ClassFeaturedOn the Contrary

Class-Based Affirmative Action

Why Labor Should Support Class-Based Affirmative Action
Richard D. Kahlenberg

In the press, debates over affirmative action in higher education pit liberals (who support taking race into account in admissions) and conservatives (who oppose it). But there is a third way on the issue—affirmative action based on class, rather than race—which is far more progressive than our current system of racial preferences. As the U.S. Supreme Court curtails the ability of universities to count skin color in admissions, the class-based approach is quickly gaining ground. This is a development that organized labor ought to cheer.

Early Liberal Opposition to Racial Preferences

In the 1950s and 1960s, many labor unions were at the forefront of fighting for civil rights and against segregation in the South because it was the right thing to do and because segregation undercut a key goal of labor organizers: working-class solidarity. Southern business leaders, by contrast, knew that dividing workers by race was the very best way to keep them splintered and therefore powerless.

After passage of the landmark 1964 Civil Rights Act, outlawing prospective racial discrimination, policymakers were faced with the critical question: how should society address the legacy of past discrimination, and our nation’s egregious history of slavery and segregation? Clearly, affirmative steps were necessary to remedy this history, lest it be perpetuated into the future. But how could this be done without playing into the hands of the right wing, which benefits when working-class people are divided by race?

In a famous 1965 address at Howard University, Lyndon Johnson argued for taking action:

You do not take a person who, for years, has been hobbled by chains and liberate him, bring him up to the starting line of a race and then say, “you are free to compete with all the others” and still justly believe that you have been completely fair.

The dilemma, however, was that if he sup- ported explicit racial preferences—in education or employment—it might cause a backlash from working-class whites who did not consider themselves privileged and would wonder, “why don’t we get a break?” Explicit preferences for jobs or spots in colleges that were based on race, and blind to class, might encourage white working-class people to vote their race (for Republicans) rather than their class (for Democrats).

Significantly, in the Howard University address, Lyndon Johnson did not call for racial preferences—an omission noted by news reports at the time—and instead called for a number of programs to lift up the poor of all races: jobs, housing, and health care.1 He issued an executive order requiring federal contractors to take “affirmative action to ensure that applicants are employed, and that employees are treated during employment, without regard to their race, creed, color or national origin.”

The father of racial preferences was Richard Nixon.

Likewise, Dr. Martin Luther King Jr., argued in his 1964 book, Why We Can’t Wait, that it was imperative to address our nation’s history of discrimination, but instead of proposing a Bill of Rights for Blacks, he advocated a Bill of Rights for the Disadvantaged of all races. On the merits, King argued, “It is a simple matter of justice that America, in dealing creatively with the task of raising the Negro from backwardness, should also be rescuing a large stratum of the forgotten white poor.”2 Politically, King also realized that racial preferences could divide the coalition of working-class whites and blacks that he saw as critical to progressive politics. He wrote to a freelance editor who was helping him write Why We Can’t Wait:

It is my opinion that many white workers whose economic condition is not too far removed from the economic condition of his black brother, will find it difficult to accept a “Negro Bill of Rights,” which seeks to give special consideration to the Negro in the context of unemployment, joblessness, etc. and does not take into sufficient account their plight (that of the white worker).3

Given these political realities, it is perhaps not surprising that the father of racial preferences was not Lyndon Johnson or Martin Luther King and instead was Richard Nixon. In 1969, Nixon proposed the Philadelphia Plan that imposed racial hiring quotas on the city’s construction industry. Bayard Rustin, the great civil rights leader and friend of labor who planned the 1963 March on Washington, was suspicious: why would Nixon, who was no great supporter of civil rights, support a policy of racial preferences? Rustin charged that Nixon was using the Philadelphia Plan to “deliberately throw black and white workers at each other’s throats.”4

Class-Based Affirmative Action as a More Progressive Option Today

Forty-five years later, Americans are still fighting over affirmative action, and the policy of racial preferences in higher education remains a fundamentally conservative practice. An ideal admissions system would consider merit in the broadest sense: not just grades and test scores and leader- ship, but all of those things in the context of what obstacles a student has had to overcome in life. Research suggests that today, those obstacles are primarily economic in nature. Anthony Carnevale at Georgetown University finds that a child growing up with socioeconomic disadvantages—in families where parents have little formal education, income, and wealth, in neighborhoods with concentrations of poverty and the like—is expected to score 399 points lower on the math and verbal sections of the SAT than the most socioeconomically advantaged children. Racial disadvantages impose an additional 56 points.5 So if we want to admit qualified students who have overcome odds, admissions officers would give considerable weight to socioeconomic disadvantage and much smaller consideration to race.

