of the implicit message--although, as I have repeatedly emphasized, not the explicit argument--of Sander's analysis: i.e., that regardless of whether affirmative action helps blacks seeking a legal education, it harms the legal profession and those it serves by allowing unqualified blacks to become lawyers. A full assessment of this critique would take me far beyond the scope of this brief Response--and beyond the scope of the criteria Sander posits for judging his claim. Given that so much of Sander's proposal turns on bar passage rates, however, it is important to say a few words about the implicit assumption that bar exam scores are a good proxy for a law school graduate's fitness to practice law. Although any argument that such exams have no relationship to quality would certainly strain credulity--just as it would be ridiculous to claim that law school grades have no predictive value in judging a student's future ability to practice law--the assumption that bar passage (let alone one's score on the bar exam) is synonymous with legal competence has not only never been proven but is, given what we know about the exam and its history, implausible on its face.

In light of the history recounted in Part I, one does not have to go as far as George Bernard Shaw's famous quip that "all professions ... are ... conspiracies against the laity" to conclude that bar examinations have as much to do with suppressing competition as ensuring competence. (114) The statistics Sander cites toward the end of his article about falling bar passage rates in recent years are strong evidence that the attempt to use the bar examination as a means of restricting competition is not merely a thing of the past. Sander notes that between 1994 and 2003, "average bar passage rates across American jurisdictions dropped as many states raised the passing threshold." (115) Given everything else he has said in his article, this is a startling trend. Just two sentences above this quotation, Sander notes that in the last decade it appears that the black-white credentials gap for law school applicants has significantly narrowed "from about 170 points in the early 1990s to perhaps 130 or 140 points now." (116) In other words, black applicants, who Sander has spent much of the last hundred pages asserting disproportionately come from the bottom of the credentials pool, are significantly better qualified by these measures than they were a decade ago. Indeed, as Sander makes clear at the beginning of the article, the entire population of law students is, on average, significantly better qualified than it was during the profession's "golden age." As he notes, "[b]y the late 1960s and early 1970s, admission to many law schools had become dramatically more competitive." (117) If anything, this trend has only accelerated in the intervening three decades. With the exception of a brief downturn during the 1991 recession and the "dot com" boom of the late 1990s, applications to law school have steadily increased from just over 60,000 in 1984 to over 100,000 today. (118) Although admissions have crept up as well, the conclusion is nevertheless inescapable that the qualifications of entering law students are higher than they have ever been before. (119)

So why then are bar passage rates decreasing? Although one could undoubtedly spin many complex scenarios to explain this trend, in this case the simple market explanation appears to be the most persuasive. Faced with a legal profession that has quadrupled in size since 1960, (120) lawyers in many jurisdictions have simply turned to lowering bar passage rates as a means of protecting themselves and their incomes from competition from new entrants. Bar passage rates have dropped sharply in states with desirable legal markets while remaining stable or actually rising in jurisdictions where lawyers are less likely to want to practice. (121) At the same time, several states have recently instituted measures to make it more difficult for lawyers who are fully licensed in one jurisdiction to practice law in another. These measures range from making lawyers who want to relocate from one state to another retake the entire bar exam to enforcing unauthorized-practice rules against out-of-state lawyers even if all they are doing is counseling an in-state client about a transaction that may not even take place within the protesting state's borders. (122) It is hard to view these trends as anything other than an attempt by in-state lawyers to protect themselves from competition from lawyers seeking to move to more lucrative markets. (123) Although these attempts have not always been successful, they nevertheless cast serious doubt on any attempt to equate bar passage with legal competence. (124)

This conclusion is particularly compelling in light of the fact that the bar exam has never been validated as a good predictor of whether a candidate will become a competent or ethical practitioner. (125) Sander spends a great deal of time demonstrating that there is a positive correlation among entering credentials--most notably the LSAT--law school grades, and bar passage. (126) Accepting arguendo that all of these connections are as strong as Sander suggests, (127) it says nothing about whether the bar exam is correlated in any significant way with competence or success in practice. Yet, Sander routinely assumes that these correlations are not only self-evidently true but are highly probative and operate in a continuous and linear manner--i.e., that the higher one's score on the bar exam, the better lawyer one is likely to become.

There is not a shred of evidence that this is true. Even if we concede that the substantive knowledge, writing skills, and analytic ability that bar exams seek to measure are relevant to legal competence, a moment's reflection makes it clear that the bar is less likely to measure these qualities accurately--let alone the full range of competencies that go into making a good lawyer--than even the typical law school exam, which, as we have seen, does so only imperfectly. One way to see this is to notice that the best law schools in the country almost never test their students by means of the kind of multiple choice questions that form the core of every bar exam. Indeed, as Sander concedes, most top law schools don't even teach the black letter rules that dominate the bar. Many schools no longer even offer the subjects--agency and trusts and estates, for example--that are routinely covered on many state bars. Clearly, in the opinion of those schools who consider themselves to be the leaders in training top lawyers, neither the methodology nor the content of the average bar exam is essential to the task of producing competent practitioners.

Moreover, even if one is inclined to believe that being able to pass the bar is an important signal about competence, the margin by which one passes the exam is almost certainly irrelevant to any judgment about future performance. Sander argues that even if he is wrong that the total number of black lawyers will increase if we eliminate affirmative action, his plan should still be adopted because it will "change dramatically ... the academic preparation of those blacks who become attorneys." (128) As proof, he cites the fact that one-fifth of black practicing lawyers have failed the bar at least once and notes that it is "a statistical certainty that many blacks who pass the bar pass by very small margins." (129) "If we believe that bar exams measure anything relevant to good lawyering," he concludes, sharply raising the first-time passage rate and the scores of those who pass the exam "would be a very good thing." (130)

At no point, however, does Sander tell us why in the world this remarkable statement is likely to be true---especially with respect to those black students who are currently passing the exam on their first attempt. To be sure, there are often significant costs associated with failing the bar for the first time. (131) Helping prospective black lawyers to avoid these costs would indeed be a very good thing, although as I will suggest, there are better ways of achieving this goal than the draconian one Sander suggests. (132) Nevertheless, it is important to emphasize that even these costs are likely to be temporary and to be dwarfed by the benefit of eventually becoming a lawyer. More importantly, whatever costs there are to failing the exam the first time, there are no costs to passing the exam by a narrow margin. Indeed, one might argue that the opposite is true: those who pass by large margins have invested way too much time in studying for the exam! (133) It is simply not true that if we think the bar measures "anything at all" that we also have to believe that incremental increases in passing scores tell us anything important about competence or future success. Even if one concedes, as Sander argues in an earlier footnote, that a one-point difference in index scores is relevant to predicting law school grades, (134) the dependent variable allegedly measured by bar passage--fitness to practice law--is so much more complex to identify and predict than the grades the LSAT claims to forecast that it is simply implausible that anything like this same kind of linear relationship exists with respect to bar exam scores, no matter how great a sample size one can amass. The implausibility of any such linear correlation is, of course, precisely why no bar authority has ever asserted that the exam they administer is capable of this kind of precision.

Indeed, the only reason to be concerned about the margin by which black candidates pass the bar is that bar officials have strong incentives to steadily increase the threshold for a passing score in order to decrease the number of new lawyers coming into the marketplace. As I have indicated above, this is just what they appear to be doing. Sander hypothesizes that bar officials might be inclined to raise the threshold even higher if they were not so worried that this would push black failure rates to unacceptable levels. (135) Although some have objected to raising passing thresholds on this ground, the real reason to oppose such increases lies elsewhere. (136) Given the dramatic increase in the overall quality of the pool of test-takers--including, as Sander documents, blacks at the bottom of the distribution--if bar officials were to take such action it would provide additional evidence that their real interest is competition, not quality.

Once we free ourselves of the assumption that those who fail to pass the bar--let alone those who have difficulty passing the bar--are for this reason alone presumptively incompetent to practice law, we can imagine a number of ways to address the depressing statistics Sander cites with respect to the odds that blacks in the bottom part of the distribution face in becoming lawyers. For starters, states could simply refrain from escalating passing scores, or better yet, return them to their 1994 levels. Given that passing rates nationally have declined from 82.3% to 74.7% during this period, (137) it is quite possible that this change alone would substantially increase the production of black lawyers. (138) Second, we could encourage those black law students who fail the exam on their first attempt to retake it. According to Wightman, a significant number of blacks who fail the bar on the first attempt never take the exam again. (139) Some of these disappointed test-takers would undoubtedly pass the exam if they took it a second or third time. (140) Given that there is no reason to believe that those who fail to pass on the first attempt are by that reason alone incompetent to practice law, we should make every effort to encourage them to do so.

One way to accomplish this objective--and to raise passage rates generally--would be to guarantee that every black student in this part of the distribution is able to take a bar review course. Sander discounts the importance of test preparation courses, but the only evidence he cites for this proposition relates to the SAT. (141) Unlike the SAT, however, the bar does not pretend to be an "aptitude" test for which coaching is presumptively unnecessary or ineffective. (142) To the contrary, the bar is much more like an "achievement" test that purports to examine learned knowledge and skill. Only in this case, the knowledge that test-takers are supposed to have is often not taught in law school. (143) So where do students learn all of this? In bar review class, of course, along with a wealth of test-taking techniques, information about past exams, tips about commonly asked questions, and, perhaps most important of all, the opportunity to practice, practice, practice. As someone who owes his own bar card to BarBri, it is simply inconceivable to me that these courses do not improve one's chances of passing the exam.

We do not have good evidence about the relative access of blacks and whites to bar review courses. But even if we assume, as one study of New York lawyers suggests, that "the vast majority of law graduates" take bar preparation courses, there are still good reasons to believe that the black lawyers we are currently discussing are more likely to be in the minority of students who do not have access to this crucial resource. (144) Bar review courses are expensive, and unlike students at elite schools, the blacks in this part of the distribution will rarely have this expense paid for by their future employers. (145) If we add the plausible assumption that on average blacks who attend low-status law schools have few resources to pay for such courses on their own--particularly more than once--then it is very likely that we could improve passage rates among this group by ensuring that more of them have the opportunity to take advantage of this crucial resource. (146)

Needless to say, if we were prepared to think about modifying the bar exam, or even more radically, of eliminating it altogether (either for certain forms of legal practice or in toto), we would likely boost the number of black lawyers (or their equivalent) even further. (147) How much, of course, would depend upon the specific modifications taken and an assessment of how these changes might track performance that may be correlated with race. (148) It is possible, for example, that blacks would do better on an exam that included an element of oral advocacy or that called for team building skill or gave credit for public service. (149) Similarly, it is quite possible that many of the tasks lawyers in this part of the distribution typically undertake--real estate closings, simple wills, uncontested divorces--do not require three years of legal education followed by a two-day bar examination and character and fitness review to be performed competently and efficiently. (150)

None of this means that we should ignore the issues related to black performance in law school and on the bar exam that Sander documents. Paying attention to these issues, however, does not require that we throw out 524 affirmative action babies (to borrow Steve Carter's admittedly infantalizing term (151)), and risk damaging the career prospects of hundreds more who currently attend top schools, in order to clean up the dirty bathwater that has unfortunately too often resulted from the way that race continues to structure and cloud our thinking about how to help our profession and our nation move beyond its racist past. I conclude with a few thoughts about how to clarify our thinking without crucifying the very black lawyers who continue to bear the brunt of our collective struggle to move beyond the problem of the color line in this new century.

