The Myth And Math of Affirmative Action
By Goodwin Liu
With the arrival of spring, thousands of high school and college seniors have been anxiously checking the mail for word from the nation's most prestigious universities. Although some envelopes are thick with good news, most are thin and disappointing. For many white applicants, the disappointment will become bitterness if they suspect the reason for their rejection was affirmative action. But such suspicions, in all likelihood, are misplaced.
Affirmative action is widely thought to be unfair because it benefits minority applicants at the expense of more deserving whites. Yet this perception tends to inflate the cost beyond its real proportions. While it is true that affirmative action gives minority applicants a significant boost in selective admissions, it is not true that most white applicants would fare better if elite schools eliminated the practice. Understanding why is crucial to separating fact from fiction in the national debate over affirmative action.
Any day now, a federal appeals court in Cincinnati will issue a decision in a major test lawsuit challenging the use of race as a factor in selective admissions. In that case, the University of Michigan denied admission in 1995 to a white undergraduate applicant named Jennifer Gratz. Charging reverse discrimination, Gratz said, "I knew of people accepted to Ann Arbor who were less qualified, and my first reaction when I was rejected was, 'Let's sue.' "
The Michigan case will likely end up at the Supreme Court. If it does, Gratz will try to follow in the footsteps of Allan Bakke, a rejected white applicant who won admission in 1978 to the University of California at Davis's medical school after convincing the high court that the school's policy of reserving 16 of 100 seats each year for minority students was unconstitutional. For many Americans, the success of Bakke's lawsuit has long highlighted what is unfair about affirmative action: Giving minority applicants a significant advantage causes deserving white applicants to lose out. But to draw such an inference in Bakke's case -- or in the case of the vast majority of rejected white applicants -- is to indulge in what I call "the causation fallacy."
There's no doubt, based on test scores and grades, that Bakke was a highly qualified applicant. Justice Lewis Powell, who authored the decisive opinion in the case, observed that Bakke's Medical College Admission Test (MCAT) scores placed him in the top tier of test-takers, whereas the average scores of the quota beneficiaries in 1974 placed them in the bottom third. Likewise, his science grade point average was 3.44 on a 4.0 scale, compared with a 2.42 average for the special admittees, and his overall GPA was similarly superior. Given these numbers, the only reason for Bakke's rejection was the school's need to make room for less qualified minority applicants, right?
Wrong. Although Justice Powell pointed out that minority applicants were admitted with grades and test scores much lower than Bakke's, he did not discuss what I found to be the most striking data that appeared in his opinion: Bakke's grades and scores were significantly higher than the average for the regular admittees. In other words, his academic qualifications were better than those of the majority of applicants admitted outside the racial quota. So why didn't he earn one of the 84 regular places?
It is clear that the medical school admitted students not only on the basis of grades and test scores, but on other factors relevant to the study and practice of medicine, such as compassion, communication skills and commitment to research. Justice Powell's opinion does not tell us exactly what qualities the regular admittees had that Bakke lacked. But it notes that the head of the admissions committee, who interviewed Bakke, found him "rather limited in his approach" to medical problems and thought he had "very definite opinions which were based more on his personal viewpoints than upon a study of the total problem." Whatever Bakke's weaknesses were, there were several reasons, apart from affirmative action, that might have led the medical school to reject his application. Grades and test scores do not tell us the whole story.
Of course, affirmative action did lower Bakke's chance of admission. But by how much? One way to answer this question is to compare Bakke's chance of admission had he competed for all 100 seats in the class with his chance of admission competing for the 84 seats outside of the racial quota. To simplify, let's assume none of the special applicants would have been admitted ahead of any regular candidate.
In 1974, Bakke was one of 3,109 regular applicants to the medical school. With the racial quota, the average likelihood of admission for regular applicants was 2.7 percent (84 divided by 3,109). With no racial quota, the average likelihood of admission would have been 3.2 percent (100 divided by 3,109). So the quota increased the average likelihood of rejection from 96.8 percent to 97.3 percent.
To be sure, Bakke was not an average applicant. Only one-sixth of regular applicants (roughly 520) received an interview. But even among these highly qualified applicants, eliminating the racial quota would have increased the average rate of admission from 16 percent (84 divided by 520) to only 19 percent (100 divided by 520). Certainly a few more regular applicants would have been admitted were it not for affirmative action. But Bakke, upon receiving his rejection letter, had no reason to believe he would have been among the lucky few.
In fact, Bakke applied in both 1973 and 1974 and, according to evidence in the lawsuit, he did not even make the waiting list in either year.
The statistical pattern in Bakke's case is not an anomaly. It occurs in any selection process in which the applicants who do not benefit from affirmative action greatly outnumber those who do.
