Today’s New York Times includes a passionate op-ed by Sherrilyn A. Ifill, the president and director-counsel of the NAACP Legal Defense and Education Fund Inc., in defense of race-based affirmative action. The future direction of such policies is likely to be decided at some time in the next two weeks when the U.S. Supreme Court issues its ruling in a challenge to racial preferences in the case of Fisher v. University of Texas.

In particular, Ifill is concerned that “an alarming number of scholars, pundits and columnists—many of them liberal—have declared that economic class, not race, should be the appropriate focus of university affirmative-action efforts.” As a longtime proponent of class-based affirmative action (author of a 1996 book, The Remedy: Class, Race and Affirmative Action, coauthor a 2012 Century Foundation report, A Better Affirmative Action: State Universities that Created Alternatives to Racial Preferences) and a liberal, to boot, let me explain why I disagree with the four central arguments Ifill advances in favor of racial preference policies.

1. Race Still Matters, Therefore Racial Preferences Are Needed.

Ifill argues that because racial discrimination continues to exist in American society—in the criminal justice system, housing, and employment—universities should be allowed to use racial preferences as a countermeasure.

Ifill is correct, of course, to suggest that discrimination continues to be a serious problem, which is why it would be crazy to repeal civil rights laws that are meant to address such discrimination. But the providing of racial preferences—the equivalent of a 310 point SAT boost in admissions to African Americans at selective private universities, according to Thomas Espenshade and Alexandra Radford—has never been held by the Supreme Court to be an appropriate remedy for ongoing societal discrimination. Instead, the argument that has prevailed is that racial preferences are necessary to promote the educational benefits of racial and ethnic diversity in the classroom.

The problem in applying the diversity argument, in turn, is that the University of Texas found a way to create higher levels of racial and ethnic diversity—by reducing reliance on test scores and giving a boost to economically disadvantaged students of all races—than they had using race. In 1996, when a lower court held Texas’s racial preference plan unconstitutional, the freshman class at U.T. Austin (using race in admissions) was 4.1 percent African American and 14.5 percent Hispanic. By 2004, race-neutral alternatives produced a class that was 4.5 percent African-American and 16.9 percent Hispanic. (Texas subsequently began using race again, which prompted the Fisher litigation.)

Moreover, class-based programs can be carefully defined in a way that captures the impact of racial discrimination—by considering, for example, whether a student lives in concentrated poverty, a reflection, in some measure, of racial housing discrimination.

2. Affirmative Action Has Majority Support.

Ifill dismisses concerns that racial preferences are unpopular, suggesting that this is not the proper measure of such policies, and then goes on to suggest “a recent New York Times poll showed that most Americans support affirmative action.” The poll indeed found that, by 53 percent to 38 percent, Americans favor “affirmative action programs for minorities in hiring, promoting and college admissions.” But the issue at stake in Fisher is not the amorphous concept of affirmative steps—which can include encouraging minority students to apply—but whether race should count in who is admitted. On that question, a recent Washington Post poll found by 76 percent to 22 percent, Americans oppose “allowing universities to consider applicants’ race as a factor in deciding which students to admit.” And a Wall Street Journal/NBC poll found that even in using the vague affirmative action language, support was at an “historic low.”

3. Universities Need Wealthy Students of Color.

Ifill also raises the concern that if universities shift the basis of affirmative action preferences from race to class, “we may simply reinforce stereotypes within the student body that will equate minority students with poverty.” This argument was also advanced by the University of Texas: that race-neutral plans produced too many low-income and working-class minority students and that Texas needed to provide a preference to wealthy minority students such as “the African American or Hispanic child of successful professionals in Dallas” who would defy stereotypes.

This line of argument highlights how far the case for race-based affirmative action has drifted from basic concepts of fairness. President Obama recognized the basic issue of equity when he said his own daughters, who are economically and educationally privileged, do not deserve a preference in college admissions. Moreover, as a practical matter, wealthy students of color are precisely the subgroup of minority students most likely to be admitted without a preference because they, like advantaged white students, are the most likely to have enjoyed educational advantages and scored well on standardized tests.

4. Race vs. Class Is a False Dichotomy.

Ifill’s final, and most theoretically plausible, argument is that pitting race and class poses a “false dichotomy” because universities can provide a leg up to students on both criteria. Indeed, universities purport to do just that, saying they consider both race and class in admissions.

But extensive research from Anthony Carnevale of Georgetown University and William Bowen, the former president of Princeton University, puts the lie to that claim. In fact, most selective universities provide very heavy preferences based on race, and virtually no consideration to economic disadvantage. From a self-interested perspective, that’s understandable. A lack of class diversity is easier to hide than a lack of racial diversity; and addressing socioeconomic diversity is more expensive because low-incomes students need greater financial aid and support on campus.

The one big exception to the rule is when universities are banned from using race—in places like California, Florida, and Washington. In order to achieve racial diversity indirectly, they adopt class-based programs in earnest.

I have been hearing the argument, “let’s pursue race and class diversity together,” for more than two decades. But somehow, as long as universities can employ robust racial preferences, the vast majority never gets around to addressing class. As a result, a generation of talented low-income and working-class students have been virtually shut out of America's competitive colleges and universities.