IV. SOME STRAIGHT TALK ABOUT STRAIGHT TALK

Rick Sander has performed a valuable--if painful--service in highlighting the fact that affirmative action entails potential risks as well as benefits for black lawyers. Even if one concedes that these risks are as great as Sander indicates, however, it is hard to see why disclosure is not the sole appropriate remedy. As an economist, Sander belongs to a profession that typically thinks that people ought to be able to make their own decisions about how to weigh the risks and benefits that they confront in their lives. To the extent that going to a more highly ranked law school, or in the case of the students at the bottom of the distribution, going to law school at all, will often produce costs (in terms of the risk of attrition, low grades, and difficulty on the bar) that outweigh the potential benefits of graduating from such an institution and becoming a lawyer, then it would seem that the solution would be to convey this information to black applicants in a form that would allow them to decide for themselves whether to take the risk.

This is indeed Sander's first line of attack. When one examines the kind of disclosure he proposes, however, it is evident that something more is at stake than simply helping black applicants make better choices about whether and where to seek a legal education. Worse yet, the manner in which Sander suggests making this disclosure seems destined to reinforce, rather than to correct, the discouragement and disengagement that Sander argues plays a crucial role in producing the disquieting statistics he cites.

Sander proposes that every law school should provide every applicant with the median index scores, class rank, and bar passage scores--broken down by race--for every applicant, student, and/or graduate. (152) Some of this data would indeed be useful for black applicants considering their chances of graduating and becoming a lawyer if they choose to attend a particular school. But other information would be equally, if not more, valuable. For example, where are graduates of this school likely to be five, ten, or fifteen years out of law school? How does this compare to where graduates of more or less highly ranked schools tend to end up? What kinds of academic support programs does a school have and how effective have such programs been in helping students with comparably low index scores improve their academic performance or pass the bar? Have there been any racial incidents at the school in the last few years and what, if anything, has the school done to assess or improve its racial climate? The point simply is that all students--especially those attending less highly ranked schools--would benefit from more information about their prospects for success in school and after graduation. Admissions data, GPAs, and bar passage rates are only a few of the numerous factors that a student would want to consider in order to make a fully informed choice.

Although more disclosure would undoubtedly be beneficial, it is not at all clear how distributing racial breakdowns of admission statistics, grades, and bar passage rates to all students will benefit black applicants or admittees, especially those who decide to brave the odds and attend a more highly ranked school than Sander's thesis would suggest that they should. (153) Rather than helping these students make a rational choice about risks and benefits, the disclosure Sander suggests seems likely to reinforce the idea, already far too prevalent among white students and faculty, that the black students who do enroll are not competent to do the work or to become lawyers. Whatever value Sander might argue that there is to demonstrating that Justice O'Connor was wrong in Grutter, or to providing white students with a road map about how affirmative action is working at a given institution, one would be hard-pressed to come up with a reason why any of this will benefit the black students and lawyers whose interests, once again, Sander asserts he is protecting.

Worse still, by focusing only on the most negative aspects of the current reality--i.e., that many black students receive low grades and have difficulty passing the bar--without giving at least equal time to the positive news that most black lawyers are leading successful and productive careers, Sander's proposed disclosure is destined to exacerbate the extent to which black law students currently feel alienated and disengaged. Sander argues that black students who find themselves in academic environments for which they are unprepared will often become "stressed out" or "disengaged" in ways that further lower their performance. (154) My interviews with black lawyers suggest that there is some truth to this theory. Several of those in my sample look back with regret on their time in law school and wish that they had taken more advantage of the opportunities that were provided. Many of those who expressed these feelings confessed that they found law school "overwhelming" and "intimidating" and that it was difficult to go from being the "golden child" who was always at the top of his or her class to someone who struggled academically.

Although this phenomenon is therefore real and important, there are three factors that Sander leaves out of his presentation that are critical to understanding this reality and what we might do to prevent it. First, it should be obvious to anyone who has ever taught in a law school, especially one that is highly ranked, that black students are far from the only ones who disengage from their studies after finding that law school is more difficult than they expected it to be. At schools such as my own, for example, it is sadly true that after first-semester grades come out, a large percentage of students do exactly what Sander suggests black students do: "blame the system--the professor, the school, or legal education generally--and . reduce [their] effort." (155) Indeed, as Sander and his coauthor Mitu Gulati discovered in their study of third-year law students, by the time many students get to this stage of their career, a very substantial percentage are no longer attending class at all. (156) As the authors conclude, this reality raises serious questions about what, if anything, we are accomplishing in the third year of law school--questions, it should go without saying, that raise issues extending far beyond the future of affirmative action.

Second, the difficulties that black students experience in law school sometimes have more to do with their failure to understand the rules of the game than the inherent difficulty of the material. Sander is correct that many black students spend just as much time studying as their white peers yet often have less to show for it. (157) He assumes that this means that these students are academically unprepared for the challenge. But there is another plausible explanation--that they do not know how to study the right way in order to maximize their chances of doing well. My interviews are filled with black lawyers who lament the fact that they had no idea how to study efficiently during their first year in law school. Some of those who fall in this category eventually did catch on and substantially improved their performance in their second and third years. As Sander predicts, however, others unfortunately became discouraged, believing that they were not going to do well in school no matter how hard they tried. What Sander leaves out of his account, however, is that in many instances, this resignation was facilitated by the low expectations of faculty and fellow students.

Sander's account of discouragement and disengagement is entirely internal. Black students in his view realize that they are overmatched and become frustrated and withdrawn, thereby diminishing their chances of succeeding even further. Strangely absent from this account is any acknowledgement of how the expectations of others contribute to this deadly cycle. Any plausible theory of motivation, however, must begin with an understanding that such feelings are profoundly interactive. People are more likely to do well when they are expected to do well. (158) More to the point, they are much less likely to succeed when they are expected to fail. (159) Unfortunately, this is precisely the message that has often been conveyed to entering black students. Sometimes the message has been express. As a respondent who entered a top ten law school in the early 1980s reports, "[T]he first week of law school.... the dean of our law school came to a BLSA [Black Law Students Association] meeting and said, 'You guys won't do as well as the other people here.' I mean, he just said it!" (160) Other times the message is the unintended but nevertheless powerful consequence of diverting all black students into academic support programs. (161) Still others have felt the sting of low expectations when professors fail to call on black students in class. (162) For another group, it has been the presumptions expressed by fellow students. (163)

There is no doubt that the very existence of affirmative action complicates all of these effects. As Sander might argue, it is precisely because of the existence of such programs that schools, professors, and students doubt the competence of incoming black students. As I have argued, research on unconscious bias suggests that the roots of these feelings extend far deeper than any particular policy or practice. For present purposes, however, the key point is that whatever one thinks about these psychological dynamics, Sander's disclosure proposals seem destined to make them worse.

Indeed, given the very "cascade effects" Sander identifies, it is hard to imagine that the implicit message of Sander's analysis--that the black beneficiaries of affirmative action are not qualified to become competent and successful practitioners--will not spread in a manner that is destined to undermine the very goal of improving the condition of the black bar that his express argument is designed to promote. Sander's assertion that eliminating affirmative action in law school will increase the production of black lawyers and improve (or at a minimum, not retard) the progress of blacks in the job market depends upon everything else staying the same. Yet given his observation that law schools are part of a connected system in which the actions of particular actors have predictable consequences on the outlook and choices of other participants, this assumption is highly dubious. At the entry level, for example, will elite colleges--or elite high schools for that matter--continue aggressively to recruit black students in a world in which law schools declare that such efforts actually harm their intended beneficiaries? Similarly, at the exit level, will law firms and other employers continue their fledgling affirmative efforts to recruit and retain black lawyers if the very institutions that produce these students announce that such policies harm blacks by placing them in situations where they will be "overmatched"? Needless to say, if affirmative action were reduced significantly at the entry level, then Sander's already optimistic projections about the number of blacks who would be admitted to law school under his proposal become even less plausible. (164) The situation would likely be even worse if legal employers follow Sander's lead and decrease their own affirmative efforts to recruit and retain black lawyers. (165) Given that there are already several efforts underway to reverse Grutter either through litigation or legislation, a realistic appraisal of the costs of implementing Sander's proposal must account for these plausible systemwide effects. (166)

Now, let me be very clear here. I am not arguing that we should hide from the troubling statistics about black performance that Sander outlines or that our entire system depends, as Sander seems to suggest, upon our collective willingness to blind ourselves to the realities of how affirmative action actually works. To the contrary, I am making a plea that any discussion about black performance must take place within a broader examination of how affirmative action--and the market for legal services generally--actually works. (167) This reality, as I have tried to suggest, is far more complex than the simplified universe of inputs and outputs that Sander's analysis seems to assume. The LSAT, law school grades, and the bar exam undoubtedly measure some qualities that are relevant to whether one is likely to become a good lawyer. To the extent that black students are performing poorly on these measures, it is a legitimate and important cause for concern. But any plan for addressing these problems must also acknowledge that the causes of these achievement gaps are multiple and complex and that their implications for legal competence are largely unknown.

What we do know is this. Affirmative action in law school admissions has played a crucial role in transforming a once exclusionary and insular profession into one that is at least tolerably diverse. Notwithstanding the difficulties Sander documents, the black lawyers who have been at the forefront of this transformation have for the most part done remarkably well--even if, like Anthony Chase, they sometimes struggle along the way. Any claim that most, or even a significant percentage, of these integration warriors would have been better off under a regime where law schools treated Bakke as an indication that affirmative efforts were no longer necessary or desirable is simply not supported by the evidence. Unfortunately, "the battle for racial inclusion" has not, as Sander asserts, "been fought and largely won." (168) Race continues to structure the opportunities and outlook of all Americans even as overt discrimination based on race recedes. Any dialogue about affirmative action, or about legal education and practice generally, must candidly acknowledge this complex reality. When we do, we are likely to be able to see problems and imagine solutions that help us to improve opportunities for all Americans.