Today, universities do the opposite. According to research by William Bowen and other scholars, being an under-represented minority (black or Latino) increases one’s chance of admissions to elite universities by 28 percentage points. Being from a low-income family does not improve one’s chances at all.6 The athletic department pushes athletes, fund- raisers push the children of alumni, and civil rights advocates push for minority applicants. No one fights for poor and working-class kids.

As a result, Carnevale finds that rich kids outnumber poor kids on selective campuses by fourteen to one.7 Universities essentially admit fairly privileged kids of all colors and call that equity. At highly selective campuses studied by Bowen and Derek Bok, 86 percent of black students were middle or upper class, and the white students are even more privileged.8 Although racial preferences do not translate into socioeconomic diversity, class-based affirmative action programs, structured properly, can produce high levels of racial diversity. At seven of ten leading public colleges where affirmative action was banned, alternatives produced black and Hispanic representations equal to or greater than those accomplished through racial preferences.9 Much publicity has accompanied drops in minority representation at the three outliers—UC (University of California) Berkeley, UCLA, and the University of Michigan—but these are all top-ranked institutions that compete with other highly regarded national universities, the vast majority of which continue to employ racial preferences. An African-American student admitted to Berkeley without a preference may well get into an even more highly ranked university, such as Stanford, which counts race in admission, and choose to attend there. Under those circumstances, it is very hard for Berkeley to compete in attracting under-represented minority students.

At seven of ten colleges where affirmative action was banned, alternatives produced black and Hispanic representations equal to or greater than those accomplished through racial preferences.

So what would happen if all universities were to rely on race-blind admissions? National research finds that economic affirmative action and percentage plans could actually produce greater levels of racial diversity than racial plans.10 The Supreme Court gave an enormous push for class-based affirmative action in its 2013 decision in Fisher v. University of Texas, in which the Court said universities could only use race if alternatives like class-based affirmative action do not produce sufficient racial diversity.

Some critics ask, “doesn’t a shift to class- based affirmative action suggest that racism is a thing of the past?” Not at all. The reason class- based affirmative action produces racial diversity is precisely because past and present discrimination means that African-American and Latino students come from families that are more likely to be poor, to live in neighborhoods with high concentrations of poverty, and to have low amounts of wealth.

In theory, universities could employ affirmative action based on race and add class to the mix, but almost none do. Today, the main universities that pay attention to socioeconomic diversity are those that are banned (often by voter referendum) from employing race in admissions. To their credit, these universities want racial diversity, so they use socioeconomic status as a proxy and end up creating both racial and economic diversity.

Universities could employ affirmative action based on race and add class to the mix, but almost none do.

Texas, for example, adopted the Top 10 Percent Plan, which admitted students to University of Texas at Austin (UT-Austin) from the top 10 percent of every high school— including those located in low-income and working-class areas that had rarely sent a student to UT-Austin in the past. When wealthy parents from tony suburbs in Texas complained that their children were being shut out, a heartening coalition of black and Hispanic urban Democrats joined forces with white working- class rural Republicans to support the 10 per- cent plan. In the U.S. Supreme Court, UT-Austin was left to argue that the reason it needed to continue to use race is that it wanted “the African-American or Hispanic child of successful professionals in Dallas.”11

Is that what the labor movement is about? Should not unions be fighting for poor and working-class kids who no one else will champion?


Race Still Matters: The Continued Need for Race-Conscious Admissions Policies
Julie J. Park

In the Grutter and Fisher cases, the Supreme Court repeatedly affirmed the use of race-conscious admissions policies when race-neutral alternatives are insufficient. Over the years, there have been two primary rationales for affirmative action. The first is the need to compensate for the systemic disadvantages accumulated and experienced by people of color due to systemic and historic patterns of discrimination. The second is the need for universities to provide learning environments that facilitate the educational benefits of engaging with a racially diverse student body. Engagement with racial diversity is associated with numerous positive college outcomes, including critical thinking, leadership, team- work, reduction in bias, and overall college satisfaction.1 Since the Bakke case, the latter (the benefits of diversity) argument has been the prime justification for allowing race-conscious admissions policies. Regardless, the “diversity defense” is intertwined with the “historic discrimination” rationale, in that the educational benefits of diversity in the college environment are effective because most students grow up in segregated environments with little meaningful pre-college engagement with diversity.2