Recent research confirms this point. Using 1989 data from a representative sample of selective schools, former university presidents William Bowen and Derek Bok showed in their 1998 book, "The Shape of the River," that eliminating racial preferences would have increased the likelihood of admission for white undergraduate applicants from 25 percent to only 26.5 percent.
The Mellon Foundation, which sponsored the study, provided me with additional data to calculate admission rates by SAT score. If the schools in the Bowen/Bok sample had admitted applicants with similar SAT scores at the same rate regardless of race, the chance of admission for white applicants would have increased by one percentage point or less at scores 1300 and above, by three to four percentage points at scores from 1150 to 1299, and by four to seven percentage points at scores below 1150.
It is true that black applicants were admitted at much higher rates than white applicants with similar grades and test scores. But that fact does not prove that affirmative action imposes a substantial disadvantage on white applicants. The extent of the disadvantage depends on the number of blacks and whites in the applicant pool. Because the number of black applicants to selective institutions is relatively small, admitting them at higher rates does not significantly lower the chance of admission for the average individual in the relatively large sea of white applicants.
In the Bowen/Bok study, for example, 60 percent of black applicants scoring 1200-1249 on the SAT were admitted, compared with 19 percent of whites. In the 1250-1299 range, 74 percent of blacks were admitted, compared with 23 percent of whites. These data indicate -- more so than proponents of affirmative action typically acknowledge -- that racial preferences give minority applicants a substantial advantage. But eliminating affirmative action would have increased the admission rate for whites from 19 percent to only 21 percent in the 1200-1249 range, and from 23 percent to only 24 percent in the 1250-1299 range.
These figures show that rejected white applicants have every reason not to blame their misfortune on affirmative action. In selective admissions, the competition is so intense that even without affirmative action, the overwhelming majority of rejected white applicants still wouldn't get in.
Still, isn't it true that minority applicants are admitted at rates up to three times higher than white applicants with similar SAT scores? Isn't that unfair?
To answer that question, it's important to observe that racial preferences are not the only preferences that cause different groups of applicants with similar test scores to be admitted at different rates. Geo- graphic, athletic and alumni preferences also weigh heavily, to the detriment of applicants such as Jennifer Gratz at Michigan. Gratz hailed from a Detroit suburb, not from a rural area or the inner city. She was not a star athlete. And her working-class parents were high school graduates, not University of Michigan alumni.
Yet preferences for athletes, though occasionally criticized, have never galvanized the kind of outrage often directed at affirmative action. Similarly, there is no organized legal campaign against geographic preferences, even though where one grows up is as much an accident of circumstance as one's skin color. And neither Gratz nor her lawyers at the Washington-based Center for Individual Rights have publicly denounced alumni preferences, much less launched a moral crusade against them.
Such preferences reflect institutional interests that are unrelated to an applicant's grades or test scores. But the same is true of affirmative action when it is used to enhance educational diversity. The question, then, is not whether unequal treatment is unfair as a general rule, but whether unequal treatment based on race should be singled out for special condemnation.
As the Supreme Court said in 1954, unequal treatment based on race can inflict on members of a disfavored race "a feeling of inferiority as to their status in the community that may affect their hearts and minds in a way unlikely ever to be undone." But social stigma is not the complaint pressed by white applicants such as Bakke or Gratz. Despite 30 years of affirmative action, white students continue to dominate most of the nation's best colleges and all of the top law and medical schools. Against this backdrop, not even the most ardent foe of affirmative action would say that it stamps white applicants with a badge of racial inferiority. Indeed, just as athletic and geographic preferences do not denigrate applicants who are uncoordinated or suburban, affirmative action is not a policy of racial prejudice.
For white applicants, the unfairness of affirmative action lies not in its potential to displace or stigmatize, but in its potential to stereotype. Minority applicants are not the only ones who contribute to educational diversity. Were a school to use race as its sole "plus" factor in admissions, then white applicants could legitimately complain that the school failed to take into account non-racial attributes essential to genuine educational diversity.
Putting the complaint in these terms is an important first step toward rethinking the conventional view that a race-conscious admissions policy pits whites against minorities in a zero-sum game. Instead of attacking affirmative action, white applicants such as Jennifer Gratz might do better to urge top schools committed to educational diversity to place a higher premium on first-generation college attendance or growing up in a blue-collar home. Ironically, the stories of affirmative action's "victims" could spur America's colleges to further widen the elite circles of educational opportunity. And that would be a result students of any color could applaud.
Goodwin Liu, a Washington attorney, was a Supreme Court law clerk during the 2000-01 term. His essay is adapted from "The Causation Fallacy: Bakke and the Basic Arithmetic of Selective Admissions," to be published this month in the Michigan Law Review.
For more information:
Association of American Colleges and Universities Statement on Affirmative Action
October 14, 2002
Supreme Court Decisions on the Michigan Cases - Education Law and Policy Alert by Art Coleman and Scott Palmer, June 23, 2003 (pdf)*