(1.) Grutter v. Bollinger, 539 U.S. 306 (2003).

(2.) Richard H. Sander, A Systemic Analysis of Affirmative Action in American Law Schools, 57 STAN. L. REV. 367, 369 (2004) (emphasis added).

(3.) Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978).

(4.) Sander, supra note 2, at 369.

(5.) Id. at 370.

(6.) I am grateful to Judge Harry Edwards for pointing out this obvious fact. As I argue below, this will also be the case if Sander's proposal is adopted, at least with respect to the spots at the bottom of the class at highly ranked schools that will be vacated by blacks and (presumably) filled by whites. I consider the implications of this fact in Part IV.

(7.) Even this statement must be qualified. As I indicate below, some number of students enter law school with no intention of ever practicing law. See infra pp. 1943-44. Nevertheless, it is probably true that even those who do not intend to become lawyers when they graduate want to have the option of doing so that comes from graduating and passing the bar.

(8.) Sander, supra note 2, at 478.

(9.) Id. at 371-72.

(10.) Id. at 372.

(11.) The comparable number for whites is over eighty-three percent. Timothy T. Clydesdale, A Forked River Runs Through Law School: Toward Understanding Race, Gender, Age, and Related Gaps in Law School Performance and Bar Passage, 29 LAW & SOC. INQUIRY 711, 727 (2004).

(12.) Sander, supra note 2, at 474 (arguing that the 14% of black applicants who would not get into any law school under his plan have "such weak academic credentials that they add only a comparative handful of attorneys to total national production"). As I argue below, Sander's statement understates the actual importance of those in this group. Nevertheless, it is true that 70% of those who Sander's plan would exclude currently do not become lawyers.

(13.) The fact that Sander claims that his proposal would increase the number of blacks who become lawyers in any given year does not change this conclusion. Although black law students who end up being among the 7.9% of new black lawyers Sander asserts his plan would produce should count as beneficiaries, if the remaining 92% of entering black lawyers--or even a percentage substantially greater than 7.9%--would actually be harmed by Sander's proposal then it would be very difficult to conclude that "most" blacks would be helped by eliminating affirmative action.

(14.) Sander, supra note 2, at 454-68. In the interest of full disclosure, I should acknowledge that like Rick Sander, I am also one of the principal investigators on the AJD Study. For the purposes of this Response, I have neither replicated Sander's analysis of the AJD Study data nor conducted my own independent review. Instead, I rely on Sander's reporting of the data and on the information contained in the Report issued on our preliminary findings by the National Association of Law Placement Foundation and the American Bar Foundation in 2004. See RONIT DINOVITZER ET AL., AFTER THE JD: FIRST RESULTS OF A NATIONAL STUDY OF LEGAL CAREERS (2004) [hereinafter AJD STUDY].

(15.) Sander, supra note 2, at 479.

David B. Wilkins, Kirkland & Ellis Professor of Law and Director of the Program on the Legal Profession and the Center on Lawyers and the Professional Services Industry, Harvard Law School. Thanks to David Chambers, Lani Guinier, and Bill Kidder for helpful comments on a prior draft. Thomas Tso provided invaluable research assistance. I am especially grateful to my friend Anthony Chase for allowing me to use his story to begin this piece.

(16.) In 1960, there were approximately 2000 black lawyers in the United States, constituting just over one percent of the profession. See RICHARD L. ABEL, AMERICAN LAWYERS 100 (1989); Harry T. Edwards, A New Role for the Black Law Graduate--A Reality or an Illusion ?, 69 MICH. L. REV. 1407, 1410 (1971) (reporting the number of black lawyers in 1960 as 2180). This meager number was only marginally higher than the 1300 black lawyers practicing when Thurgood Marshall joined the bar in 1930.

(17.) For excellent accounts of the legal profession's history of discrimination and exclusion, see ABEL, supra note 16; JEROLD S. AUERBACH, UNEQUAL JUSTICE: LAWYERS AND SOCIAL CHANGE IN MODERN AMERICA (1976); J. CLAY SMITH, JR., EMANCIPATION: THE MAKING OF THE BLACK LAWYER, 1844-1944 (1993).

(18.) Just to hit a few of the highlights: no black received a legal education prior to the Civil War; the ABA did not admit black members until 1943; and in 1947, William Coleman could not get a job in his native Philadelphia notwithstanding graduating number one in his class at Harvard Law School, being an editor on the Harvard Law Review, and clerking for Justice Felix Frankfurter. See, e.g., RICHARD KLUGER, SIMPLE JUSTICE 292-93 (1977) (describing Coleman's experience); SMITH, supra note 17, at 541-45 (noting that the ABA did not admit blacks until 1943); J. Clay Smith, Jr., In Freedom's Birthplace: The Making of George Lewis Ruffin, the First Black Law Graduate of Harvard University, 39 HOWARD L.J. 201, 214-16 (1995) (reporting that Ruffin was the first black graduate of an American law school, in 1869).

(19.) See GERALDINE R. SEGAL, BLACKS IN THE LAW: PHILADELPHIA AND THE NATION 218 (1983).

(20.) ERWIN O. SMIGEL, THE WALL STREET LAWYER 45 (2d ed. 1969).

(21.) AUERBACH, supra note 17, at 121. The phrase comes from George Wickersham, a former U.S. attorney general and the founder of the Wall Street law firm of Cadwalader, Wickersham & Taft that continues to bear his name. See also MARCIA GRAHAM SYNNOTT, THE HALF-OPENED DOOR: DISCRIMINATION AND ADMISSIONS AT HARVARD, YALE, AND PRINCETON, 1900-1970, at 18 (1979) (quoting a dean at Columbia as boasting that the use of standardized tests had allowed the school to cut the number of Jews it admitted by half).

(22.) See Note, The Jewish Law Student and New York Jobs--Discriminatory Effects in Law Firm Hiring Practices, 73 YALE L.J. 625 (1964).

(23.) See CYNTHIA FUCHS EPSTEIN, WOMEN IN LAW (1993).

(24.) SEGAL, supra note 19, at 24 (footnotes omitted) (paraphrasing Secretary of Labor Willard Wirtz).

(25.) See MARC GALANTER & THOMAS PALAY, TOURNAMENT OF LAWYERS: THE TRANSFORMATION OF THE BIG LAW FIRM 20-36 (1991) (referring to the "Golden Age" of the large law firm "circa 1960"). In subsequent work, Galanter makes clear that unlike some who celebrate this period as a halcyon time, any realistic appraisal of the profession's past must account for the many not-so-golden exclusionary practices that were the hallmark of the era. See Marc Galanter, Lawyers in the Mist: The Golden Age of Legal Nostalgia, 100 DICK. L. REV. 549 (1996) (reviewing MARY ANN GLENDON, A NATION UNDER LAWYERS (1994)).

(26.) SMIGEL, supra note 20, at 37.

(27.) See id. at 106. As one partner observed:

   There are intangibles. We see a man for long hours over the years,
   see his wife, know his family background, what outside charity
   activities he participates in. You get to know these people over
   a ten year period. You see them in your home or when you're away
   on a trip with them--the word comes down about them from judges
   and clients. We encourage extracurricular activities. On a personal
   level, if we never see a man at functions we wonder if he has the
   qualities we want--if he measures up. He must be able to play team
   ball--if he can't we will not take him, for we are also looking for
   personal qualities, including his ability to go along with you.

Id.

(28.) It bears repeating that blacks are by no means the only, or arguably even the largest, beneficiaries of affirmative action in the last forty years. Since the 1960s, Hispanics, Asians, Native Americans, and white women were all considered part of the effort to reverse the exclusionary history described above. Like Sander, however, for present purposes I confine my analysis to the experience of black law students and lawyers.

(29.) Sander, supra note 2, at 377-80.

(30.) I recognize that the Supreme Court rejected this justification for affirmative action in Bakke. As I indicated above, however, notwithstanding his long discussion about Grutter, Sander is not making an argument about constitutional law. Indeed, Sander is quite clear that he rejects the primary argument advanced by opponents of affirmative action in Bakke that such policies are unconstitutional because they harm the interests of whites. To the contrary, Sander argues that whites are actually benefited by affirmative action because they are protected from harder grading and lower bar passage rates. I return to these arguments below.

(31.) See U.S. Census Bureau, Census 2000 Special EEO Tabulation, at http:// www.census.gov/eeo2000/index.html (last visited Apr. 23, 2005) (reporting that there were 33,865 black, non-Hispanic lawyers in 2000). Given that Sander projects that approximately two thousand new black lawyers enter the bar every year, the current size of the black bar is undoubtedly closer to forty thousand. Sander, supra note 2, at 473.

(32.) See generally David B. Wilkins, From "Separate Is Inherently Unequal" to "Diversity Is Good for Business": The Rise of Market-Based Diversity Arguments and the Fate of the Black Corporate Bar, 117 HARV. L. REV. 1548 (2004).

(33.) As Sander acknowledges, the two such studies that are most important are WILLIAM G. BOWEN & DEREK BOK, THE SHAPE OF THE RIVER: LONG-TERM CONSEQUENCES OF CONSIDERING RACE IN COLLEGE AND UNIVERSITY ADMISSIONS (1998), and David L. Chambers et al., Michigan's Minority Graduates in Practice: The River Runs Through Law School, 25 LAW & SOC. INQUIRY 395 (2000). In 2000, I conducted a comprehensive survey of all living black Harvard Law School graduates (n = 657) which also concluded that these graduates were quite successful in their careers. DAVID B. WILKINS ET AL., HARVARD LAW SCHOOL: REPORT ON THE STATE OF BLACK ALUMNI, 1869-2000 (2002). To be sure, each of these studies has weaknesses. I return to some of these below. See infra note 40. Nevertheless, it is worth noting that there has never been any comprehensive investigation of the careers of the black recipients of affirmative action that has reached a contrary conclusion. Research about the long-term effects of affirmative action in other fields is consistent with this conclusion. See Harry Holzer & David Neumark, Assessing Affirmative Action, 38 J. ECON. LITERATURE 483, 559 (2000) (finding on the basis of a thorough review of available studies that there is "very little compelling evidence of deleterious efficiency effects of affirmative action" and that "the empirical case against affirmative action on the grounds of efficiency is weak at best").