Some who oppose race-conscious affirmative action heartily agree that this country still suffers from the cumulative effects of discrimination and that universities benefit from diversity. In spite of this, they do not see the need for continued race-conscious policies, arguing that we can achieve racial diversity without such measures. However, we still need policies that consider race as one of many factors in the admissions process, or at minimum, the option to use them when needed. Thus, my focus in this piece is on why we specifically need the flexibility to consider race as one of numerous factors in the admissions process (alongside class and other relevant attributes) and less on the broad legal rationale for affirmative action (historic discrimination or the diversity defense). These reasons include the need to use race- conscious policies when race-neutral policies do not result in sufficient diversity, the need for diversity throughout a college campus, and the need for relative equal status among students.

Racial diversity is associated with numerous positive college outcomes, including critical thinking, leadership, teamwork, reduction in bias, and overall college satisfaction.

First, while some institutions like the University of Georgia and University of Washington have been successful in recruiting students of color via race-neutral preferences, states need the option to use race-conscious policies in the event that race-neutral policies on their own are insufficient. Black enrollment has eroded substantially in the University of California (UC) system since the banning of affirmative action in 1996 via Proposition 209. Even with its comprehensive review system, which individually examines students’ applications to capture measures of economic disadvantage, the black student populations in particular at UCLA (3.7 percent), UC-San Diego (2 percent), and UC-Irvine (2.6 percent) are abysmal. In an amicus brief from Fisher v. Texas, the chancellors and president of the UC system comment on the effectiveness of race- neutral approaches:

To date, however, those measures have enjoyed only limited success. They have not enabled the University of California fully to reverse the precipitous decline in minority admission and enrollment that followed the enactment of Proposition 209, nor to keep pace with the growing population of underrepresented minorities in the applicant pool of qualified high school graduates. These effects have been most severe and most difficult to reverse at the University’s most highly-ranked and competitive campuses.3

Some individuals have argued that institutions such as UCLA and Berkeley have been unable to attract diverse student bodies because they draw from a more national pool of students, and thus, they lose students to competing institutions that can still use affirmative action, such as the Ivies.4

This overlooks institutions such as UC-San Diego and UC-Irvine, which draw more heavily from the state, but still are unable to draw black students with race-neutral policies. Institutions within a single state vary considerably in their ability to recruit and retain a diverse student population. Each institution needs the freedom to experiment with its admissions policies and, if necessary, utilize race-conscious admissions tools to meet goals around equity and diversity.

Related, a single institution having a certain percentage of minority students does not mean the campus environment is equitable. The University of Texas–Austin (UT-Austin) is a key example of this, where the university recruited the vast majority of students through its race-neutral Top 10 (or in reality, top 8) Percent Plan, wherein students ranking in the top 8 percent of their high school class are guaranteed admission into the state’s public institutions. Following the ruling in Grutter that universities can consider race in a holistic assessment of applicants, Texas had some lee- way to re-implement race-conscious admissions, which they did for a very small portion of the incoming class. The reasoning was that while the Top 10 Percent Plan resulted in levels of racial diversity comparable with Texas in previous years, UT-Austin was aware that there were still substantive swatches of campuses— particular majors such those related to STEM (science, technology, engineering, and math) fields and classroom environments—that lacked diversity. Given that the Grutter ruling is contingent on the educational benefits of diversity, diversity cannot just be a macro-level phenomenon it needs to penetrate into every sphere of the university to facilitate student learning— and the classroom is an obvious key site. Thus, race-conscious policies are still a necessary tool given an institution’s compelling interest in providing learning environments that maximize the educational benefits of diversity such as team work, critical thinking, and reduced prejudice.

States need the option to use race- conscious policies in the event that race-neutral policies on their own are insufficient.