(34.) Sander is ambiguous about whether he is claiming that affirmative action has always harmed black law students (or harmed them since the "second phase" of these programs after Bakke) or that, whatever value these programs had in the past, such efforts are now unnecessary or counterproductive. Given his criticism of law school admissions in the years since Bakke and his extensive reliance on the Bar Passage Study, which chronicles a class of law students who have now been in practice for more than a decade, it seems fair to charge him with the stronger claim that affirmative action has been harmful to black lawyers for some time. In order to remove any ambiguity on my part, however, I have tried to be clear throughout whether I am addressing the strong or the weak form of Sander's argument.

(35.) See, e.g., David B. Wilkins, Doing Well by Doing Good? The Role of Public Service in the Careers of Black Corporate Lawyers, 41 HOUS. L. REV. 1 (2004); Wilkins, supra note 32; David B. Wilkins, Partners Without Power?: A Preliminary Look at Black Partners in Corporate Law Firms, 2 J. INST. FOR STUDY LEGAL ETHICS 15 (1999); David B. Wilkins, Rollin' on the River: Race, Elite Schools, and the Equality Paradox, 25 LAW & Soc. INQUIRY 527 (2000) [hereinafter Wilkins, Rollin' on the River]; David B. Wilkins & G. Mitu Gulati, Why Are There So Few Black Lawyers in Corporate Law Firms?: An Institutional Analysis, 84 CAL. L. REV. 493 (1996); David B. Wilkins, Book Review, On Being Good and Black, 112 HARV. L. REV. 1924 (1999) [hereinafter Wilkins, Book Review] (reviewing PAUL M. BARRETT, THE GOOD BLACK: A TRUE STORY OF RACE IN AMERICA (1999)). The arguments presented in these articles are expanded and illuminated by material from over two hundred in-depth interviews with black lawyers in my forthcoming book, The Black Bar: The Legacy of Brown v. Board of Education and the History of Race and the American Legal Profession (forthcoming 2005). The citations to interviews below come from that project. In order to preserve confidentiality, respondents are identified only by number and the date of the interview.

(36.) See John F. Dovidio et al., On the Nature of Prejudice: Automatic and Controlled Processes, 33 J. EXPERIMENTAL SOC. PSYCHOL. 510, 517 (1997); Russell H. Fazio et al., Variability in Automatic Activation as an Unobtrusive Measure of Racial Attitudes: A Bona Fide Pipeline?, 69 J. PERSONALITY & SOC. PSYCHOL. 1013, 1022 (1995); see also Mahzarin R. Banaji, The Opposite of a Great Truth Is Also True: Homage to Koan #7, in PERSPECTIVISM IN SOCIAL PSYCHOLOGY: THE YIN AND YANG OF SCIENTIFIC PROGRESS 127 (John T. Jost et al. eds., 2003); Nilanjana Dasgupta et al., Automatic Preference for White Americans: Eliminating the Familiarity Explanation, 36 J. EXPERIMENTAL SOC. PSYCHOL. 316, 324-25 (2000); William von Hippel et al., The Linguistic Intergroup Bias as an Implicit Indicator of Prejudice, 33 J. EXPERIMENTAL SOC. PSYCHOL. 490, 507 (1997).

(37.) See, e.g., David A. Thomas, Racial Dynamics in Cross-Race Developmental Relationships, 38 ADMIN. SCI. Q. 169 (1993); see also John F. Dovidio, On the Nature of Contemporary Prejudice: The Third Wave, 57 J. SOC. ISSUES 829, 844-45 (2001).

(38.) Sander, supra note 2, at 456.

(39.) Id.

(40.) See, e.g., Chambers et al., supra note 33, at 501 (finding no correlation between undergraduate grades and future income and only a modest relationship--explaining approximately 5% of the variation--between law school grades and future income); James B. Rebitzer & Lowell J. Taylor, Efficiency Wages and Employment Rents: The Employer-Size Wage Effect in the Job Market for Lawyers, 13 J. LAB. ECON. 678, 690 (1995) (reporting that in their analysis of partner income, variables corresponding to law school prestige, law review membership, and top law school grades "are not statistically significant at the 5% level"). Neither of these comparisons proves that law school grades or other standard entry-level credentials have no effect on future earnings, let alone on the more general issue of career success. To name only the most obvious limitation, both studies suffer from classic "restricted range" problems. The Michigan study is only relevant to the career success of Michigan graduates. Given the credentials of most law firm partners, the Rebitzer and Taylor study is also only measuring a relatively narrow range of potential entering credentials. Indeed, given the description of how elite employers hire and retain lawyers contained in the next Part, it would be quite surprising if there were no correlation between these traditional entering credentials and career success. My point simply is that whatever relationship exists is likely to be much more attenuated--and much less independently grounded (since the value of grades and law school status reside as much in the fact that they are easily visible and rankable signals that reinforce the self-perceptions of those applying them)--than Sander's easy equation of grades with "cognitive skill and subject mastery" would lead one to believe. Sander, supra note 2, at 456. At a minimum, neither study--nor any other of which I am aware--has ever established the kind of robust relationship between grades and career success that would justify the power that these traditional credentials are often assumed to have. See also Lani Guinier, Lessons and Challenges of Becoming Gentlemen, 24 N.Y.U. REV. L. & Soc. CHANGE 1, 12 n.37 (1998) (reporting on a study done by a prominent law firm that found, with the exception of those at the very top, that there was little correlation between grades and quality with respect to the lawyers in the firm); Luariz Vold, Legal Preparation Tested by Success in Practice, 33 HARV. L. REV. 168, 174-75 (1919) (study of all those admitted to the North Dakota bar between 1902 and 1913, comparing success in court to law school grades, and finding that the highest-ranked students academically were less successful in court than the next lower tier of students). Research in other fields is consistent with this conclusion. See Holzer & Neumark, supra note 33, at 544 (noting that while "[t]here is some evidence of lower qualifications for minorities hired under affirmative action programs," "[e]vidence of lower performance among these minorities appears much less consistently or convincingly").

(41.) Once again, research on the continuing existence of discrimination in labor markets generally is consistent with this conclusion. See Holzer & Neumark, supra note 33, at 499 ("Taken together, the various studies summarized above suggest that, while differences in educational attainment and cognitive skills account for large fractions of racial differences in wages, employer discrimination continues to play a role in generating different labor market outcomes by race and sex.").

(42.) Sander, supra note 2, at 373 ("[B]lack graduates at Harvard and Yale have their pick of jobs.").

(43.) Sander's concession with respect to graduation rates for blacks at top schools is clear. See id. at 437 (noting that over 95% of blacks attending the most elite schools graduate); see also UCLA Law Professor Proposes to Bump Black Law Students Down into Less Selective Law Schools, J. BLACKS HIGHER EDUC., Autumn 2004, http://www.jbhe.com/ news_views/45_sander_lawschool.html (reporting, in a survey of twenty-six highly ranked law schools, that black graduation rates average better than 90%--including at nine schools such as Columbia, Georgetown, and the University of Michigan, which report three-year graduation rates of 100%--and that the rates at top schools like Harvard, Stanford, and Yale are all above 90%). As former Stanford dean Kathleen Sullivan notes in the Journal of Blacks in Higher Education article, when black students do not finish in three years, it is often due to a family emergency or other temporary situation, and the student eventually returns to school to complete his or her degree. Id. (quoting Sullivan as concluding "[o]ur record is, in fact, pretty much perfect"). It is important to note that these graduation rates appear higher than those in the Bar Passage Study. This is another indication, as Sander concedes, that black performance has improved since 1991. Sander is more equivocal with respect to bar passage rates among this group. Thus, he concedes that controlling for all other variables, students at more highly ranked schools have higher bar passage rates. Cf Sander, supra note 2, at 449. He also asserts, however, that blacks who attend more highly ranked schools than their entering credentials warrant will have lower passage rates as a result of getting low grades. He never tests this proposition directly, however, with respect to black graduates of top schools nor reports black passage rates by school tier.

(44.) Sander, supra note 2, at 465.

(45.) Id. at 466.

(46.) Id. at 465-66.

(47.) Id. at 483. This number is almost equal to the number of black students who enrolled in such institutions in 1991. See id. at 415 tbl.3.1 (showing 147 blacks in top ten schools). Since the number of blacks attending law school has not doubled during this period, this suggests that the concentration of black students in top schools may be even greater today than it was in 1991.

(48.) Id. at 454.

(49.) I arrive at this number by adding the blacks in the top ten schools in Table 3.1 (147), id. at 415 tbl.3.1, to the number in other "national" schools (272) in Table 5.3, id. at 431 tbl.5.3.

(50.) See id. at 416 (noting that the "seven law schools that have historically served minorities" are "obviously a special case"). As many have observed, many black students choose to attend historically black colleges and law schools even though they could attend more highly ranked schools. See generally JULIAN B. ROEBUCK & KOMANDURI S. MURTY, HISTORICALLY BLACK COLLEGES AND UNIVERSITIES: THEIR PLACE IN AMERICAN HIGHER EDUCATION (1993).

(51.) Sander, supra note 2, at 461 n.261 (defining law school categories). If we take as a rough guesstimate that there are an additional 100 black students in the schools ranked thirtieth through fortieth, somewhat fewer than the thirteen blacks per school in the top thirty, the black enrollment in the top forty schools would be 519, or roughly 36% of total black enrollment in just over 26% of law schools, excluding those that are historically black.

(52.) Id. at 464 tbl.7.4. Although the p-value of school status for Tier 3 schools (.006) is substantially greater than it is for the first two (< .0001 for Tier 1 and .0002 for Tier 2), as Sander acknowledges earlier, anything below .05 is typically considered highly significant. See id. at 438 n.190.

(53.) The fact that elite schools have always led with respect to issues of race increases the risk that black students will fall farther than Sander predicts. If schools at the top are perceived as reducing their commitment to affirmative action, and to diversity more generally, it is quite possible that those in the lower tiers will do so as well. Today, having a reasonably diverse student body is generally considered a positive credential in assessing school status. If the top schools were to lose half or more of their black students, the relationship between diversity and eliteness might drift back to where it was in the days before affirmative action, particularly among schools that like to think of themselves as being more elite than they are in fact. In such a world, even though blacks who are currently attending Tier 2 schools may be eligible to be admitted to schools in Tier 4, they may not in fact be admitted since all law schools, as Sander makes clear, have many more qualified applicants than they have places to offer them. Given the substantial earning differences between Tiers, black students from Tier 2 schools who fall to Tier 5 schools (or students who fall from Tier 1 to Tier 4) will not receive the benefits Sander posits.

(54.) Indeed, even if blacks improve their grades by as much as Sander predicts, so long as those grades remain average, they are unlikely to produce the payoff Sander suggests. Sander assumes that there is a continuous trade-off between grades and income. Sander, supra note 2, at 465 ("[O]ne standard-deviation improvement at a school produces, on average, a 12.3% rise in earnings."). This is not, however, how law firms tend to think about grades. Rather than operating on the basis of a continuous curve, most law firms employ a set of grade cutoffs that are expressly tied to school status to guide their hiring decisions. By all accounts, these cutoffs rise sharply once one moves outside of the top schools. While students at top schools will be considered--although they may not be hired--if their grades are anywhere near the respectable range, those from lower-tier schools must be at the top of their class. Moreover, these cutoffs are also more rigidly enforced at lower-tier schools, since schools in this range routinely allow firms to prescreen resumes in order to ensure that applicants in fact meet their hiring criteria. In our survey of third-year law students, Mitu Gulati and I found that the students at both high- and low-status schools are well aware of the implications of these practices. See David B. Wilkins & G. Mitu Gulati, What Law Students Think They Know About Elite Law Firms: Preliminary Results of a Survey of Third Year Law Students, 69 U. CIN. L. REV. 1213, 1232-34 (2001) [hereinafter Wilkins & Gulati, Third Year Law Students] (reporting that elite school respondents believed that law firms value law school prestige significantly more than grades, whereas those at less elite schools came to the opposite conclusion and believed that they needed significantly better academic credentials to be hired by a large law firm). Thus, even if a black student who is currently near the bottom of her class at Vanderbilt would finish in the middle of her class at the University of Tennessee, to borrow one of Sander's examples, it is far from clear that she would make the grade cutoff employed by the smaller number of top firms that interview at the latter campus. I return below to Sander's implicit claim that affirmative action in hiring will change this dynamic.

(55.) There are good reasons to suspect that Sander underestimates even this benefit. Although getting a low grade is certainly an indication that one did not learn all that one could have in a given class, it does not mean that no learning took place at all. Given that top schools are likely to have significantly greater resources to teach a broader range of subjects than their less elite competitors--and to teach common subjects in a more advanced and sophisticated manner--even a student who is performing poorly may nevertheless be acquiring valuable knowledge in the classroom that he or she might not get at a lower-ranked school. Indeed, the very study Sander cites to support his claim that blacks do better at less elite schools also finds that schools with greater resources increase long-term earnings:

   We do find that students who attend colleges with higher average
   tuition costs tend to earn higher income years later, after
   adjusting for student characteristics.... Because tuition and
   expenditures per student are positively correlated, these
   results suggest that tuition matters because higher cost schools
   devote more resources to student instruction.

Stacy Berg Dale & Alan B. Krueger, Estimating the Payoff to Attending a More Selective College: An Application of Selection on Observables and Unobservables, 117 Q.J. ECON. 1491, 1524 (2002), cited in Sander, supra note 2, at 453 n.242. Significantly, Dale and Krueger find that these benefits are especially important for students from low-income backgrounds. See id. at 1525 (finding "that the returns to school characteristics such as average SAT score or tuition are greatest for students from more disadvantaged backgrounds"). As I indicate below, many black students fall into this category. The evidence Sander relies on from Linda Loury and David Garman, see Linda Datcher Loury & David Garman, College Selectivity and Earnings, 13 J. LAB. ECON. 289 (1995), cited in Sander, supra note 2, at 451 n.225, also does not establish that blacks do not benefit from being admitted to more selective schools. See Holzer & Neumark, supra note 33, at 549 (discussing the Loury and Garman study and concluding that "blacks with lower SAT scores are not necessarily worse off when they are admitted to a more selective college than they would have otherwise been; instead, their average gain from having been admitted is not as high as that observed among whites"). Ian Ayres and Rick Brooks make a similar argument (much more rigorously) in their response to Sander in this issue. See Ian Ayres & Richard Brooks, Does Affirmative Action Reduce the Number of Black Lawyers?, 57 STAN. L. REV. 1807, 1825-26 (2005).

(56.) ROBERT GRANFIELD, MAKING ELITE LAWYERS: VISIONS OF LAW AT HARVARD AND BEYOND (1992). I often tell students that this is the best book written about Harvard Law School that no one has ever read.

(57.) Id. at 123-67 (arguing that professional socialization encourages students to suppress competition for grades while at the same time leads them to transform their initial interest in public service and social justice into a belief that the best and perhaps only way to further these ends is to take a job in a large corporate law firm).

(58.) Although schools outside of the top tier do this as well, because of their broader alumni base and greater resources, it is safe to say that elite schools are able to engage in more of this kind of socialization than their less highly ranked peers.

(59.) SMIGEL, supra note 20, at 39. He also concluded that 64% had also attended one of nineteen elite colleges, and fully 55% were also alumni of elite boarding schools. Id. at 7273.

(60.) See ROBERT L. NELSON, PARTNERS WITH POWER: THE SOCIAL TRANSFORMATION OF THE LARGE LAW FIRM 132-33 (1988) (noting that four large Chicago law firms increased their hiring of graduates from regional and local schools during the mid-1970s to early 1980s).

(61.) See Scott Jaschik & Douglas Lederman, If You Went Here, You'd Be Sitting Pretty Now, N.Y. TIMES, Apr. 25, 2004, [section] 4A (Education Life Supplement), at 37 (reporting that according to a study by Brian Leiter of the University of Texas School of Law, "Harvard, Chicago, Michigan, Yale [and the] University of Virginia" send the most graduates to top law firms, and that at firms like New York's Cravath, Swaine & Moore, almost half of the firm's lawyers are graduates of Harvard, Columbia, and NYU). Based on this and other evidence, Jaschik and Lederman conclude, "If you want the option of joining the fast track, a prestige school is essential." Id.; see also Angela Cheng, NLJ 250: Most Mentioned Law Schools, NAT'L L.J., Nov. 15, 2004, at S13 (noting that although law firm recruiters reported that they hired students from over two hundred law schools, elite schools like Georgetown, Harvard, and Virginia topped the list).

(62.) See AJD STUDY, supra note 14, at 44 tbl.5.2.

(63.) Id.

(64.) Wilkins & Gulati, supra note 35, at 741 tbl.5.

(65.) Id.

(66.) See Benjamin G. Davis, The Color Line in International Arbitration: An American Perspective, 14 AM. REV. INT'L ARB. 461, 517 (2003) (surveying the backgrounds of lawyers involved in international arbitration and concluding that "[t]his combination of top-ranked schools, contacts and interaction appears to be a non-trivial set of relations enhancing international commercial arbitration specialist human capital").

(67.) For example, the fact that a large California firm continues to place substantial weight on law school grades even for lateral partnership candidates was sufficiently newsworthy that it provoked a lengthy profile in the New York Law Journal. Anthony Lin, Law School Grades Paramount at Gibson Dunn--Even for Laterals, N.Y.L.J., Feb. 10, 2003, at 1. In addition to pointing out how rare the firm's practice was--and how amusing it was to many of the firm's competitors (and insulting to some of its candidates)--the article also stressed that the firm applied its cutoff in a manner consistent with the primary value of graduating from an elite law school. As the managing partner made clear, "[t]he grade requirement varies according to law school ... so that lawyers who graduated from schools such as Harvard are accepted if they rank within the top quarter of their class. Graduates of lower-ranked schools need to do considerably better." Id. This is simply an extension of the metric described above that requires graduates of lower-status schools to receive significantly higher grades than their peers from elite schools in order to receive equivalent consideration.

(68.) Consider the reaction that those who constantly talk about their grade point average typically receive, particularly the longer that they have been out of school.

(69.) See ABEL, supra note 16, at 206 (noting the connection between firm status and the educational background of a firm's lawyers); NELSON, supra note 60, at 66 (same).

(70.) As the article on Gibson Dunn makes clear, this is a primary reason why that firm continues to ask lateral partners about their grades. See Lin, supra note 67, at 1 (noting that "in the status-conscious legal profession, high law school grades can have durable value as totems of prestige, feeding partners' perceptions of themselves as an intellectual or meritocratic elite"). Even some of these other academically related honors, however, are more available to graduates from top schools. For example, an overwhelming majority of Supreme Court law clerks come from schools ranked at the very top of the law school hierarchy. Similarly, because of their greater resources, elite schools offer more opportunities for students to acquire valuable credentials, for example, by participating in one of the many journals these schools tend to operate outside of the school's main law review. I am grateful to Bill Kidder for reminding me of this fact.

(71.) In our study of black Harvard Law School graduates, for example, almost fifty percent of those from the 1970s had parents who worked in blue collar jobs, with many of the rest coming from households where one or both parents were low-level office workers or civil servants. Although the percentage of blacks from such backgrounds has declined with each passing decade, nearly one-third of all black graduates from the 1990s continue to hail from working class roots. See WILKINS ET AL., supra note 33, at 28.

(72.) Interview 19, at 8 (June 9, 1997) ("When I first arrived at Yale... I was invited to a reception at [a dean's] home, where he showed me his corkscrew collection. I really thought I was in the twilight zone viewing that!"). As this son of a factory worker, who is now a senior lawyer in the general counsel's office of a major corporation, went on to sheepishly admit, "I collect vintage watches and corkscrews myself." Compare id. with Interview 96, at 28 (Oct. 11, 1998) (recalling how a fellow black summer clerk from Howard "showed up in an orange velvet sports jacket" and remembering thinking "[t]his is not what you ought to be doing at this place"). The informant, a black lawyer from a small town in Michigan, had learned what he "ought to be doing" at the large in-house legal department where he and the unsuspecting student from Howard were both interning during his time as a student at the University of Michigan Law School.

(73.) See Chambers et al., supra note 33, at 418 n.30 (noting that minorities "are significantly more likely than whites to feel they benefited from friends made at Michigan and from contacts with Michigan alumni after graduation"). To be sure, not every black law student has taken full advantage of this important opportunity. Several black lawyers in my sample acknowledge that they spent far too little time interacting with their white peers while in law school. Although many believe that their careers have benefited from contacts with fellow black students over the years, a point I discuss below, most of those in this group now view their decision to confine the majority of their social interaction to other blacks as a serious mistake--one that they are working hard to see that today's black students do not repeat.

(74.) See id. at 419 (noting that minorities "place a higher value on the prestige of a Michigan Law School degree than whites do"). Black Harvard Law School graduates appear to place similar weight on the prestige of their degree. See WILKINS ET AL., supra note 33, at 49 tbl.25 (reporting that respondents selected the prestige of their degree as one of the most important factors in their success); see also Wilkins & Gulati, Third Year Law Students, supra note 54, at 1230 (reporting that minority students rate law school prestige as more important than grades and that white students reach the opposite conclusion). In a poignant and revealing story, a respondent who attended a local law school recounted how his mentor mistakenly assumed that he had gone to Harvard. "I often wonder if he had known [where I went to law school] what kind of assumptions he would have made." Interview 108, at 51 (Sept. 16, 1998).

(75.) As one respondent summed it up:

   I have rarely encountered any problems that are a result of race in
   my career as an attorney.... Being fair-skinned and basically
   projecting what I would describe as fairly straightforward upper
   middle class values in the way in which I basically function,
   operate, interact and deal with people, I'm not sure that the
   first thing a lot of people ever think about when they meet me is
   that I'm black. I mean, sooner or later they do, for many reasons
   that I know and many reasons that I don't know. But the fact of the
   matter is that, the combination of I think those particular
   characteristics combined with my educational background and
   pedigree, and then throw in a lot of experiences that I've had at a
   fairly early point in my life that basically resulted in my
   developing the ability to sort of negotiate the world in which I
   was operating in, made it possible for me to kind of think about
   opportunity without necessarily first thinking about the possibility
   of impediments based on race. And I must say that in most cases that
   has been borne out. I think that 1 have been fortunate, and I think
   many of my colleagues, people of my vintage have been fortunate in
   arriving on the scene with the right credentials at a time when
   institutions were absolutely desperate to find the right black. And
   I know that when you walk in with a Harvard degree, a Brooks
   Brothers suit and a great set of credentials and experiences that a
   lot of white guys are likely to say, hey, what's the big deal. Got
   one; hey, this is easy; this isn't hard.

Interview 162, at 7 (May 1, 2000).

(76.) For a more detailed description of this study, see Wilkins & Gulati, supra note 35, at 561-63, app. at 622 tbl.3 & 623 tbl.4.

(77.) Wilkins & Gulati, supra note 35, at 561-62; Wilkins, Rollin' on the River, supra note 35, at 534. Although Georgetown was not one of our thirteen elite schools, it is the best law school in the Washington, D.C., area. Given the substantial local effects in the market for lawyers, although Georgetown may not count as an elite school nationally, it does in the Washington, D.C. market. I return to these local effects below.

(78.) See Wilkins & Gulati, supra note 35, at 563-64, 624 app. tbl.5.

(79.) See id. at 741 tbl.5 (noting that 70% of the partners in a sample of five large law firms graduated from elite law schools).

(80.) See CHI. COMM. ON MINORITIES IN LARGE LAW FIRMS, DIRECTORY OF MINORITY PARTNERS (1996-1997).

(81.) Michele Landis Dauber, in her response in this issue, cites an impressive array of research from scholars, employing a variety of methodologies, indicating that blacks receive less benefit from their credentials in the labor market than do whites, particularly in high-status professions such as law. See Michele Landis Dauber, The Big Muddy, 57 STAN. L. REV. 1899, 1903-05 (2005). David Chambers et al. reach a similar conclusion in their response with respect to the importance of elite credentials for black law professors. See David L. Chambers et al., The Real Impact of Eliminating Affirmative Action in American Law Schools: An Empirical Critique of Richard Sander's Study, 57 STAN. L. REV. 1855, 1896 (2005) (reporting that 60.1% of all black law professors attended top twenty schools, with 48.1% graduating from the top ten). All of this research is consistent with the view that blacks need better credentials ifs they are to have an equal chance to succeed in markets where their value continues to be discounted. I discuss how the existence of affirmative action affects these findings below.

(82.) Ann Davis, Big Jump in Minority Associates, but ..., NAT'L L.J., Apr. 29, 1996, at 1, 22.

(83.) This is especially true for blacks who are trying to build national, as opposed to local, careers. The black lawyers who find themselves pushed down into Tier 4 or 5 schools under Sander's plan will be much more likely than they are today to build careers close to where their schools are located. Their degrees will simply be more valuable the closer they stay to home. For those who are bumped down to local and commuter schools, home may be the only place their degrees have value, A few graduates from Suffolk University Law School (almost all of whom were in the top of their class) receive job offers from large law firms in Boston. Many more populate the small and midsized firms in New England that ply their trade in the local civil and criminal courts and represent local businesses in their dealings with others of similar size and with city and state officials. Virtually none go on to work in Chicago or Los Angeles. As a manager of a website which helps law students find jobs recently put it, "Going to Loyola Law School can lead you to a pretty lucrative, satisfying life in Los Angeles.... But if you want to have any chance of going national, you really have to be at U.C.L.A. or U.S.C." Jaschik & Lederman, supra note 61 (quoting Adam Avitable, a manager of Legal Authority). The more blacks attend schools like Suffolk, the fewer black lawyers we are likely to see on the national stage.

(84.) Sander, supra note 2, at 468.

(85.) I chronicle these efforts in Wilkins, supra note 32, at 1568-71. Once again, it is important to note that this finding is consistent with research in other areas demonstrating that successful blacks help other blacks to succeed. See Holzer & Neumark, supra note 33, at 499 (discussing studies that show that "black owners/managers hire more black employees than white owners/managers in similar locations").

(86.) Sander, supra note 2, at 466.

(87.) Richard Connelly, Preconceived Notions: Recruiters Pigeonhole TSU Graduates, Leaving Top Students in the Cold, TEX. LAW., May 17, 1993, at S-1.

(88.) Id.

(89.) Id.

(90.) Id.

(91.) See also Rita H. Jensen, Minorities Didn't Share in Firm Growth, NAT'L L.J., Feb. 19, 1990, at 1 (chastising firms for not recruiting at historically black law schools such as Howard).

(92.) Sander, supra note 2, at 371 n.7. As indicated above, noticeably absent from this list is any unfairness to white applicants.

(93.) See Mortgage Group Joins NAACP to Help Blacks Purchase Homes, CHI. TRIB., Jan. 22, 1999, at 1 (describing a partnership spearheaded by Franklin Raines, the company's black former CEO). Raines is a graduate of Harvard Law School. Needless to say, Raines's recent departure from Fannie Mae in no way undermines the value of the work that he did while there.

(94.) See America's Top Black Lawyers, BLACK ENTERPRISE, Nov. 2003, at 121, 128, 146; Coke Names Patrick, Civil Rights Lawyer, as General Counsel, WALL ST. J., Jan. 25, 2001, at B2. Deval Patrick is also a Harvard Law School graduate.

(95.) Vivian Chen, Master of the Game, MINORITY L.J., Summer 2001, at 17, 20 (describing Merck General Counsel Kenneth Frazier's role in the company's decision to reduce the cost of the company's AIDS drugs in Africa). Frazier, like Raines and Patrick, is a black graduate of Harvard Law School.

(96.) See Jonathan Alter, The Audacity of Hope, NEWSWEEK, Dec. 27, 2004, at 74. As anyone who has followed Obama's meteoric rise already knows, the fact that he is a Harvard Law School graduate has been central to his appeal.

(97.) Indeed, there are good grounds for worrying that under Sander's plan the graduates of highly ranked schools might not be as successful as they are today. In addition to whatever educational benefits it brings, the "critical mass" of blacks who currently attend top schools also generates cohort effects that have played an important role in career building. If the number of black students attending these institutions were to decline significantly, future black graduates of these institutions would be less able to call on black classmates for information, opportunities, and support.

(98.) Judge Edwards sums up this value with his usual eloquence:

   [W]hen I graduated from Cornell, and later with honors from the
   University of Michigan Law School, no one doubted that I had made it
   on my own merits; indeed, I often received backhanded compliments
   from people who said that I must be talented, because I had been
   admitted to Cornell and Michigan and had succeeded despite my race.
   I had the advantage of an assumption that I was smart, and I
   believed it.

See Harry T. Edwards, Personal Reflections on 30 Years of Legal Education for Minority Students, 37 LAW QUADRANGLE NOTES 38, 39 (1994). As Judge Edwards goes on to say, this is one respect in which his life has been easier than that of his son, who has grown up without such a presumption. Id. at 40. Notwithstanding this cost, however, Judge Edwards makes clear that affirmative action policies such as those instituted by his alma mater, the University of Michigan, have played a crucial role in opening the doors of opportunity for black lawyers. Id. at 39-43.

(99.) Sander, supra note 2, at 437 tbl.5.5, 446 tbl.6.2.

(100.) Id. at 473 tbl.8.2.

(101.) Id. at 474.

(102.) Id. This is an example of where Sander cites the more shocking first-time failure rate (81%) in text and only gives the still low, but significantly higher (this time, almost 40% higher) eventual passage rate in the footnote. As I indicate below, it is the latter statistic that is the most important.

(103.) I arrived at this number by subtracting the number of black lawyers eventually passing the bar under the present system in Sander's Table 8.2 (1981), id. at 473 tbl.8.2, from the number Sander claims will eventually become lawyers under his scenario (2150), id. Once again, the fact that Sander predicts that first-time passage rates will increase substantially under his plan is less important than the total number of new black lawyers that his proposal would ultimately produce.

(104.) I say the middle because, as we have seen, blacks at the top of the distribution-the ones most likely to attend elite law schools--almost never fail to graduate from law school and eventually pass the bar.

(105.) Sander, supra note 2, at 481-82.

(106.) Id. at 473 tbl.8.2. The number is the difference between "Graduates (2004 or Later)" (2802) and "Graduates Taking the Bar" (2552). This number may actually overstate the percentage of law graduates who do not take the bar, since the Bar Passage Study included some respondents for whom bar information could not be determined. As elsewhere, I am simply relying on Sander's reporting of the data, upon which he bases his own calculations. I am grateful to David Chambers for bringing this possible discrepancy to my attention.

(107.) See Clydesdale, supra note 11, at 726 (discussing the importance of "life events" in analyzing the results of the Bar Passage Study).

(108.) Given that scores on standardized tests, which under Sander's formula are the main criteria constituting this group, are highly correlated with wealth, it is likely that those who fall in this group are more likely to come from working class families and to have attended less prestigious secondary and undergraduate schools. See Jesse M. Rothstein, College Performance Predictions and the SAT, 121 J. ECONOMETRICS 297, 298 (2004) (discussing the "acknowledged correlation between SAT scores and student socioeconomic status"); see also Linda F. Wightman, The Threat to Diversity in Legal Education, 72 N.Y.U. L. REV. 1, 42-43 (1997) (reporting that 50.7% of blacks in the Bar Passage Study came from lower-middle-class backgrounds). As a result, these black students are likely to have fewer employment opportunities when they graduate from college.