Third, race-conscious affirmative action, in combination with class-based affirmative action, is still necessary because middle-class under-represented minority (URM, meaning African-American, Latino/a, Native American, Southeast Asian, and Pacific Islander) students play a critical role in the broader campus racial climate. To have healthy inter-group relations, it is important to have students of all races and social classes represented so that no single racial/ethnic group is represented primarily by one socioeconomic bracket. Class-based affirmative action would assist in increasing representation of low-income students of all races, which is severely needed. Still, measures are needed to encourage the recruitment and retention of middle-class URM students

To create a more fluid campus environment where students are interacting across race, institutions need socioeconomic and racial diversity to prevent a double balkanization across race and class lines. Institutions need lower and middle- and upper-income white students, as well as lower- and middle- and upper-income students of color. However, when the URM population consists overwhelmingly of low- income students of color, the social distance between white and minority groups widens, which has implications for how students engage with each other on campus. In contrast, if you have students across the social class spectrum from all groups, you have more opportunities for students to mix across racial/ethnic lines: middle-income black students may socialize with lower-income black students, but they will likely have much in common with middle- income white students. They become a bridge group that helps alleviate balkanization on campus.

Institutions need socioeconomic and racial diversity to prevent a double balkanization across race and class lines.

When people of different backgrounds come together, having a similar status is an important part of supporting positive inter-group relations, a concept known as “relative equal status.”5 While students have some equal status due to their enrollment as students, race and class both present challenges to relative equal status. For instance, a middle-class black student has greater relative equal status with a middle-class white peer, versus a low-income black student and an affluent white student. There is still social distance between students of different races, but some commonality across class lines.

In my research with colleagues Nida Denson and Nicholas Bowman, we found that cross- racial interaction, a key catalyst in triggering the educational benefits of diversity, was maximized at universities that were both racially and socioeconomically diverse. The effects associated with the socioeconomic diversity of an institution were indirect, while the effects attributed to racial diversity on cross-racial interaction were direct and comparatively stronger.6 We suggest that socioeconomic diversity helps “prime” racial diversity—in essence, it helps make racial diversity work better by increasing the likelihood that students of different racial groups will have some commonality across social class regardless of race, which creates a more fluid and less balkanized campus environment.

Some might argue that middle-income URM students are undeserving of any preference in the admissions system. However, there may be middle-income URM students who fall just outside of a high school ranking percentage or SAT threshold who display the ability and potential to succeed in a highly competitive university environment. Universities need the flexibility to consider these students. Their presence contributes to a more robust learning environment and may support the retention of lower-income students of color by creating a more fluid campus environment.

Policies that only consider race and not class neglect low-income students of color, which is a disservice to everyone. However, policies that consider only class and not race also have the potential to limit or hinder the campus environment for diversity. Nurturing a successful cam- pus racial climate is not simply a matter of gathering a certain number of students of color so universities can rest easier knowing they have “enough” diversity. It involves attending to the nuances that shape inter-group relations and influence student learning, key outcomes that are vital to preparing students for citizen- ship in a diverse democracy. This means paying attention to how policies may affect dynamics, such as relative equal status, as well as under- standing that policies that may work effectively in one context (e.g., the Texas model) may not work as well in other contexts.

One area where I can wholeheartedly agree with class-based affirmative action advocates is the need for broadening opportunity for low- income white students, who also have a vital role to play in the campus racial climate. Just as universities need students of color from every socioeconomic background, enrolling greater numbers of lower-income white students will prevent the consolidation of privilege along racial and socioeconomic lines for whites (i.e., if all majority race students are also affluent). Studies have found that levels of cross-racial interaction are higher for lower-income white students, in part because they are more likely to attend high schools with greater levels of racial diversity.7

There is no one-size-fits-all model for institutions when it comes to attracting racial and socioeconomic diversity. However, private and public universities need the flexibility and freedom to consider race in combination with class in their admissions policies to attract student bodies that are both racially and socioeconomically diverse, for the sake of both equity and diversity.


Richard D. Kahlenberg Responds

Julie J. Park makes the argument that because race still matters in American society, racial preferences should continue to be employed by selective colleges and universities. Moreover, she shows particular concern about the plight of middle-class and more advantaged minority students who she worries will not benefit from class-based affirmative action. I disagree with both strands of her analysis.

Park is clearly right that race matters in American society. But it does not necessarily follow that racial preferences are the best policy response. The original supporters of race-based affirmative action argued that because preferences represent a departure from the noble idea that a person’s skin color should not count in deciding who gets ahead in society, those preferences would be temporary in nature. In the 1960s, Whitney Young Jr. of the National Urban League called for “a decade of discrimination in favor of Negro youth,” while in the 1970s, Eleanor Holmes Norton, chair of the Equal Employment Opportunity Commission under President Jimmy Carter, acknowledged that “there is a general consensus in our society” that affirmative action “ought to be temporary.”1 Park’s argument—that racial preferences should exist as long as racial discrimination is part of America’s reality— means we will likely have to count race in college admissions in the foreseeable future.