(109.) It is important to pause to note here that this is not significantly below the median starting salary for all of the graduates of such institutions who go into solo or small firm practice. See AJD STUDY, supra note 14, at 44 tbl.5.2 (showing the median starting salary for Tier 4 graduates who go into solo practice as $57,500 and for those who work in firms of two to twenty lawyers as $54,500). It is also higher than the median salary for graduates from these institutions who start their career in state or local government ($45,000) and within striking distance of the median salary of those who begin their careers with the federal government ($56,000), which is also very close to the mean ($56,182) for all graduates of these institutions. Id.

(110.) See NAT'L CTR. FOR EDUC. STATISTICS, U.S. DEP'T OF EDUC., A DESCRIPTIVE SUMMARY OF 1999-2000 BACHELOR'S DEGREE RECIPIENTS 1 YEAR LATER WITH AN ANALYSIS OF TIME TO DEGREE: STATISTICAL ANALYSIS REPORTS 127 (2003).

(111.) Indeed, as Linda Wightman speculates, this may account for why a substantial number of blacks who fail the exam for the first time never retake it. See Linda F. Wightman, Through a Different Lens: A Reply to Stephan Thernstrom, 15 CONST. COMMENT. 45, 55-56 (1998) ("The findings about single attempts also raise questions, however, about the utility of bar passage as a criterion in evaluating the success of admission decisions. How many of those graduates who did not attempt the bar a second time did so because passing was not necessary to entry or success in their chosen career?"). I return to this issue below.

(112.) See, e.g., CARROLL SERON, THE BUSINESS OF PRACTICING LAW: THE WORK LIVES OF SOLO AND SMALL-FIRM ATTORNEYS (1996); Stephen Daniels & Joanne Martin, "'The Impact That It Has Had Is Between People's Ears": Tort Reform, Mass Culture, and Plaintiffs' Lawyers, 50 DEPAUL L. REV. 453 (2000).

(113.) See Sander, supra note 2, at 437. As I indicate later in this Part, this is partially because many blacks who fail the first time never retake the exam. See infra p. 1952.

(114.) GEORGE BERNARD SHAW, THE DOCTOR'S DILEMMA, at xv (1911). The claim that the primary goal of professional licensing is to protect the income and status of practitioners has been frequently invoked by critics on both the right and the left. See, e.g., MILTON FRIEDMAN, CAPITALISM AND FREEDOM 144-49 (1962); Richard L. Abel, Why Does the ABA Promulgate Ethical Rules?, 59 TEX. L. REV. 639 (1981).

(115.) Sander, supra note 2, at 475.

(116.) Id.

(117.) Id. at 377.

(118.) LAW SCH. ADMISSION COUNCIL, MINORITY DATA BOOK (2002); Law Sch. Admission Council, Volume Summary Data, at http://www.lsat.org/ LSAC.asp?url=/lsac/ lsac-volume-summary.asp (last visited Apr. 18, 2005).

(119.) See Daniel R. Hansen, Do We Need the Bar Examination?: A Critical Evaluation of the Justifications for the Bar Examination and Proposed Alternatives, 45 CASE W. RES. L. REV. 1191, 1216 & nn.131-32 (1995) (documenting the rise in quality of law students); William C. Kidder, The Bar Examination and the Dream Deferred: A Critical Analysis of the MBE, Social Closure, and Racial and Ethnic Stratification, 29 LAW & Soc. INQUIRY 547 (2004) (documenting the rise in the quality of law students).

(120.) See Robert C. Clark, Why So Many Lawyers? Are They Good or Bad?, 61 FORDHAM L. REV. 275 (1992) (documenting the rapid escalation in the number of lawyers since 1960); see also U.S. CENSUS BUREAU, STATISTICAL ABSTRACT OF THE UNITED STATES 2004-2005, at 385 (2004) (reporting that there are now 952,000 lawyers in the United States).

(121.) Passage rates in several Sun Belt states fell significantly between 1994 and 2002. For example, Florida's passage rate fell from 84% in 1994 to 68% in 2002. California's dropped from 56% to 45% during this period, while Georgia's fell from 75% to 70%. States with large legal markets also experienced significant declines. Illinois's passage rate, for example, dropped by a whopping 22 percentage points (from 94% to 72%) during this period, while New York's passage rate declined from 73% to 61%. By contrast, states like Mississippi and North Dakota, both of which plausibly would like to raise the number of lawyers practicing within their borders, actually increased their passage rates (in Mississippi from 69% to 81% and in North Dakota from 78% to 85%) during this period. The statistics in this paragraph are from Nat'l Conference of Bar Examiners, 1994 Statistics, B. EXAMINER, May 1995, at 8, 9-10, available at http://www.ncbex.org/stats/ pdff1994stats.pdf; Nat'l Conference of Bar Examiners, 2002 Statistics, B. EXAMINER, May 2003, at 6, 6-7, available at http://www.ncbex.org/stats/pdf/2002stats.pdf.

(122.) See Andrew M. Perlman, A Bar Against Competition: The Unconstitutionality of Admission Rules for Out-of-State Lawyers, 18 GEO. J. LEGAL ETHICS 135, 142-43 (2004) (noting that many states require attorneys seeking to relocate to retake the entire bar exam and that many of those who do not require special attorneys' exams); see also Birbrower, Montalbano, Condon & Frank, P.C.v. Superior Court, 949 P.2d 1 (Cal. 1998) (holding that lawyers who were fully licensed in New York were guilty of the unauthorized practice of law in California for representing a California client in connection with a pending arbitration).

(123.) See, e.g., Kidder, supra note 119, at 549-55; Perlman, supra note 122, at 146-50. As Kidder makes clear, even if one accounts for the fact that some of the change in passage rates is a result of the changes in the applicant pool, the protectionist motivations of many bar officials are nevertheless apparent in light of the many misleading statements by such officials concerning the need to make the exam tougher because of the declining quality of applicants. Kidder, supra note 119, at 551-55.

(124.) Following Birbrower the California Legislature passed a statute allowing out-ofstate lawyers to represent clients in arbitration proceedings (the type of proceeding at issue in Birbrower) so long as there is also a local lawyer acting as "counsel of record." More recently, the ABA Committee on Multijurisdictional Practice has proposed rules that would also take some of the sting out of the Birbrower restrictions. Perlman, supra note 122, at 140 n.20. As Rick Abel's trenchant work has repeatedly taught us, however, the fact that lawyers are not always successful at restricting competition does not mean that they do not continue to try. See ABEL, supra note 16.

(125.) See DEBORAH L. RHODE, IN THE INTERESTS OF JUSTICE: REFORMING THE LEGAL PROFESSION 151 (2000) ("No effort has been made to correlate performance on admission exams with performance in practice. The most that bar officials can establish is a correlation between examination scores and law school grades.").

(126.) Sander, supra note 2, at 418-25.

(127.) It is not surprising that LSAT scores are highly correlated with bar passage. Both exams share many of the same properties, i.e., both use multiple choice and/or short essay formats and are administered under extreme time constraints. As a result, it is not surprising that those who do well on one are likely to do well on the other--and that issues of stereotype threat are likely to be activated in both situations. See Claude M. Steele & Joshua Aronson, Stereotype Threat and the Test Performance of Academically Successful African Americans, in THE BLACK-WHITE TEST SCORE GAP 401 (Christopher Jencks & Meredith Phillips eds., 1998). As Sander concedes, law school exams also share some of these qualities. I am inclined to agree with Sander that there are also important differences between standardized tests and exams. Having said that, however, it is important to note that the evidence Sander proffers for the proposition that all of these devices are not simply measuring the same limited set of qualities under essentially similar conditions is quite weak. Sander points to evidence from two data sources that the black-white gap in legal writing courses is equal to, if not larger than, the gap for courses employing traditional exams. Sander, supra note 2, at 424 n. 164. As he concedes, however, the sample size for both data sources is tiny and the samples are arguably quite skewed. Moreover, as he also notes but then dismisses, legal writing classes are not graded anonymously. Although I tend to agree with Sander that professors--or, more likely, the recent graduates or adjuncts who actually grade student papers in these courses--are not likely to be consciously biased against blacks, as my interviews have revealed time and time again, the stereotype that "blacks can't write" is so pervasive in our culture that it is quite possible that it often penetrates at a subconscious level the highly discretionary subjective judgments that those grading these exercises invariably employ. At a minimum, Sander's concession that "more research on this point is needed" is a dramatic understatement. Id. at 434 n. 182.

(128.) Id. at 477.

(129.) Id.

(130.) Id. at 478.

(131.) In addition to the cost of retaking the exam, and the undeniable psychological trauma associated with failing such a high-stakes test, employers may also take adverse action against those who fail the bar. We know very little, however, about how important such sanctions are for graduates of low-status schools, many of whom may he working as solo practitioners or in other settings where failing the bar is either more common or the consequences of failing less severe than we tend to think of them as being for the graduates of elite schools.

(132.) Similarly, it would unquestionably be a good thing to help more black students graduate from law school. The question, however, is whether there are other ways to achieve this goal that do not require the substantial costs associated with dismantling affirmative action outlined above.

(133.) As I often tell my students, you do not want to get an "A" on the bar exam.

(134.) Sander, supra note 2, at 423 n.159. Needless to say, even this is a highly debatable proposition.

(135.) Id. at 481.

(136.) See Kidder, supra note 119, at 569-75 (noting opposition to raising thresholds on the ground that it would disproportionately hurt black applicants).

(137.) See Sander, supra note 2, at 475.

(138.) As Sander acknowledges, rising passing scores disproportionately burden black test-takers. Id. at 475-76. It is impossible to say how much rolling back passing rates would improve black bar passage without knowing much more about the demographics and passage rates in particular states. Nevertheless, even if the 8.6% increase in the overall passage rate that would result from returning to 1994 levels produced only a 3% or 4% rise in black passage rates, this would translate into almost half of the number of additional black lawyers that Sander posits would result from eliminating affirmative action. Cf. id. at 476 (noting that the black first-time passage rate in California fell by almost 15% when the state raised its minimum passing score).

(139.) "[B]lack examinees [who] failed the first attempt at the bar and never attempted it again ... represent nearly half of those in the failed category" of bar passage rates. Wightman, supra note 111, at 55.

(140.) Even if passage rates generally decline with repeated attempts, so long as they do not decline to zero or we have some other reason for thinking that those who do not retake the test are especially inclined to fail, we can safely predict that some blacks who currently stop after the first failure would eventually become lawyers. Cf. id. at 54-55 ("Among the passing black examinees, 222 required more than one attempt. Among those, 69 percent passed on their second attempt and 93 percent passed by the third attempt.").

(141.) Sander, supra note 2, at 423 & n.160.

(142.) Whether this is in fact true of the SAT---or even more importantly for our purposes, the LSAT--I leave for others to debate. For an early study suggesting that the LSAT is indeed coachable, see Brian Powell & Lala Cart Steelman, Equity and the LSAT, 53 HARV. EDUC. REV. 32, 40-41 (1983) (finding gains as high as 33-40 points on a 200-600 point scale from only two hours of coaching).

(143.) This is especially true, as I have suggested, in elite schools, but it is generally true across the board. Even schools that "teach to the bar" are likely to fail to cover many areas that actually appear on the exam, and few students will have taken all of the bar-related courses that are offered. Moreover, even local and regional schools typically teach federal (or generically national) rules with little emphasis on the peculiar variations and practices in their own jurisdiction, let alone differences that might exist elsewhere. Yet these differences are the standard grist for the mill of the typical state bar exam.

(144.) See Comm. on Legal Educ. & Admission to the Bar, Ass'n of the Bar of the City of N.Y., Report on Admission to the Bar in New York in the Twenty First Century--A Blueprint for Reform, 47 REC. ASS'N B. CITY N.Y. 464, 484, 503-04 (1992).

(145.) It is important to emphasize that the costs of a bar preparation course must be added to the substantial cost of studying for the exam, which can take literally hundreds of hours. See Kristen Booth Glen, When and Where We Enter: Rethinking Admission to the Legal Profession, 102 COLUM. L. REV. 1696, 1734 (2002) (suggesting that 350 hours of study is traditionally the time needed to study for the bar). For elite law school graduates, this time is typically subsidized (if not paid for outright) by their future employers. Once again, this subsidy will not be available to most of the graduates from the low-status law schools we are discussing who must find ways to support themselves and pay the fees associated with the exam itself before they can even think about paying for a bar review course.

(146.) See Andrea A. Curcio, A Better Bar: Why and How the Existing Bar Exam Should Change, 81 NEB. L. REV. 363, 391 (2002) (citing evidence from the Bar Passage Study that blacks are more likely to have financial or family responsibilities that plausibly affect access to bar review courses and bar passage rates); see also COMM. ON BAR ADMISSIONS AND LAWYER PERFORMANCE & RICHARD A. WHITE, AALS SURVEY OF LAW SCHOOLS ON PROGRAMS AND COURSES DESIGNED TO ENHANCE BAR EXAMINATION PERFORMANCE 21 (2001), available at http://www.aals.org/Bar2001Report.pdf (noting that CUNY reported an increase in its MBE scores upon providing its students with the PMBR course free of charge).

(147.) For an excellent review of possible alternatives to the current bar exam, see Kidder, supra note 119, at 563-82.

(148.) Whether any particular set of skills or abilities is caused by race is, of course, a different and much more controversial question. Like Sander, I am skeptical of any claim that suggests that there are innate differences between blacks and whites and therefore doubt that any causal connection between race and performance exists. That said, it is certainly plausible that given a variety of psychological, cultural, and measurement-related issues, blacks are more likely to score better on some kinds of exams than others. See Steele & Aronson, supra note 127.

(149.) For one such proposal by the dean of CUNY Queens Law School, see Kristin Booth Glen, Thinking Outside the Bar Exam Box: A Proposal to "MacCrate" Entry to the Profession, 23 PACE L. REV. 343 (2003).

(150.) Indeed, many of these low-level individual services are currently performed by paralegals who have no formal legal training but who must now work under the nominal supervision of attorneys (who, of course, continue to extract the majority of the gains).

(151.) See STEPHEN L. CARTER, REFLECTIONS OF AN AFFIRMATIVE ACTION BABY (1991).

(152.) Sander, supra note 2, at 482.

(153.) As David Chambers has suggested to me in comments on a prior draft, why not simply give every admitted student their own index score and the chances of graduating for students in various index ranges? This would allow all students who are being admitted with low scores to understand the risk that they might not graduate (or, in the case of those attending elite schools, to be reassured about the fact that they are quite likely to graduate) without also reinforcing stereotypes and motivational problems of the kind described in the text.

(154.) Sander, supra note 2, at 450.

(155.) Id. at 450-51; see also Note, Making Docile Lawyers: An Essay on the Pacification of Law Students, 111 HARV. L. REV. 2027 (1998) (arguing that Harvard law students who do not get high grades routinely disengage from their studies).

(156.) Mitu Gulati et al., The Happy Charade: An Empirical Examination of the Third Year of Law School, 51 J. LEGAL EDUC. 235, 244 (2001).

(157.) Sander, supra note 2, at 453.

(158.) Recall Judge Edwards's statement that "I had the . assumption that I was smart, and I believed it." Edwards, supra note 98, at 39 (second emphasis added).

(159.) Geoffrey Cohen suggests that black students may be especially sensitive to negative expectations. See Geoffrey L. Cohen et al., The Mentor's Dilemma: Providing Critical Feedback Across the Racial Divide, 25 PERSONALITY & SOC. PSYCHOL. BULL. 1302 (1999).

(160.) Interview 55, at 14 (July 11, 1997).

(161.) A respondent who went to Emory Law School in the mid-1980s underscores how such well-intentioned efforts can quickly turn into a self-fulfilling prophecy:

   The most frustrating aspect of Emory was that ... you'd go to
   these special programs for minority students, and they'd
   basically tell you walking in the door that black students at
   Emory don't do well. And it becomes a self-fulfilling
   prophecy.... You go through your first semester. You really
   don't know what you're doing. You haven't really been able to
   make contact with some of the other students to get the
   understanding of how you need to be studying and how your
   outlines need to be, and you don't know to go and talk to your
   professors and things of that nature. All you hear is black
   students at Emory don't do well. So, first semester grades
   come in, many of the students didn't do extremely well, and so
   therefore it became, well, why should I even try because no
   matter how hard I try, I'm not going to do well.

Interview 140, at 6 (Aug. 9, 1999); cf. Clydesdale, supra note 11 (finding that participation in a pre-law school academic support program is negatively correlated with bar passage).

(162.) As one typical respondent reports:

   In civil procedure.... the professor asked a very difficult
   question. It was my second week of class. I had no idea
   what he was talking about. I didn't realize no one else
   did either at the time, but I never will forget. He went
   down the row, he called on eight people, and he got to
   me and he skipped over me and went on to the next person.
   I guess he said, "Well, I know be doesn't know."

Interview 125, at 5 (Aug. 11, 1999). As the respondent went on to make clear: "It made me feel even more inferior, that he had assumed that because I am black--that was the only reason I could think of why he would skip over me, that I didn't know--I wouldn't have a reasonable response to the question." Id.

(163.) As one respondent who went to law school in the year Bakke was under consideration described how he was treated by other students: "lilt just seemed like we were under an insidious, constant attack. And you put that on top of just the general first year experience, adjusting to the first year of law school, didn't make for a very, very pleasant experience." Interview 29, at 31 (June 27, 1997). As Clydesdale notes, black students are significantly more likely than others to report having experienced racist incidents while in school. See Clydesdale, supra note 11, at 757.

(164.) Virtually every law school "norms" the grades of all of its applicants to account for perceived differences in quality among undergraduate institutions. Clearly, if black applicants were to come increasingly from less prestigious colleges and universities, this would negatively affect their chances of getting into a good law school.

(165.) Notwithstanding all of the talk about the fact that "diversity is good for business," it is plain that many law firm partners and other similar employers believe that they are engaging in "affirmative action" when they hire black lawyers with credentials--or from institutions--that, if presented by a white candidate, would be unlikely to lead to a decision to hire. See Wilkins, Book Review, supra note 35, at 1955-57 (discussing how many partners believe in the "affirmative action myth" that all black lawyers--even those who meet all of a firm's traditional qualifications--are in fact the product of affirmative action). Indeed, even apart from their views about affirmative action, the logic of Sander's proposal is likely to lead many employers to rely more on law school status as a proxy for potential than they do today. After all, under Sander's plan the "overmatched" black students who are not admitted to top schools will be replaced by white students who are, according to their entering credentials, not overmatched. Although some of these white students (or others of their white peers--once again confining my analysis to blacks and whites) will get the same low grades

the black students used to get (since most schools grade on a curve), their "quality" (as measured by their index scores) will actually be higher. Therefore, a rational employer persuaded by Sander's argument about the correlation among index scores, grades, and merit may very well put more stock in law school status under Sander's plan than they do today. This process is likely to damage the employment prospects of black students without elite credentials even further. I am grateful to my research assistant Thomas Tso for pressing this point. Finally, black law students might decrease their effort to do well in law school if they see the prospect of getting a good job after graduation as becoming too remote. See Wilkins & Gulati, supra note 35, at 602-03 (arguing that affirmative action in recruiting actually raises the incentives for black lawyers by helping them to believe that they have a realistic chance to succeed). Given these combined effects, it is not surprising that Holzer and Neumark conclude that "in our view discrimination [in the labor market] persists even in the face of affirmative action, and would likely worsen in its absence...." Holzer & Neumark, supra note 33, at 501.

(166.) See Court Revives Diversity Issue in Michigan, N.Y. TIMES, June 14, 2004, at A11 (reporting that a court has cleared the way for a Michigan ballot initiative outlawing affirmative action); Daniel Golden, Not Black and White: Colleges Cut Back Minority Programs After Court Rulings, WALL ST. J., Dec. 30, 2003, at A1 (reporting on various court challenges to minority scholarships and other similar programs after Grutter).

(167.) More needs to be known about the process of legal education as well. As Tim Clydesdale notes, entering credentials do not fully explain black performance in law school. Clydesdale, supra note 11, at 711. Something in the law school environment is further depressing black achievement. Whether that something is discrimination, stereotype threat, culture, or expectations, any meaningful attempt to improve black performance must account for this reality. See also LANI GUINIER ET AL., BECOMING GENTLEMEN: WOMEN, LAW SCHOOL, AND INSTITUTIONAL CHANGE 76-77 (1997) (making a similar argument about the performance of female students).

(168.) Sander, supra note 2, at 483.