Permanent racial preference policies are problematic for a variety of reasons, but particularly for those of us on the liberal end of the political spectrum because they divide the multiracial coalition of working-class people. As Georgetown law professor Sheryll Cashin argues in her new book, Place, Not Race, a pol- icy of racial preference “pushes away potential allies” in a way that pleases right-wingers. She asks: Why expend energy “on a policy that primarily benefits the most advantaged children of color, while contributing to a divisive politics that makes it difficult to create quality K-12 education for all children?”2

Park suggests that middle-class and advantaged students of color will be shut out if we shift from preferences based on race to class, but middle class and upper-class minority students are precisely the ones who are most likely to be admitted without a preference because they have the advantages of growing up in more affluent households. (Almost nine in ten African- American students on selective campuses are middle or upper class.) Where is the concern for disadvantaged minority students who would benefit from class-based affirmative action but are virtually absent from selective campuses today?

Where is the concern for disadvantaged minority students who would benefit from class-based affirmative action but are virtually absent from selective campuses today?

Park says she favors using class alongside race, but I have been hearing that argument for two decades, and in the vast majority of schools, the predominant focus remains on racial diversity to the exclusion of socioeconomic status. That is understandable. A lack of socioeco- nomic diversity is much less visible to the naked eye than a lack of racial diversity. And it is much more expensive for universities to admit low-income students of all colors than it is to admit privileged students of all races.

When race is taken off the table, universities suddenly focus on class as a proxy for race, which is a healthy development. Isn’t it time for a better affirmative action that focuses on those incredible students, from all racial backgrounds, who have overcome tremendous odds imposed by our nation’s growing economic inequality?


Julie J. Park Responds

In his piece, Richard D. Kahlenberg does a masterful job of demonstrating the emotional pull behind class-based affirmative action. Why wouldn’t anyone want to give low-income and working-class kids a leg up in the admissions race? I am in complete agreement with the need for class-based preferences in admissions. Still, I maintain my original position that universities need the additional option to consider race in combination with class and numerous other factors when they consider a candidate for admissions.

To reiterate, different institutions within a single state vary considerably in their ability to attract racial/ethnic diversity. Harkening to my home state of Ohio, hypothetically, the Ohio State University may be able to attract an impressive amount of diversity if they utilized an admissions system based on accepting a given percentage of top-ranked students from each high school, including those in working-class and poor neighbor- hoods. However, Miami University, an institution known for its homogeneity, would have a much more difficult time. Certainly, an institution like Miami would benefit from greater socioeconomic diversity, but socioeconomic diversity would not necessarily translate into racial diversity given the demography of the state (i.e., the higher absolute number of white students in the state); hence the need to still consider race in addition to class.

[Given the demography of some states] socioeconomic diversity would not necessarily translate into racial diversity.

Further, affirmative action in higher education does not just affect undergraduate admissions; it affects graduate and professional school admissions, a key influence on the leadership of society. Being able to take race into account in the admissions process is critical to raise up the next generation of scholars, doctors, lawyers, andother professionals who are equipped to navigate a global economy and serve a diverse citizenry. I chair the admissions committee for my program. We look little at GRE scores, finding them generally unhelpful at testifying to a person’s ability to succeed in our field. We pay careful attention to an applicant’s personal statement, undergraduate transcript, interview, and letters of recommendation. We have far more excellent applicants than we do slots for our cohorts, so we consider the type of learning environment we want to create, taking into account the need for diversity  across race, class, gender, and life experience. To create an optimum learning environment, we need the ability to see the full applicant and consider what he or she will bring to the table.

Richard D. Kahlenberg and I both agree that some sort of affirmative action is needed to broaden opportunity in higher education, but we disagree about how to get there. I agree fully that affirmative action is in continual need of improvement, and a greater emphasis on class is needed. Some public institutions may experiment with combined plans, wherein a sizable portion of a class could be admitted through race-neutral means (e.g., percentage plans) but the university would still elect to admit a portion of its class via a more holistic process that would consider students outside of that threshold, taking race and other attributes into account. Affirmative action of any sort is limited in its ability to heal the racial wounds of America’s past, but the diversity it produces is essential for preparing students for leadership and active citizenship in a diverse democracy. Through a combination of race and class-conscious admissions, I hope we will get closer to a vision of higher education that will broaden opportunity and reflect the diversity of our great country.


This article can be downloaded at: